\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Warren Iopa v. Saltchuk-Young Brothers, Ltd.\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-04\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n WARREN K. IOPA, Claimant, No. 17-70415\\n Petitioner,\\n BRB No.\\n v. 16-0232\\n\\n SALTCHUK-YOUNG BROTHERS,\\n LIMITED, Employer; SIGNAL OPINION\\n MUTUAL INDEMNITY ASSOCIATION,\\n LTD., Carrier; DIRECTOR, OFFICE OF\\n WORKERS’ COMPENSATION\\n PROGRAM,\\n Respondents.\\n\\n On Petition for Review of an Order of the\\n Benefits Review Board\\n\\n Submitted February 15, 2019 *\\n Honolulu, Hawaii\\n\\n Filed March 4, 2019\\n\\n Before: Richard C. Tallman, Jay S. Bybee,\\n and N. Randy Smith, Circuit Judges.\\n\\n Per Curiam Opinion\\n\\n *\\n The panel unanimously concludes this case is suitable for decision\\nwithout oral argument. See Fed. R. App. P. 34(a)(2).\\n\\f2 IOPA V. SALTCHUK-YOUNG BROTHERS\\n\\n SUMMARY **\\n\\n\\n Longshore Act / Attorneys’ Fees\\n\\n The panel affirmed a decision by the Benefits Review\\nBoard upholding an administrative law judge’s decision\\nstriking, as untimely, a petition for payment of a claimant’s\\nattorneys’ fees under the Longshore and Harbor Workers’\\nCompensation Act.\\n\\n Following claimant’s successful litigation of claims for\\ntemporary disability benefits under the Longshore Act, the\\nALJ held that he was entitled to reasonable fees and costs.\\nClaimant’s counsel filed a fee petition for work done before\\nthe Office of Workers’ Compensation Programs, and\\nsubsequently filed a corrected petition with the Office of\\nALJs. The ALJ struck the first petition as improperly filed\\nand dismissed the second petition as untimely.\\n\\n The panel held that the ALJ properly used the excusable\\nneglect standard in evaluating the circumstances for the\\nuntimely fee petition. The panel also held that the ALJ\\nproperly applied the four-factor test in Pioneer Investment\\nServices Co. v. Brunswick Associates Ltd. Partnership, 507\\nU.S. 378 (1993), in finding that there was no excusable\\nneglect.\\n\\n\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f IOPA V. SALTCHUK-YOUNG BROTHERS 3\\n\\n COUNSEL\\n\\nJay L. Friedheim, Honolulu, Hawaii; Lara D. Merrigan,\\nMerrigan Legal, San Rafael, California; for Petitioner.\\n\\nNormand R. Lezy, Cox Wooton Lerner Griffin & Hansen\\nLLP, for Respondents Young Brothers Limited and Signal\\nMutual Indemnity Association Ltd.\\n\\nKathleen H. Kim, Office of the Solicitor, United States\\nDepartment of Labor, Washington, D.C., for Respondent\\nDirector, Office of Workers’ Compensation Program.\\n\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n Petitioner Warren Iopa appeals the United States\\nDepartment of Labor’s Benefits Review Board’s (“BRB”)\\norder affirming an Administrative Law Judge’s (“ALJ”)\\ndecision striking as untimely a petition for payment of his\\nattorney’s fees under the Longshore and Harbor Workers’\\nCompensation Act (“Longshore Act”), 33 U.S.C. §§ 901–\\n50, filed more than nine months past the ALJ-ordered\\ndeadline. We now consider for the first time in our circuit\\nwhether striking an untimely petition for attorney’s fees\\nunder the Longshore Act is proper only given extreme\\ncircumstances, or whether excusable neglect is the proper\\nstandard by which to evaluate such petitions. We hold that\\nthe excusable neglect analysis is proper and affirm the\\nBRB’s decision to uphold the ALJ’s dismissal order.\\n\\f4 IOPA V. SALTCHUK-YOUNG BROTHERS\\n\\n I\\n\\n Following Iopa’s successful litigation of claims for\\ntemporary disability benefits under the Longshore Act, the\\nALJ held that he was entitled to reasonable attorney’s fees\\nand costs, and that a fee petition had to be filed within\\n21 days of the award order entered July 31, 2014. See\\n20 C.F.R. § 702.132(a). On June 8, 2015, Iopa’s counsel\\ninstead improperly filed a fee petition for work done before\\nthe Office of Workers’ Compensation Programs (“OWCP”).\\nAt the request of the ALJ’s office, counsel filed a corrected\\npetition with the Office of Administrative Law Judges\\n(“OALJ”) on October 27, 2015. The ALJ then issued an\\norder striking the first petition due to his lack of authority to\\naward attorney’s fees for work done before the OWCP, and\\nstriking the second petition based on a finding of\\nuntimeliness without excusable neglect.\\n\\n II\\n\\n We have jurisdiction under 33 U.S.C. § 921(c). We\\nreview BRB decisions under the Longshore Act “for errors\\nof law and for adherence to the substantial evidence\\nstandard.” Gen. Const. Co. v. Castro, 401 F.3d 963, 965 (9th\\nCir. 2005) (quoting Alcala v. Dir., OWCP, 141 F.3d 942, 944\\n(9th Cir. 1998)). We conduct de novo review on questions\\nof law, including questions of statutory interpretation, under\\nthe Longshore Act. See Pedroza v. BRB, 624 F.3d 926, 930\\n(9th Cir. 2010). “Because the [BRB] is not a policymaking\\nentity, we accord no special deference to its interpretation of\\nthe Longshore Act.” Price v. Stevedoring Servs. of Am., Inc.,\\n697 F.3d 820, 825 (9th Cir. 2012).\\n\\f IOPA V. SALTCHUK-YOUNG BROTHERS 5\\n\\n III\\n\\n Iopa’s counsel argues that the ALJ did not apply the\\nproper standard in evaluating the circumstances for the\\nuntimely fee petition and, alternatively, even if the proper\\nstandard was applied, substantial evidence does not support\\nthe ALJ’s decision to strike fees.\\n\\n A\\n\\n Iopa asserts that Longshore Act fee petitions are subject\\nto the relatively lenient standard adopted by the BRB in\\n1986: “The loss of an attorney’s fee is a harsh result and\\nshould not be imposed on counsel as a penalty except in the\\nmost extreme circumstances.” Paynter v. Dir., OWCP,\\n9 Black Lung Rep. (Juris) 1-190, at *1 (Ben. Rev. Bd. 1986).\\nIn 2015, however, the Rules of Practice and Procedure for\\nAdministrative Hearings Before the OALJ were revised to\\ninclude, inter alia, the following provision: “When an act\\nmay or must be done within a specified time, the judge may,\\nfor good cause, extend the time . . . [o]n motion made after\\nthe time has expired if the party failed to act because of\\nexcusable neglect.” 29 C.F.R. § 18.32(b)(2) (emphasis\\nadded). This rule applies to claims brought before an ALJ\\nin the Department of Labor, including Longshore Act\\nclaims. See id. § 18.10(a). While Paynter may have\\npreviously served as the primary guide in determining\\nwhether to strike a fee petition, the 2015 revision of the\\nRules of Practice and Procedure for Hearings Before the\\nOALJ requiring a showing of “excusable neglect” for\\nuntimely claims cannot be ignored. See id. § 18.32(b)(2).\\n\\n In determining whether circumstances constitute\\nexcusable neglect, the Supreme Court set forth the following\\nfour-factor test in Pioneer Investment Services Co. v.\\nBrunswick Associates Ltd. Partnership: “the danger of\\n\\f6 IOPA V. SALTCHUK-YOUNG BROTHERS\\n\\nprejudice to the debtor, the length of the delay and its\\npotential impact on judicial proceedings, the reason for the\\ndelay, including whether it was within the reasonable control\\nof the movant, and whether the movant acted in good faith.”\\n507 U.S. 380, 395 (1993). We and our sister circuits have\\nadopted the Supreme Court’s four-factor test. See Pincay v.\\nAndrews, 389 F.3d 853, 855–60 (9th Cir. 2004) (weighing\\nPioneer factors in untimely filing of notice of appeal under\\nFederal Rule of Appellate Procedure 4(a)(1)(A)); Briones v.\\nRiviera Hotel & Casino, 116 F.3d 379, 381–82 (9th Cir.\\n1997) (adopting the Pioneer test in evaluating motions for\\nrelief under Federal Rule of Civil Procedure 60(b)(1)); In re\\nO’Brien Envtl. Energy, Inc., 188 F.3d 116, 125 n.7 (3d Cir.\\n1999) (recognizing Pioneer as providing “guidance not just\\nwith regard to [bankruptcy] Rule 9006, but in other . . . non-\\nbankruptcy contexts discussing the issue of excusable\\nneglect”); Pratt v. Philbrook, 109 F.3d 18, 19 (1st Cir. 1997)\\n(explaining that “the Pioneer test for ‘excusable neglect’ was\\nintended to extend beyond the bankruptcy context”). We\\nhold that applying the Pioneer factors to the instant case is\\nappropriate and consistent with post-Pioneer case law\\nanalyzing “excusable neglect” in various regulatory\\ncontexts. The ALJ did not, therefore, commit an error of law\\nby applying an improper standard.\\n\\n B\\n\\n The ALJ’s four-factor Pioneer analysis and subsequent\\nconclusion that Iopa’s counsel did not establish excusable\\nneglect was supported by substantial evidence. The ALJ\\nfound that the first factor—prejudice—weighed against a\\nfinding of excusable neglect. The ALJ determined that\\nRespondents demonstrated they would be prejudiced by the\\ndelayed filing, because their “memory of the details of the\\ncase” and ability “to recall each back and forth between the\\n\\f IOPA V. SALTCHUK-YOUNG BROTHERS 7\\n\\nparties for the purpose of contesting the validity or amount\\nof time claimed for a given line item” was affected by the\\nsubstantial delay. A reasonable factfinder would not be\\ncompelled to disagree with this analysis and finding.\\n\\n The ALJ found that the second factor—the length of\\ndelay—weighed strongly against a finding of excusable\\nneglect, because the delay was substantial. That finding is\\nsupported by the fact that the petition was filed\\napproximately 280 days past the established deadline of 21\\ndays. The fact that Iopa’s lawyer waited another month to\\ncorrect his petition after being instructed by the ALJ to file\\nthe proper petition with the OALJ instead of the OWCP\\nfurther supports this finding.\\n\\n The ALJ also found that the third factor—the reasons for\\ndelay—weighed against a finding of excusable neglect. The\\nALJ’s determination that “none of [the reasons for delay] are\\nconvincing or persuasive” or were beyond the control of\\ncounsel is supported by case law. Although Iopa’s counsel\\nnoted several challenges in managing his caseload,\\nparticularly following the departure of the associate who\\nmanaged this case, the Supreme Court has held that “we give\\nlittle weight to the fact that counsel was experiencing\\nupheaval in his law practice.” Pioneer, 507 U.S. at 398; see\\nalso In re Enron Corp., 419 F.3d 115, 126–27 (2d Cir. 2005)\\n(holding a party being too busy with negotiations was not\\nexcusable neglect); In re Harlow Fay, Inc., 993 F.2d 1351,\\n1352 (8th Cir. 1993) (holding counsel’s relocation to a\\ndifferent state and reduction in staff was not excusable\\nneglect); cf. Selph v. Council of L.A., 593 F.2d 881, 884 (9th\\nCir. 1979) (explaining that “excusable neglect is not meant\\nto cover the usual excuse that the lawyer is too busy, which\\ncan be used, perhaps truthfully, in almost every case”)\\n(citation omitted).\\n\\f8 IOPA V. SALTCHUK-YOUNG BROTHERS\\n\\n The ALJ found that the fourth factor—good faith—had\\nno weight in this case. Even if the ALJ had found that\\ncounsel acted in good faith, that factor does not require a\\nfinding of excusable neglect when weighed against the other\\nthree factors. See In re Veritas Software Corp. Sec. Litig.,\\n496 F.3d 962, 973 (9th Cir. 2007) (affirming the district\\ncourt’s denial of a fee application as untimely where two\\nfactors favored a finding of excusable neglect, stating that an\\nexcusable neglect determination is “committed to the\\ndiscretion of the district court” and “[w]hile the district court\\nwould not have abused its discretion in granting [the] fee\\napplication, it did not abuse its discretion in denying it”).\\nWe affirm the BRB’s decision upholding the ALJ’s finding\\nof untimeliness absent excusable neglect.\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1270,"cells":{"id":{"kind":"number","value":4595922,"string":"4,595,922"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. Antonio Gilton"},"decision_date":{"kind":"string","value":"2019-03-04"},"docket_number":{"kind":"string","value":"16-10109"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: J. Clifford Wallace, M. Margaret McKeown, and Jay S. Bybee, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"type\": \"010combined\", \"text\": \"FILED\\n FOR PUBLICATION\\n MAR 04 2019\\n UNITED STATES COURT OF APPEALS MOLLY C. DWYER, CLERK\\n U.S. COURT OF APPEALS\\n\\n\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 16-10109\\n\\n Plaintiff-Appellant, D.C. No.\\n 3:13-cr-00764-WHO-1\\nv.\\n\\nREGINALD ELMORE, AKA Fat Reg; OPINION\\nESAU FERDINAND, AKA Sauce,\\n\\n Defendants,\\n\\nBARRY GILTON, AKA Prell; ADRIAN\\nGORDON, AKA Tit; MONZELL\\nHARDING, Jr.; CHARLES HEARD,\\nAKA Cheese; LUPE MERCADO; PAUL\\nROBESON, AKA P World; ALFONZO\\nWILLIAMS, AKA Fonz, AKA Relly;\\nJAQUAIN YOUNG, AKA Loc,\\n\\n Defendants,\\n\\nand\\n\\nANTONIO GILTON, AKA TG, AKA\\nTone,\\n\\n Defendant-Appellee.\\n\\n\\n Appeal from the United States District Court\\n for the Northern District of California\\n William Horsley Orrick, District Judge, Presiding\\n\\f Argued and Submitted January 7, 2019\\n San Diego, California\\n\\nBefore: J. Clifford Wallace, M. Margaret McKeown, and Jay S. Bybee, Circuit\\nJudges.\\n\\n Opinion by Judge Bybee\\n\\nBYBEE, Circuit Judge:\\n\\n Following the June 2012 murder of Calvin Sneed, police obtained a warrant\\n\\nauthorizing the seizure of Antonio Gilton’s historical cell-site location information.\\n\\nGilton was subsequently charged with four counts relating to the murder of Sneed.\\n\\nHe moved to suppress the location data, arguing that the warrant issued without\\n\\nprobable cause and that the officers’ reliance on the warrant was not in good faith.\\n\\nThe district court granted Gilton’s motion, concluding that the warrant was so\\n\\ndeficient in indicia of probable cause that no officer could have relied on the\\n\\nwarrant in good faith.\\n\\n Although we agree with the district court that the warrant authorizing the\\n\\nseizure of Gilton’s location data was not supported by probable cause, we conclude\\n\\nthat the deficiencies were not so stark as to render the officers’ reliance on the\\n\\nwarrant “entirely unreasonable.” See United States v. Leon, 468 U.S. 897, 923\\n\\n(1984). We reverse.\\n\\n I. FACTS AND PROCEEDINGS\\n\\n\\n 2\\n\\f A\\n\\n At approximately 2:00 a.m. on June 4, 2012, officers of the San Francisco\\n\\nPolice Department responded to a report of shots fired in the area of Meade and Le\\n\\nConte Avenues, San Francisco. When the officers arrived, they found Calvin\\n\\nSneed slumped in the driver’s seat of a crashed Toyota Camry with a gunshot\\n\\nwound to his head. Sneed was later pronounced dead.\\n\\n Standing distraught next to the car was Sneed’s minor girlfriend, L.G. She\\n\\ntold the police that approximately eight months before the shooting, she had moved\\n\\nfrom San Francisco to Los Angeles for a “new start,” and that she had been staying\\n\\nwith her “elder brother,” Antonio Gilton (hereinafter “Gilton”).1 In Los Angeles,\\n\\nL.G. met and began dating Calvin Sneed. L.G. subsequently learned that Sneed\\n\\nwas pimping young women for prostitution in the Los Angeles area, and she began\\n\\nto advertise herself on various prostitution websites.\\n\\n L.G.’s friends and family eventually discovered what she was doing, and\\n\\nL.G.’s mother traveled to Los Angeles on May 31, 2012, to persuade L.G. to return\\n\\nto San Francisco. L.G. did not return with her mother. Three days later, however,\\n\\n\\n\\n\\n 1\\n On appeal, Antonio Gilton informs us that he is actually L.G.’s cousin, not\\nher older brother. All parties agree, however, that L.G. was referring to Antonio\\nGilton when she mentioned her “elder brother.”\\n 3\\n\\fL.G. and Sneed traveled together to San Francisco, where Sneed dropped L.G. off\\n\\nat her parents’ house around 4:00 p.m.\\n\\n L.G. stayed with her parents until approximately 12:15 a.m. on June 4, 2012,\\n\\nwhen she texted Sneed to come and pick her up. L.G. began to argue with her\\n\\nmother about returning to Los Angeles with Sneed, but her father merely told her\\n\\n“you grown” and instructed her “before you leave, turn the lights off” before he\\n\\nexited the room.\\n\\n Around 1:56 a.m., L.G. was waiting for Sneed outside her parents’ home and\\n\\nnoticed a silver SUV parked nearby with its lights on. As Sneed’s car arrived, the\\n\\nsilver SUV drove off, but as Sneed drove past where L.G. was standing and turned\\n\\nthe corner, L.G. saw the SUV reappear and accelerate towards Sneed’s vehicle.\\n\\nL.G. heard gunshots and saw a muzzle flash coming from the SUV. L.G. then\\n\\nheard a crash and ran to Sneed’s car, where she found him “slumped in the driver’s\\n\\nseat with a gunshot wound to his head.” L.G. cooperated with police and allowed\\n\\nthem to search her cell phone. During the search, the police identified and L.G.\\n\\nconfirmed cell phone numbers for her father, her mother, Gilton, and L.G.’s\\n\\nyounger brother.\\n\\n Later that day, the police received confidential information implicating\\n\\nL.G.’s father, Barry Gilton, and a second, unidentified individual in the murder.\\n\\n\\n 4\\n\\fAs a result, the police obtained historical cell-site location information (“CSLI”)\\n\\ndata for Barry Gilton’s phone pursuant to an exigent circumstances request to T-\\n\\nMobile.2 Although Barry Gilton told police that he had been at home in bed after\\n\\n12:15 a.m. on the night of the murder, the CSLI data indicated that between 12:49\\n\\na.m. and 2:19 a.m. that night, Barry Gilton’s cell phone traveled from near his\\n\\nhome to the Western Addition. The cell phone then returned to the vicinity of the\\n\\nGilton home around the time of the murder before traveling towards the northern\\n\\narea of the Mission after the shooting. The police also obtained video surveillance\\n\\nfrom a camera near the site of the murder that showed a light colored mid-size\\n\\nSUV believed to be the vehicle used in the shooting.\\n\\n Based on all the information set forth above, San Francisco Police\\n\\nDepartment (SFPD) Sergeant Gary Watts submitted a fourteen-page affidavit in\\n\\nsupport of a state search warrant. The application sought CSLI data for two cell\\n\\n\\n\\n 2\\n Cell sites usually consist of a set of radio antennas mounted on a tower,\\nalthough “they can also be found on light posts, flagpoles, church steeples, or the\\nsides of buildings.” Carpenter v. United States, 138 S. Ct. 2206, 2211 (2018).\\n“Each time [a] phone connects to a cell site, it generates a time-stamped record\\nknown as cell-site location information (CSLI).” Id. This CSLI data indicates the\\ngeneral geographic area in which the cell phone user was located when his or her\\nphone connected to the network. Because most smartphones tap into the wireless\\nnetwork “several times a minute whenever their signal is on . . . modern cell\\nphones generate increasingly vast amounts of increasingly precise CSLI.” Id. at\\n2211–12.\\n 5\\n\\fphone numbers, a phone associated with an unknown individual and the phone\\n\\nassociated with Gilton. Sergeant Watts laid out in some detail what the police\\n\\nknew about the crime, Sneed’s relationship with L.G., and information learned\\n\\nfrom the confidential informant. He related how L.G. was related to Gilton, that\\n\\nshe had been staying with Gilton in Los Angeles (where she met Sneed and began\\n\\nengaging in prostitution), that Gilton’s phone number was in her cell phone, and\\n\\nthat the murder was likely committed by a family member or members. Watts\\n\\naverred that “there appear[ed] to be probable cause to believe that the cell phone\\n\\nnumbers provided [would] tend to show . . . possible first-hand knowledge of those\\n\\npersons responsible for the shooting of . . . Calvin Sneed” and that “the cell-site\\n\\ntower locations used on the date and times listed could possibly lead to the proper\\n\\nidentity and the whereabouts of additional persons associated with this crime.”\\n\\nWatts also stated that he had “discussed the merits of the case with the District\\n\\nAttorney’s Office.”\\n\\n That same day, June 6, 2012, a judge of the Superior Court of California,\\n\\nCounty of San Francisco issued a warrant to Sprint for the seizure of cell phone\\n\\nrecords for Gilton’s phone number. The warrant identified three categories of\\n\\ninformation for seizure: (1) subscriber and billing information; (2) all incoming\\n\\nand outgoing calls and text messages from the period of May 1, 2012, to June 6,\\n\\n\\n 6\\n\\f2012; and (3) cell-site location information (CSLI). Only the third category—the\\n\\nCSLI data—is at issue in this appeal.\\n\\n B\\n\\n Gilton was indicted by a federal grand jury in 2013, on four counts related to\\n\\nSneed’s murder. In 2015, Gilton filed a motion to suppress the CSLI obtained by\\n\\nthe San Francisco police pursuant to the Sprint warrant. The district court granted\\n\\nGilton’s motion to suppress. The court concluded first that Gilton had a reasonable\\n\\nexpectation of privacy in the historical CSLI he sought to suppress. Accordingly,\\n\\nthe court determined that probable cause was required to obtain that CSLI from\\n\\nSprint. The court concluded, however, that “[t]he affidavit in support of the Sprint\\n\\nwarrant plainly failed to provide a substantial basis for concluding that there was\\n\\nprobable cause to search” because the affidavit “hardly mention[ed] Gilton” and\\n\\ndid not provide a substantial basis for inferring that Gilton was in the San\\n\\nFrancisco area at the time of the shooting. Even assuming that the warrant\\n\\nsupported an inference of a “family-based attack,” the district court concluded that\\n\\nthe “facts pointed to one particular family member being involved: [Barry] Gilton,\\n\\nnot [Antonio] Gilton.” Finally, the district court concluded that the “good faith”\\n\\nexception did not apply because “it was entirely unreasonable to believe that the\\n\\n\\n\\n\\n 7\\n\\faffidavit’s passing, innocuous references to A. Gilton established probable cause to\\n\\nobtain his cell phone data.”\\n\\n The government appealed the district court’s order granting the motion to\\n\\nsuppress. We originally heard argument in March 2017, but then withdrew the\\n\\ncase from submission pending the final decision of the Supreme Court in United\\n\\nStates v. Carpenter, 138 S. Ct. 2206 (2018). Following the Court’s resolution of\\n\\nCarpenter, we requested supplemental briefing and heard reargument in January\\n\\n2019. We agree with the district court that the police lacked probable cause to\\n\\nobtain Gilton’s CSLI data, but we disagree with the district court’s conclusion that\\n\\nthe good faith exception does not apply. Accordingly, we reverse.\\n\\n II. ANALYSIS\\n\\n In Carpenter, the Supreme Court held that “an individual maintains a\\n\\nlegitimate expectation of privacy in the record of his physical movements as\\n\\ncaptured through CSLI.” 138 S. Ct. at 2217. Consequently, Fourth Amendment\\n\\nprotections generally require the government to “obtain a warrant supported by\\n\\nprobable cause before acquiring [CSLI] records.” Id. at 2221.\\n\\n Here, the government did in fact obtain a warrant authorizing the acquisition\\n\\nof Gilton’s CSLI data. Our analysis is thus confined to the questions of whether\\n\\nthat warrant was supported by probable cause, and, if not, whether the search\\n\\n\\n 8\\n\\fshould nevertheless be upheld on the basis of the officers’ good faith reliance on\\n\\nthe warrant.\\n\\n A\\n\\n We turn first to the issue of whether the Sprint warrant was supported by\\n\\nprobable cause. Probable cause exists where the totality of the circumstances\\n\\nindicates a “fair probability that . . . evidence of a crime will be found in a\\n\\nparticular place.” Illinois v. Gates, 462 U.S. 213, 238 (1983). This standard does\\n\\nnot require the affidavit to establish that the evidence is in fact in the place to be\\n\\nsearched, or even that it is more likely than not to be there. United States v.\\n\\nFernandez, 388 F.3d 1199, 1254 (9th Cir. 2004), modified 425 F.3d 1248 (9th Cir.\\n\\n2005). Rather, the issuing judge “need only conclude that it would be reasonable\\n\\nto seek the evidence in the place indicated in the affidavit.” Id. (quotation\\n\\nomitted).\\n\\n In general, the issuing judge’s finding of probable cause is entitled to “great\\n\\ndeference.” United States v. Krupa, 658 F.3d 1174, 1177 (9th Cir. 2011).\\n\\n“[N]otwithstanding the deference that magistrates deserve,” however, we may\\n\\ndetermine that a warrant was invalid where “the magistrate’s probable-cause\\n\\ndetermination reflected an improper analysis of the totality of the circumstances.”\\n\\n\\n\\n\\n 9\\n\\fLeon, 468 U.S. at 915. In addition, the decision of the magistrate “cannot be a\\n\\nmere ratification of the bare conclusions of others.” Gates, 462 U.S. at 239.\\n\\n Here, police sought a warrant to obtain subscriber and billing information,\\n\\nincoming and outgoing calls, incoming and outgoing texts, and CSLI information\\n\\nfor phones belonging to Gilton and a second, unidentified person. The application\\n\\nwas supported by a lengthy affidavit describing the results of interviews, video\\n\\nsurveillance, and an exigent circumstances request to T-Mobile. The affidavit thus\\n\\nprovided the magistrate with a basis for determining the existence of probable\\n\\ncause, see id., and the magistrate’s finding of probable cause is appropriately\\n\\nentitled to deference, see Krupa, 658 F.3d at 1177.\\n\\n Although the affidavit explained in some detail the need for the information\\n\\nfrom the unidentified person’s phone, the affidavit mentions Gilton only three\\n\\ntimes. First, the affidavit records L.G.’s testimony that she “was staying with her\\n\\nelder brother in L.A.” Second, the affidavit indicates that one of the phone\\n\\nnumbers discovered in L.G.’s phone belonged to Gilton. Third, the affidavit states\\n\\nthat L.G. identified which number belonged to Gilton. As the district court pointed\\n\\nout, the affidavit does not indicate that police had reason to believe that Gilton\\n\\n“was . . . in or around San Francisco on or around June 4, 2012.”\\n\\n\\n\\n\\n 10\\n\\f The affidavit’s scant and innocuous references to Gilton do not establish a\\n\\n“fair probability” that evidence of the crime would be found in Gilton’s location\\n\\ndata. Rather, these facts merely indicate that L.G. had the phone number of a\\n\\nfamily member with whom she had lived in Los Angeles. Surely, it is common for\\n\\nan adolescent to store the phone number of the relative with whom she is living,\\n\\nand the mere existence of a familial connection between L.G. and Gilton is not\\n\\nsufficient to render it “reasonable” to search for evidence of the crime in Gilton’s\\n\\nlocation data. See United States v. Grant, 682 F.3d 827, 836–37 (9th Cir. 2012)\\n\\n(stating that an association “through family . . . affiliation” is insufficient to\\n\\nestablish probable cause).\\n\\n The government depends on two related inferences to support the finding of\\n\\nprobable cause. First, the government contends that the totality of the\\n\\ncircumstances supports an inference that “[t]he murder of Calvin Sneed was a\\n\\nfamily solution to a family problem.” Second, because Gilton is related to L.G.\\n\\nand because L.G. met Sneed while living with Gilton, the government asserts that\\n\\nthe magistrate could reasonably infer that Gilton was well aware of Sneed’s\\n\\nrelationship with L.G., that Gilton had reason to be upset with Sneed, and that he\\n\\nhad a motive to commit the crime. The government argues that these two\\n\\ninferences tie Gilton to the murder. We disagree.\\n\\n\\n 11\\n\\f First, we agree with the district court that the affidavit “pointed to one\\n\\nparticular family member being involved in the attack: B. Gilton, not A. Gilton.”\\n\\nAlthough the affidavit indicated that more than one person was likely involved in\\n\\nthe shooting and that the second person might have been a family member, nothing\\n\\nin the affidavit suggested that the identity of the second person was Gilton. Nor do\\n\\nwe agree with the government’s contention that a magistrate could reasonably\\n\\nbelieve that Gilton was the accomplice because of his supposed motive. As the\\n\\ndistrict court pointed out, there is no evidence in the affidavit that Gilton “had\\n\\ncommunicated with family members (or anyone else) about [L.G.’s] relationship\\n\\nwith Sneed,” although that surely was one of the reasons police wanted Gilton’s\\n\\ncall and text records. There is thus no basis in the affidavit to support the inference\\n\\nthat Gilton was upset with Sneed, that he had motive to commit the crime, and,\\n\\nimportantly, that he was in the San Francisco area on the night of the murder. We\\n\\nconclude, as did the district court, that the affidavit’s passing references to Gilton\\n\\nare insufficient to support a reasonable inference that evidence of a crime would be\\n\\nfound in his CSLI data.\\n\\n B\\n\\n Having determined that the warrant was not supported by probable cause,\\n\\nwe turn now to the question of whether the officers nevertheless relied in good\\n\\n\\n 12\\n\\ffaith on the warrant they obtained. The Supreme Court has made quite clear that\\n\\n“[t]he Fourth Amendment contains no provision expressly precluding the use of\\n\\nevidence obtained in violation of its commands.” Leon, 468 U.S. at 906. Because\\n\\n“[t]he wrong condemned by the Amendment is ‘fully accomplished’ by the\\n\\nunlawful search or seizure itself, . . . the exclusionary rule is neither intended nor\\n\\nable to ‘cure the invasion of the defendant’s rights which he has already suffered.’”\\n\\nId. (first quoting United States v. Calandra, 414 U.S. 338, 354 (1974); then\\n\\nquoting Stone v. Powell, 428 U.S. 465, 540 (1976)). As a result, “the use of fruits\\n\\nof a past unlawful search or seizure ‘work[s] no new Fourth Amendment wrong.’”\\n\\nId. (alteration in original) (quoting Calandra, 414 U.S. at 354).\\n\\n The exclusionary rule is thus not “a personal constitutional right of the party\\n\\naggrieved,” but rather “a judicially created remedy designed to safeguard Fourth\\n\\nAmendment rights generally through its deterrent effect.” Calandra, 414 U.S. at\\n\\n348. As such, the question of “[w]hether the exclusionary sanction is appropriately\\n\\nimposed in a particular case” is an entirely separate issue “from the question\\n\\nwhether the Fourth Amendment rights of the party seeking to invoke the rule were\\n\\nviolated by police conduct.” Leon, 468 U.S. at 906 (quoting Gates, 462 U.S. at\\n\\n223). Thus, having determined that the SFPD’s search of Gilton’s CSLI data\\n\\n\\n\\n\\n 13\\n\\fviolated his Fourth Amendment rights, we must address separately the question of\\n\\nwhether exclusion is an appropriate remedy.\\n\\n The question of application of the exclusionary rule is answered by weighing\\n\\n“[t]he substantial social costs exacted by the exclusionary rule” with the benefit of\\n\\nincreased deterrence of police misconduct. Id. at 906–07. “If . . . the exclusionary\\n\\nrule does not result in appreciable deterrence” of police misconduct, “then, clearly,\\n\\nits use in the instant situation is unwarranted.” United States v. Janis, 428 U.S.\\n\\n433, 454 (1976). To have any appreciable deterrent benefit, the exclusion of\\n\\nevidence “must alter the behavior of individual law enforcement officers or the\\n\\npolicies of their departments.” Leon, 468 U.S. at 918.\\n\\n In Leon, the Supreme Court took up the issue of whether the exclusionary\\n\\nrule should be applied in cases where the police seized evidence “in reasonable,\\n\\ngood-faith reliance on a search warrant that is subsequently held to be defective.”\\n\\nId. at 905. The Court reasoned that “[t]he deterrent purpose of the exclusionary\\n\\nrule necessarily assumes that the police have engaged in willful, or at the very least\\n\\nnegligent, conduct which has deprived the defendant of some right,” and that\\n\\n“[w]here the official action was pursued in complete good faith . . . the deterrence\\n\\nrationale loses much of its force.” Id. at 919 (quoting United States v. Peltier, 422\\n\\nU.S. 531, 539 (1975)). In such cases, “excluding the evidence will not further the\\n\\n\\n 14\\n\\fends of the exclusionary rule in any appreciable way.” Id. at 920 (quoting Stone,\\n\\n428 U.S. at 539–40).\\n\\n The Court emphasized that “[t]his is particularly true . . . when an officer\\n\\nacting with objective good faith has obtained a search warrant from a judge or\\n\\nmagistrate and acted within its scope.” Id. Because “[i]t is the magistrate’s\\n\\nresponsibility to determine whether the officer’s allegations establish probable\\n\\ncause[,] . . . [i]n the ordinary case, an officer cannot be expected to question the\\n\\nmagistrate’s probable cause determination.” Id. at 921. As a result, “[i]n most\\n\\nsuch cases, there is no police illegality” and “[p]enalizing the officer for the\\n\\nmagistrate’s error, rather than his own, cannot logically contribute to the deterrence\\n\\nof Fourth Amendment violations.” Id. at 920–21. The Court concluded that “the\\n\\nmarginal or nonexistent benefits produced by suppressing evidence obtained in\\n\\nobjectively reasonable reliance on a subsequently invalidated search warrant\\n\\ncannot justify the substantial costs of exclusion.” Id. at 922.\\n\\n The existence of a search warrant does not automatically preclude\\n\\napplication of the exclusionary rule, however. Rather, the Court instructed that we\\n\\nshould conduct our analysis “on a case-by-case basis,” id. at 918, and recognized\\n\\nfour exceptions to the exception where suppression “remains an appropriate\\n\\nremedy,” id. at 923. First, in cases where “the magistrate or judge . . . was misled\\n\\n\\n 15\\n\\fby information in an affidavit that the affiant knew was false or would have known\\n\\nwas false except for his reckless disregard of the truth.” Id. Second, “in cases\\n\\nwhere the issuing magistrate wholly abandoned his judicial role” by acting as “an\\n\\nadjunct law enforcement officer” or mere “rubber stamp” for the police. Id. at 914,\\n\\n923 (citing Lo-Ji Sales, Inc. v. New York, 442 U.S. 319 (1979)). Third, in cases\\n\\nwhere “an affidavit [is] so lacking in indicia of probable cause as to render official\\n\\nbelief in its existence entirely unreasonable.” Id. at 923 (citation omitted). Finally,\\n\\nin cases where the warrant is “so facially deficient—i.e., in failing to particularize\\n\\nthe place to be searched or the things to be seized—that the executing officers\\n\\ncannot reasonably presume it to be valid.” Id.\\n\\n Here, the parties do not urge and the evidence does not support a conclusion\\n\\nthat the officers wilfully mislead the magistrate, that the magistrate wholly\\n\\nabdicated his role as a neutral and detached judicial officer, or that the warrant was\\n\\nfacially deficient. Thus, the sole question before us is whether Officer Watts’s\\n\\naffidavit was “so lacking in indicia of probable cause” that no reasonable officer\\n\\nwould believe that probable cause existed to search Gilton’s CSLI data. We\\n\\nconclude that the deficiencies in probable cause discussed in subpart II.A., supra,\\n\\nare not so stark as to render official belief in the existence of probable cause\\n\\n“entirely unreasonable.”\\n\\n\\n 16\\n\\f First, we observe that the application for the warrant was issued only two\\n\\ndays after the murder. Second, the affidavit in support of the application was\\n\\nlengthy and laid out for the magistrate what the police knew about the murder and\\n\\nhow they had obtained the information they had, including the results of interviews\\n\\nwith family members and witnesses, video surveillance from a camera on a house\\n\\nnear the murder scene, information received from an anonymous caller who may\\n\\nhave had first-hand facts, and cell phone information obtained on an emergency\\n\\nbasis from the T-Mobile Law Enforcement Center. Third, the police suspected that\\n\\nboth L.G.’s father, Barry Gilton, and a female family member had been involved in\\n\\ninstigating Sneed’s murder. The police also had reason to believe Barry Gilton and\\n\\nat least one other person were in the car when the shooting occurred. Fourth, the\\n\\naffidavit indicated that police had discussed the case with the District Attorney’s\\n\\noffice.\\n\\n For the reasons we have previously explained, all of these facts did not add\\n\\nup to probable cause to obtain Gilton’s CSLI data. The police likely had probable\\n\\ncause to obtain the other information they sought from Gilton’s cellphone—his\\n\\nincoming and outgoing calls and texts—because L.G. had been living with Gilton\\n\\nin Los Angeles when she was dating Sneed, and Gilton may well have been in\\n\\ncontact with L.G.’s family in San Francisco about Sneed. Gilton does not\\n\\n\\n 17\\n\\fchallenge the scope of the warrant with respect to his calls and texts, only his CSLI\\n\\ninformation. What the affidavit is missing is an explanation for why police\\n\\nthought Gilton’s CSLI information was relevant, since there was no\\n\\nsuggestion—other than the cryptic information that multiple family members were\\n\\nlikely involved and that there was a second person in the car, who might or might\\n\\nnot have been a family member—that Gilton was in San Francisco rather than\\n\\nhome in Los Angeles. But we cannot say that no reasonable officer would have\\n\\nrelied on the warrant obtained from the Superior Court just two days into their\\n\\ninvestigation.\\n\\n Gilton points primarily to our holding in Grant to support his contention that\\n\\nthe affidavit was so lacking in indicia of probable cause that no reasonable officer\\n\\nwould depend on the warrant. See 682 F.3d at 836–41. In Grant, the police\\n\\nobtained a warrant to search Grant’s home nine months after the homicide. Id. at\\n\\n828. They had “no evidence suggesting that Grant was involved.” Id. at 832.\\n\\nRather, the police suspected Grant’s two sons. Police had no evidence that one of\\n\\nthe boys had visited his father after the homicide, id. at 833, and only scant\\n\\nevidence that the second son had visited his father some six months after the\\n\\nmurder (and two months prior to the search), id. at 833–34. There was no evidence\\n\\nthat the murder weapon was in Grant’s home. Id. In fact, the police found nothing\\n\\n\\n 18\\n\\fin Grant’s home related to the murder, but charged him with being a felon in\\n\\npossession of other firearms. Id. We concluded that there was no “plausible\\n\\nconnection between Grant’s house and any evidence of the murder” and\\n\\ndetermined that “the magistrate so obviously erred in approving the warrant that\\n\\nthe officers executing it could not have relied on it in good faith.” Id. at 841\\n\\n(quotation marks and citation omitted). Gilton argues that as in Grant, the\\n\\naffidavit’s sparse references to him rendered the magistrate’s error “so obvious[]”\\n\\nthat the officers executing the warrant “could not have relied on it in good faith.”\\n\\nSee id.\\n\\n Grant is distinguishable from this case. Here, Gilton was more closely\\n\\nconnected to the suspected crime than the father in Grant. The affidavit here\\n\\nindicated that Gilton had contact with L.G. up until the day preceding the murder.\\n\\nThe police suspected that the murder was family related, stemming from Sneed’s\\n\\npromoting the prostitution of the under-aged L.G., and coordinated by at least two\\n\\nfamily members. Thus, rather than appearing as a last-ditch, needle-in-the-\\n\\nhaystack effort—as the search in Grant did—the conclusion that Gilton might have\\n\\nbeen involved the events leading to Sneed’s murder was not at all far-fetched. To\\n\\nhold that the police could not have relied in good faith on the magistrate’s\\n\\ndetermination here would be to “[p]enaliz[e] the officer for the magistrate’s error,\\n\\n\\n 19\\n\\frather than his own,” and thus “cannot logically contribute to the deterrence of\\n\\nFourth Amendment violations.” See Leon, 468 U.S. at 921.\\n\\n This is particularly true in light of the fact that in 2012, no circuit court had\\n\\nyet held the Fourth Amendment applicable to CSLI data. See United States v.\\n\\nThompson, 866 F.3d 1149 (10th Cir. 2017), vacated 138 S. Ct. 2706 (2018);\\n\\nUnited States v. Stimler, 864 F.3d 253 (3d Cir. 2017), vacated in part by United\\n\\nStates v. Goldstein, 902 F.3d 411 (3d Cir. 2018); United States v. Graham, 824\\n\\nF.3d 421 (4th Cir. 2016) (en banc), abrogated by Carpenter, 138 S. Ct. 2206\\n\\n(2018); United States v. Carpenter, 819 F.3d 880 (6th Cir. 2016), rev’d 138 S. Ct.\\n\\n2206 (2018); United States v. Davis, 785 F.3d 498 (11th Cir. 2015) (en banc),\\n\\nabrogated by Carpenter, 138 S. Ct. 2206 (2018); In re Application of U.S. for\\n\\nHistorical Cell Site Data, 724 F.3d 600 (5th Cir. 2013), abrogated by Carpenter,\\n\\n138 S. Ct. 2206 (2018). In light of the prevailing belief in 2012 that CSLI data was\\n\\nnot protected by the Fourth Amendment, we conclude that there was no “willful”\\n\\nor “grossly negligent” error here where the officers nevertheless took the\\n\\nprecautionary step of seeking a warrant and provided ample factual background by\\n\\nwhich the magistrate could reach his own determination of the existence of\\n\\nprobable cause. Although we conclude that the magistrate’s determination was\\n\\nerroneous, we hold that the police here were not required to second-guess the\\n\\n\\n 20\\n\\fdetermination of a neutral and detached magistrate. As such, we conclude that\\n\\napplication of the exclusionary rule to Gilton’s CSLI data would have no\\n\\n“appreciable deterrent” effect and is thus unwarranted.3\\n\\n III. CONCLUSION\\n\\n We REVERSE the district court’s order granting Gilton’s motion to\\n\\nsuppress.\\n\\n\\n\\n\\n 3\\n Having concluded that application of the good faith exception is warranted\\nunder Leon, we do not reach the government’s arguments that the good faith\\nexception is also warranted on the basis of Davis v. United States, 564 U.S. 229\\n(2011) (upholding the good faith exception where officers reasonably relied on\\nbinding appellate precedent) or Illinois v. Krull, 480 U.S. 340 (1987) (upholding\\nthe good faith exception where officers reasonably relied on a statute).\\n 21\\n\\f FILED\\nUSA V. ANTONIO GILTON, No. 16-10109 MAR 4 2019\\n MOLLY C. DWYER, CLERK\", \"author\": \"BYBEE, Circuit Judge:\"}, {\"type\": \"dissent\", \"author\": \"MCKEOWN\", \"text\": \"MCKEOWN, J., dissenting U.S. COURT OF APPEALS\\n\\n\\n I respectfully dissent. The warrant affidavit for Antonio Gilton’s cell-site\\n\\nlocation information (“CSLI”) so thoroughly lacked probable cause that it was\\n\\nobjectively unreasonable for the officer to have relied on it. The affidavit’s only\\n\\nstatement vaguely implicating Antonio was a suggestion that a Gilton family\\n\\nmember may have been involved in the murder. As any reasonable officer should\\n\\nhave known, “none of the facts in the affidavit, singly or en masse, provide a\\n\\nreasonable basis from which to infer that” Gilton’s CSLI connected him to the\\n\\nmurder. United States v. Grant, 682 F.3d 827, 841 (9th Cir. 2012). Weak\\n\\ninferences from vague facts do not amount to probable cause as to specific\\n\\nindividuals. These are precisely the circumstances where the good faith exception\\n\\ncannot save a defective warrant. Thus, I would affirm the district court’s order\\n\\ngranting Gilton’s motion to suppress.\\n\\n I join the majority’s analysis in Part II.A with respect to the absence of\\n\\nprobable cause. As the majority and district court concluded, “the [warrant]\\n\\naffidavit’s passing references to Gilton are insufficient to support a reasonable\\n\\ninference that evidence of a crime would be found in his CSLI data.” Maj. Op. at\\n\\n14–15. Taking a closer look at the warrant reveals why the good faith exception\\n\\ndoes not apply.\\n\\n 1\\n\\f The warrant was issued as part of the San Francisco Police Department’s\\n\\ninvestigation of the murder of Calvin Sneed. Sneed was shot and killed in San\\n\\nFrancisco. His minor girlfriend, L.G., who was waiting at her parents’ house for\\n\\nSneed to pick her up, was found standing next to his car after the attack. L.G.’s\\n\\nmother had recently gone to Los Angeles to persuade her to move home because\\n\\nSneed was pimping L.G. and other girls. In Los Angeles, L.G. lived with Antonio,\\n\\nwhom she referred to as her “brother,” though he was a cousin. Police believed\\n\\nthat someone in the family, most likely L.G.’s father, Barry Gilton, was\\n\\nresponsible for the murder. In the affidavit, there are only three references to\\n\\nAntonio:\\n\\n • L.G. said she had been living with Antonio in Los Angeles.\\n\\n • One of the numbers in L.G.’s phone was Antonio’s cell number.\\n\\n • L.G. identified the number as belonging to Antonio.\\n\\n These unremarkable facts do nothing to tie Antonio to the murder—and they\\n\\nare even more benign when considering Antonio’s familial relationship with L.G.\\n\\nNothing indicates Antonio, who was living in Los Angeles, was in or near San\\n\\nFrancisco at the time, that he had any connection to the home where L.G.’s parents\\n\\nlived, or that he had any connection to Sneed or knowledge of his pimping\\n\\nactivities. The only link was that Antonio was a member of the extended Gilton\\n\\n\\n\\n 2\\n\\ffamily. Conspicuously absent was any statement by the officer that he believed\\n\\nthat Antonio’s CSLI would lead to information about the Sneed shooting.\\n\\n Even accepting the government’s suggestion that the affidavit supported a\\n\\nreasonable inference about a family-based killing, that inference does not support\\n\\nany plausible connection to Antonio. In the course of their investigation, the police\\n\\nreceived confidential information indicating that two family members were\\n\\ninvolved in the shooting. That information implicated L.G.’s father, Barry Gilton,\\n\\nand indicated that the second family member was a woman. Nothing suggested\\n\\ntwo male family members were involved. Thus, when the warrant issued, any\\n\\nrelationship between Antonio and the murder was purely speculative.\\n\\n The flimsy basis for the warrant is central to analyzing the government’s\\n\\ninvocation of the good faith exception. The majority cites the appropriate standard\\n\\nunder United States v. Leon, 468 U.S. 897 (1984), but then gives it a rote\\n\\napplication. Good faith reliance is the pivotal question. As Leon teaches,\\n\\nsuppression “remains an appropriate remedy” when “a warrant [is] based on an\\n\\naffidavit ‘so lacking in indicia of probable cause as to render official belief in its\\n\\nexistence entirely unreasonable.’” Id. at 923 (quoting Brown v. Illinois, 422 U.S.\\n\\n590, 611 (1975)). Simply having a warrant is not a free pass or substitute for good\\n\\nfaith.\\n\\n\\n\\n\\n 3\\n\\f We rejected good faith reliance in Grant because the target had no\\n\\n“independent connection to the homicide” and his limited association “through\\n\\nfamily and gang affiliation” provided an insufficient link to the crime. Grant, 682\\n\\nF.3d at 836–37. In analyzing good faith, we circled back to the absence of\\n\\nprobable cause, noting that Grant’s link with another family member was “so weak\\n\\nas to be unreasonable to rely on for probable cause.” Id. at 836. This weak link\\n\\nfared no better under Leon’s analysis.\\n\\n Antonio’s case is no different. Although Grant involved a warrant issued\\n\\nnine months after the murder, the timing of the Gilton warrant—two days after the\\n\\nmurder—does nothing to buttress good faith. What matters is that the warrant\\n\\nincluded only the three innocuous facts about Antonio. As we know from Grant,\\n\\n“[a] reasonable officer would know that probable cause is not supplied by stating\\n\\neverything one knows about a particular item one would like to find to solve a\\n\\nmurder.” Id. at 841. Grant fits Antonio’s situation to a tee—the mere family\\n\\nconnection, which is all that was present here, is simply not enough to justify\\n\\napplying the good faith exception.\\n\\n An alternative ground for applying the good faith rule is “when the police\\n\\nconduct a search in compliance with binding precedent that is later overruled.”\\n\\nDavis v. United States, 564 U.S. 229, 232 (2011). This relatively narrow\\n\\n\\n\\n\\n 4\\n\\fexpansion of the good faith exception has no bearing on this case and the majority\\n\\nexplicitly notes it does not apply Davis. Maj. Op. at 23, n.3.\\n\\n Yet the majority blesses the officer’s good faith reliance on the warrant\\n\\nbased in part on Carpenter v. United States, which was handed down during this\\n\\nappeal. 138 S. Ct. 2206 (2018). In Carpenter, the Supreme Court held that\\n\\nacquiring CSLI is a search requiring a warrant supported by probable cause. Id. at\\n\\n2217, 2221. In its use of Carpenter, the majority misses the point. Carpenter only\\n\\naffects a question not at issue here: the warrant requirement for CSLI. As we all\\n\\nagree, the police got a warrant for Antonio’s cell data. What they did not do is rely\\n\\nin good faith on the probable cause determination. Carpenter did not alter that\\n\\ndoctrine, and it is not relevant to our inquiry.\\n\\n Even under the Davis rule, Carpenter does not provide a relevant change in\\n\\nlaw permitting the application of the good faith exception. The notion that\\n\\nCarpenter put into play the Supreme Court’s precedent that “a person has no\\n\\nlegitimate expectation of privacy in information he voluntarily turns over to third\\n\\nparties,” Smith v. Maryland, 442 U.S. 735, 743–44 (1979), was laid to rest by the\\n\\nCourt’s clear statement that “[w]e . . . decline to extend Smith and [United States\\n\\nv.] Miller[, 425 U.S. 435 (1976)] to the collection of CSLI.” Carpenter, 138 S. Ct.\\n\\nat 2220. And it is surely not clear that cell phone customers “voluntarily” share\\n\\ntheir location data with their cellular provider “in any meaningful way.” In re\\n\\n 5\\n\\fApplication of the U.S. for an Order Directing a Provider of Elec. Commc’n Serv.\\n\\nto Disclose Record to Gov’t, 620 F.3d 304, 317 (3d Cir. 2010). The best the\\n\\ngovernment can claim is lack of clarity; no appellate precedent has been overruled.\\n\\n Nonetheless, the majority shoehorns Carpenter into a watered-down\\n\\napplication of Davis’ good faith exception. The majority argues that, before\\n\\nCarpenter, many circuits thought acquiring CSLI did not require a warrant, so\\n\\nthere is no significant error here because the police sought a warrant and then\\n\\nrelied on the state of the law that did not protect CSLI. This bootstrapping ignores\\n\\nDavis’ requirement that officers conduct their searches in line with relevant,\\n\\nbinding precedent. The majority dilutes that rule and expands the good faith\\n\\nexception beyond Davis’ narrow boundaries without providing a clear limiting\\n\\nprinciple. There was no change in precedent related to probable cause, which is\\n\\nthe only precedent relevant here because the police already had gone the warrant\\n\\nroute. Compounding the problem, the majority asserts that it does not rely on\\n\\nDavis. Maj. Op. at 23, n.3. Thus, we are left to understand the majority’s\\n\\nexpansion of the good faith exception as unhinged from Davis, when in fact it\\n\\nerodes the narrow limits announced in that case.\\n\\n Recognizing the tough hurdle under Davis, the government falls back on the\\n\\nStored Communications Act, which permits retrieval of phone data under relaxed\\n\\nstandards. 18 U.S.C. § 2703(d). Never mind that the police did not seek or even\\n\\n 6\\n\\fallude to a Section 2703(d) order for Antonio’s CSLI. Under the circumstances\\n\\nhere, Leon’s good faith exception benchmarks the officer’s good faith reliance on\\n\\nthe actual warrant issued by the state court in San Francisco, not a hypothetical,\\n\\nalternative order that might have issued under a federal statute never referenced to\\n\\nthe state court. Such an expansion of Leon is unsupported.\\n\\n I agree with the district court’s grant of Antonio Gilton’s motion to suppress\\n\\nthe cell phone evidence. It was the right answer and one that faithfully enforces\\n\\nthe purpose of deterrence with respect to an obviously defective warrant.\\n\\n\\n\\n\\n 7\"}]}"},"judge1":{"kind":"string","value":"J CLIFFORD WALLACE"},"judge2":{"kind":"string","value":"M MARGARET MCKEOWN"},"judge3":{"kind":"string","value":"JAY S BYBEE"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":2,"string":"2"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":1,"string":"1"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4595922/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. Antonio Gilton\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-04\nOpinion type: majority\nHas dissenting opinion: Yes\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFILED\n FOR PUBLICATION\n MAR 04 2019\n UNITED STATES COURT OF APPEALS MOLLY C. DWYER, CLERK\n U.S. COURT OF APPEALS\n\n\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 16-10109\n\n Plaintiff-Appellant, D.C. No.\n 3:13-cr-00764-WHO-1\nv.\n\nREGINALD ELMORE, AKA Fat Reg; OPINION\nESAU FERDINAND, AKA Sauce,\n\n Defendants,\n\nBARRY GILTON, AKA Prell; ADRIAN\nGORDON, AKA Tit; MONZELL\nHARDING, Jr.; CHARLES HEARD,\nAKA Cheese; LUPE MERCADO; PAUL\nROBESON, AKA P World; ALFONZO\nWILLIAMS, AKA Fonz, AKA Relly;\nJAQUAIN YOUNG, AKA Loc,\n\n Defendants,\n\nand\n\nANTONIO GILTON, AKA TG, AKA\nTone,\n\n Defendant-Appellee.\n\n\n Appeal from the United States District Court\n for the Northern District of California\n William Horsley Orrick, District Judge, Presiding\n\f Argued and Submitted January 7, 2019\n San Diego, California\n\nBefore: J. Clifford Wallace, M. Margaret McKeown, and Jay S. Bybee, Circuit\nJudges.\n\n Opinion by Judge Bybee\n\nBYBEE, Circuit Judge:\n\n Following the June 2012 murder of Calvin Sneed, police obtained a warrant\n\nauthorizing the seizure of Antonio Gilton’s historical cell-site location information.\n\nGilton was subsequently charged with four counts relating to the murder of Sneed.\n\nHe moved to suppress the location data, arguing that the warrant issued without\n\nprobable cause and that the officers’ reliance on the warrant was not in good faith.\n\nThe district court granted Gilton’s motion, concluding that the warrant was so\n\ndeficient in indicia of probable cause that no officer could have relied on the\n\nwarrant in good faith.\n\n Although we agree with the district court that the warrant authorizing the\n\nseizure of Gilton’s location data was not supported by probable cause, we conclude\n\nthat the deficiencies were not so stark as to render the officers’ reliance on the\n\nwarrant “entirely unreasonable.” See United States v. Leon, 468 U.S. 897, 923\n\n(1984). We reverse.\n\n I. FACTS AND PROCEEDINGS\n\n\n 2\n\f A\n\n At approximately 2:00 a.m. on June 4, 2012, officers of the San Francisco\n\nPolice Department responded to a report of shots fired in the area of Meade and Le\n\nConte Avenues, San Francisco. When the officers arrived, they found Calvin\n\nSneed slumped in the driver’s seat of a crashed Toyota Camry with a gunshot\n\nwound to his head. Sneed was later pronounced dead.\n\n Standing distraught next to the car was Sneed’s minor girlfriend, L.G. She\n\ntold the police that approximately eight months before the shooting, she had moved\n\nfrom San Francisco to Los Angeles for a “new start,” and that she had been staying\n\nwith her “elder brother,” Antonio Gilton (hereinafter “Gilton”).1 In Los Angeles,\n\nL.G. met and began dating Calvin Sneed. L.G. subsequently learned that Sneed\n\nwas pimping young women for prostitution in the Los Angeles area, and she began\n\nto advertise herself on various prostitution websites.\n\n L.G.’s friends and family eventually discovered what she was doing, and\n\nL.G.’s mother traveled to Los Angeles on May 31, 2012, to persuade L.G. to return\n\nto San Francisco. L.G. did not return with her mother. Three days later, however,\n\n\n\n\n 1\n On appeal, Antonio Gilton informs us that he is actually L.G.’s cousin, not\nher older brother. All parties agree, however, that L.G. was referring to Antonio\nGilton when she mentioned her “elder brother.”\n 3\n\fL.G. and Sneed traveled together to San Francisco, where Sneed dropped L.G. off\n\nat her parents’ house around 4:00 p.m.\n\n L.G. stayed with her parents until approximately 12:15 a.m. on June 4, 2012,\n\nwhen she texted Sneed to come and pick her up. L.G. began to argue with her\n\nmother about returning to Los Angeles with Sneed, but her father merely told her\n\n“you grown” and instructed her “before you leave, turn the lights off” before he\n\nexited the room.\n\n Around 1:56 a.m., L.G. was waiting for Sneed outside her parents’ home and\n\nnoticed a silver SUV parked nearby with its lights on. As Sneed’s car arrived, the\n\nsilver SUV drove off, but as Sneed drove past where L.G. was standing and turned\n\nthe corner, L.G. saw the SUV reappear and accelerate towards Sneed’s vehicle.\n\nL.G. heard gunshots and saw a muzzle flash coming from the SUV. L.G. then\n\nheard a crash and ran to Sneed’s car, where she found him “slumped in the driver’s\n\nseat with a gunshot wound to his head.” L.G. cooperated with police and allowed\n\nthem to search her cell phone. During the search, the police identified and L.G.\n\nconfirmed cell phone numbers for her father, her mother, Gilton, and L.G.’s\n\nyounger brother.\n\n Later that day, the police received confidential information implicating\n\nL.G.’s father, Barry Gilton, and a second, unidentified individual in the murder.\n\n\n 4\n\fAs a result, the police obtained historical cell-site location information (“CSLI”)\n\ndata for Barry Gilton’s phone pursuant to an exigent circumstances request to T-\n\nMobile.2 Although Barry Gilton told police that he had been at home in bed after\n\n12:15 a.m. on the night of the murder, the CSLI data indicated that between 12:49\n\na.m. and 2:19 a.m. that night, Barry Gilton’s cell phone traveled from near his\n\nhome to the Western Addition. The cell phone then returned to the vicinity of the\n\nGilton home around the time of the murder before traveling towards the northern\n\narea of the Mission after the shooting. The police also obtained video surveillance\n\nfrom a camera near the site of the murder that showed a light colored mid-size\n\nSUV believed to be the vehicle used in the shooting.\n\n Based on all the information set forth above, San Francisco Police\n\nDepartment (SFPD) Sergeant Gary Watts submitted a fourteen-page affidavit in\n\nsupport of a state search warrant. The application sought CSLI data for two cell\n\n\n\n 2\n Cell sites usually consist of a set of radio antennas mounted on a tower,\nalthough “they can also be found on light posts, flagpoles, church steeples, or the\nsides of buildings.” Carpenter v. United States, 138 S. Ct. 2206, 2211 (2018).\n“Each time [a] phone connects to a cell site, it generates a time-stamped record\nknown as cell-site location information (CSLI).” Id. This CSLI data indicates the\ngeneral geographic area in which the cell phone user was located when his or her\nphone connected to the network. Because most smartphones tap into the wireless\nnetwork “several times a minute whenever their signal is on . . . modern cell\nphones generate increasingly vast amounts of increasingly precise CSLI.” Id. at\n2211–12.\n 5\n\fphone numbers, a phone associated with an unknown individual and the phone\n\nassociated with Gilton. Sergeant Watts laid out in some detail what the police\n\nknew about the crime, Sneed’s relationship with L.G., and information learned\n\nfrom the confidential informant. He related how L.G. was related to Gilton, that\n\nshe had been staying with Gilton in Los Angeles (where she met Sneed and began\n\nengaging in prostitution), that Gilton’s phone number was in her cell phone, and\n\nthat the murder was likely committed by a family member or members. Watts\n\naverred that “there appear[ed] to be probable cause to believe that the cell phone\n\nnumbers provided [would] tend to show . . . possible first-hand knowledge of those\n\npersons responsible for the shooting of . . . Calvin Sneed” and that “the cell-site\n\ntower locations used on the date and times listed could possibly lead to the proper\n\nidentity and the whereabouts of additional persons associated with this crime.”\n\nWatts also stated that he had “discussed the merits of the case with the District\n\nAttorney’s Office.”\n\n That same day, June 6, 2012, a judge of the Superior Court of California,\n\nCounty of San Francisco issued a warrant to Sprint for the seizure of cell phone\n\nrecords for Gilton’s phone number. The warrant identified three categories of\n\ninformation for seizure: (1) subscriber and billing information; (2) all incoming\n\nand outgoing calls and text messages from the period of May 1, 2012, to June 6,\n\n\n 6\n\f2012; and (3) cell-site location information (CSLI). Only the third category—the\n\nCSLI data—is at issue in this appeal.\n\n B\n\n Gilton was indicted by a federal grand jury in 2013, on four counts related to\n\nSneed’s murder. In 2015, Gilton filed a motion to suppress the CSLI obtained by\n\nthe San Francisco police pursuant to the Sprint warrant. The district court granted\n\nGilton’s motion to suppress. The court concluded first that Gilton had a reasonable\n\nexpectation of privacy in the historical CSLI he sought to suppress. Accordingly,\n\nthe court determined that probable cause was required to obtain that CSLI from\n\nSprint. The court concluded, however, that “[t]he affidavit in support of the Sprint\n\nwarrant plainly failed to provide a substantial basis for concluding that there was\n\nprobable cause to search” because the affidavit “hardly mention[ed] Gilton” and\n\ndid not provide a substantial basis for inferring that Gilton was in the San\n\nFrancisco area at the time of the shooting. Even assuming that the warrant\n\nsupported an inference of a “family-based attack,” the district court concluded that\n\nthe “facts pointed to one particular family member being involved: [Barry] Gilton,\n\nnot [Antonio] Gilton.” Finally, the district court concluded that the “good faith”\n\nexception did not apply because “it was entirely unreasonable to believe that the\n\n\n\n\n 7\n\faffidavit’s passing, innocuous references to A. Gilton established probable cause to\n\nobtain his cell phone data.”\n\n The government appealed the district court’s order granting the motion to\n\nsuppress. We originally heard argument in March 2017, but then withdrew the\n\ncase from submission pending the final decision of the Supreme Court in United\n\nStates v. Carpenter, 138 S. Ct. 2206 (2018). Following the Court’s resolution of\n\nCarpenter, we requested supplemental briefing and heard reargument in January\n\n2019. We agree with the district court that the police lacked probable cause to\n\nobtain Gilton’s CSLI data, but we disagree with the district court’s conclusion that\n\nthe good faith exception does not apply. Accordingly, we reverse.\n\n II. ANALYSIS\n\n In Carpenter, the Supreme Court held that “an individual maintains a\n\nlegitimate expectation of privacy in the record of his physical movements as\n\ncaptured through CSLI.” 138 S. Ct. at 2217. Consequently, Fourth Amendment\n\nprotections generally require the government to “obtain a warrant supported by\n\nprobable cause before acquiring [CSLI] records.” Id. at 2221.\n\n Here, the government did in fact obtain a warrant authorizing the acquisition\n\nof Gilton’s CSLI data. Our analysis is thus confined to the questions of whether\n\nthat warrant was supported by probable cause, and, if not, whether the search\n\n\n 8\n\fshould nevertheless be upheld on the basis of the officers’ good faith reliance on\n\nthe warrant.\n\n A\n\n We turn first to the issue of whether the Sprint warrant was supported by\n\nprobable cause. Probable cause exists where the totality of the circumstances\n\nindicates a “fair probability that . . . evidence of a crime will be found in a\n\nparticular place.” Illinois v. Gates, 462 U.S. 213, 238 (1983). This standard does\n\nnot require the affidavit to establish that the evidence is in fact in the place to be\n\nsearched, or even that it is more likely than not to be there. United States v.\n\nFernandez, 388 F.3d 1199, 1254 (9th Cir. 2004), modified 425 F.3d 1248 (9th Cir.\n\n2005). Rather, the issuing judge “need only conclude that it would be reasonable\n\nto seek the evidence in the place indicated in the affidavit.” Id. (quotation\n\nomitted).\n\n In general, the issuing judge’s finding of probable cause is entitled to “great\n\ndeference.” United States v. Krupa, 658 F.3d 1174, 1177 (9th Cir. 2011).\n\n“[N]otwithstanding the deference that magistrates deserve,” however, we may\n\ndetermine that a warrant was invalid where “the magistrate’s probable-cause\n\ndetermination reflected an improper analysis of the totality of the circumstances.”\n\n\n\n\n 9\n\fLeon, 468 U.S. at 915. In addition, the decision of the magistrate “cannot be a\n\nmere ratification of the bare conclusions of others.” Gates, 462 U.S. at 239.\n\n Here, police sought a warrant to obtain subscriber and billing information,\n\nincoming and outgoing calls, incoming and outgoing texts, and CSLI information\n\nfor phones belonging to Gilton and a second, unidentified person. The application\n\nwas supported by a lengthy affidavit describing the results of interviews, video\n\nsurveillance, and an exigent circumstances request to T-Mobile. The affidavit thus\n\nprovided the magistrate with a basis for determining the existence of probable\n\ncause, see id., and the magistrate’s finding of probable cause is appropriately\n\nentitled to deference, see Krupa, 658 F.3d at 1177.\n\n Although the affidavit explained in some detail the need for the information\n\nfrom the unidentified person’s phone, the affidavit mentions Gilton only three\n\ntimes. First, the affidavit records L.G.’s testimony that she “was staying with her\n\nelder brother in L.A.” Second, the affidavit indicates that one of the phone\n\nnumbers discovered in L.G.’s phone belonged to Gilton. Third, the affidavit states\n\nthat L.G. identified which number belonged to Gilton. As the district court pointed\n\nout, the affidavit does not indicate that police had reason to believe that Gilton\n\n“was . . . in or around San Francisco on or around June 4, 2012.”\n\n\n\n\n 10\n\f The affidavit’s scant and innocuous references to Gilton do not establish a\n\n“fair probability” that evidence of the crime would be found in Gilton’s location\n\ndata. Rather, these facts merely indicate that L.G. had the phone number of a\n\nfamily member with whom she had lived in Los Angeles. Surely, it is common for\n\nan adolescent to store the phone number of the relative with whom she is living,\n\nand the mere existence of a familial connection between L.G. and Gilton is not\n\nsufficient to render it “reasonable” to search for evidence of the crime in Gilton’s\n\nlocation data. See United States v. Grant, 682 F.3d 827, 836–37 (9th Cir. 2012)\n\n(stating that an association “through family . . . affiliation” is insufficient to\n\nestablish probable cause).\n\n The government depends on two related inferences to support the finding of\n\nprobable cause. First, the government contends that the totality of the\n\ncircumstances supports an inference that “[t]he murder of Calvin Sneed was a\n\nfamily solution to a family problem.” Second, because Gilton is related to L.G.\n\nand because L.G. met Sneed while living with Gilton, the government asserts that\n\nthe magistrate could reasonably infer that Gilton was well aware of Sneed’s\n\nrelationship with L.G., that Gilton had reason to be upset with Sneed, and that he\n\nhad a motive to commit the crime. The government argues that these two\n\ninferences tie Gilton to the murder. We disagree.\n\n\n 11\n\f First, we agree with the district court that the affidavit “pointed to one\n\nparticular family member being involved in the attack: B. Gilton, not A. Gilton.”\n\nAlthough the affidavit indicated that more than one person was likely involved in\n\nthe shooting and that the second person might have been a family member, nothing\n\nin the affidavit suggested that the identity of the second person was Gilton. Nor do\n\nwe agree with the government’s contention that a magistrate could reasonably\n\nbelieve that Gilton was the accomplice because of his supposed motive. As the\n\ndistrict court pointed out, there is no evidence in the affidavit that Gilton “had\n\ncommunicated with family members (or anyone else) about [L.G.’s] relationship\n\nwith Sneed,” although that surely was one of the reasons police wanted Gilton’s\n\ncall and text records. There is thus no basis in the affidavit to support the inference\n\nthat Gilton was upset with Sneed, that he had motive to commit the crime, and,\n\nimportantly, that he was in the San Francisco area on the night of the murder. We\n\nconclude, as did the district court, that the affidavit’s passing references to Gilton\n\nare insufficient to support a reasonable inference that evidence of a crime would be\n\nfound in his CSLI data.\n\n B\n\n Having determined that the warrant was not supported by probable cause,\n\nwe turn now to the question of whether the officers nevertheless relied in good\n\n\n 12\n\ffaith on the warrant they obtained. The Supreme Court has made quite clear that\n\n“[t]he Fourth Amendment contains no provision expressly precluding the use of\n\nevidence obtained in violation of its commands.” Leon, 468 U.S. at 906. Because\n\n“[t]he wrong condemned by the Amendment is ‘fully accomplished’ by the\n\nunlawful search or seizure itself, . . . the exclusionary rule is neither intended nor\n\nable to ‘cure the invasion of the defendant’s rights which he has already suffered.’”\n\nId. (first quoting United States v. Calandra, 414 U.S. 338, 354 (1974); then\n\nquoting Stone v. Powell, 428 U.S. 465, 540 (1976)). As a result, “the use of fruits\n\nof a past unlawful search or seizure ‘work[s] no new Fourth Amendment wrong.’”\n\nId. (alteration in original) (quoting Calandra, 414 U.S. at 354).\n\n The exclusionary rule is thus not “a personal constitutional right of the party\n\naggrieved,” but rather “a judicially created remedy designed to safeguard Fourth\n\nAmendment rights generally through its deterrent effect.” Calandra, 414 U.S. at\n\n348. As such, the question of “[w]hether the exclusionary sanction is appropriately\n\nimposed in a particular case” is an entirely separate issue “from the question\n\nwhether the Fourth Amendment rights of the party seeking to invoke the rule were\n\nviolated by police conduct.” Leon, 468 U.S. at 906 (quoting Gates, 462 U.S. at\n\n223). Thus, having determined that the SFPD’s search of Gilton’s CSLI data\n\n\n\n\n 13\n\fviolated his Fourth Amendment rights, we must address separately the question of\n\nwhether exclusion is an appropriate remedy.\n\n The question of application of the exclusionary rule is answered by weighing\n\n“[t]he substantial social costs exacted by the exclusionary rule” with the benefit of\n\nincreased deterrence of police misconduct. Id. at 906–07. “If . . . the exclusionary\n\nrule does not result in appreciable deterrence” of police misconduct, “then, clearly,\n\nits use in the instant situation is unwarranted.” United States v. Janis, 428 U.S.\n\n433, 454 (1976). To have any appreciable deterrent benefit, the exclusion of\n\nevidence “must alter the behavior of individual law enforcement officers or the\n\npolicies of their departments.” Leon, 468 U.S. at 918.\n\n In Leon, the Supreme Court took up the issue of whether the exclusionary\n\nrule should be applied in cases where the police seized evidence “in reasonable,\n\ngood-faith reliance on a search warrant that is subsequently held to be defective.”\n\nId. at 905. The Court reasoned that “[t]he deterrent purpose of the exclusionary\n\nrule necessarily assumes that the police have engaged in willful, or at the very least\n\nnegligent, conduct which has deprived the defendant of some right,” and that\n\n“[w]here the official action was pursued in complete good faith . . . the deterrence\n\nrationale loses much of its force.” Id. at 919 (quoting United States v. Peltier, 422\n\nU.S. 531, 539 (1975)). In such cases, “excluding the evidence will not further the\n\n\n 14\n\fends of the exclusionary rule in any appreciable way.” Id. at 920 (quoting Stone,\n\n428 U.S. at 539–40).\n\n The Court emphasized that “[t]his is particularly true . . . when an officer\n\nacting with objective good faith has obtained a search warrant from a judge or\n\nmagistrate and acted within its scope.” Id. Because “[i]t is the magistrate’s\n\nresponsibility to determine whether the officer’s allegations establish probable\n\ncause[,] . . . [i]n the ordinary case, an officer cannot be expected to question the\n\nmagistrate’s probable cause determination.” Id. at 921. As a result, “[i]n most\n\nsuch cases, there is no police illegality” and “[p]enalizing the officer for the\n\nmagistrate’s error, rather than his own, cannot logically contribute to the deterrence\n\nof Fourth Amendment violations.” Id. at 920–21. The Court concluded that “the\n\nmarginal or nonexistent benefits produced by suppressing evidence obtained in\n\nobjectively reasonable reliance on a subsequently invalidated search warrant\n\ncannot justify the substantial costs of exclusion.” Id. at 922.\n\n The existence of a search warrant does not automatically preclude\n\napplication of the exclusionary rule, however. Rather, the Court instructed that we\n\nshould conduct our analysis “on a case-by-case basis,” id. at 918, and recognized\n\nfour exceptions to the exception where suppression “remains an appropriate\n\nremedy,” id. at 923. First, in cases where “the magistrate or judge . . . was misled\n\n\n 15\n\fby information in an affidavit that the affiant knew was false or would have known\n\nwas false except for his reckless disregard of the truth.” Id. Second, “in cases\n\nwhere the issuing magistrate wholly abandoned his judicial role” by acting as “an\n\nadjunct law enforcement officer” or mere “rubber stamp” for the police. Id. at 914,\n\n923 (citing Lo-Ji Sales, Inc. v. New York, 442 U.S. 319 (1979)). Third, in cases\n\nwhere “an affidavit [is] so lacking in indicia of probable cause as to render official\n\nbelief in its existence entirely unreasonable.” Id. at 923 (citation omitted). Finally,\n\nin cases where the warrant is “so facially deficient—i.e., in failing to particularize\n\nthe place to be searched or the things to be seized—that the executing officers\n\ncannot reasonably presume it to be valid.” Id.\n\n Here, the parties do not urge and the evidence does not support a conclusion\n\nthat the officers wilfully mislead the magistrate, that the magistrate wholly\n\nabdicated his role as a neutral and detached judicial officer, or that the warrant was\n\nfacially deficient. Thus, the sole question before us is whether Officer Watts’s\n\naffidavit was “so lacking in indicia of probable cause” that no reasonable officer\n\nwould believe that probable cause existed to search Gilton’s CSLI data. We\n\nconclude that the deficiencies in probable cause discussed in subpart II.A., supra,\n\nare not so stark as to render official belief in the existence of probable cause\n\n“entirely unreasonable.”\n\n\n 16\n\f First, we observe that the application for the warrant was issued only two\n\ndays after the murder. Second, the affidavit in support of the application was\n\nlengthy and laid out for the magistrate what the police knew about the murder and\n\nhow they had obtained the information they had, including the results of interviews\n\nwith family members and witnesses, video surveillance from a camera on a house\n\nnear the murder scene, information received from an anonymous caller who may\n\nhave had first-hand facts, and cell phone information obtained on an emergency\n\nbasis from the T-Mobile Law Enforcement Center. Third, the police suspected that\n\nboth L.G.’s father, Barry Gilton, and a female family member had been involved in\n\ninstigating Sneed’s murder. The police also had reason to believe Barry Gilton and\n\nat least one other person were in the car when the shooting occurred. Fourth, the\n\naffidavit indicated that police had discussed the case with the District Attorney’s\n\noffice.\n\n For the reasons we have previously explained, all of these facts did not add\n\nup to probable cause to obtain Gilton’s CSLI data. The police likely had probable\n\ncause to obtain the other information they sought from Gilton’s cellphone—his\n\nincoming and outgoing calls and texts—because L.G. had been living with Gilton\n\nin Los Angeles when she was dating Sneed, and Gilton may well have been in\n\ncontact with L.G.’s family in San Francisco about Sneed. Gilton does not\n\n\n 17\n\fchallenge the scope of the warrant with respect to his calls and texts, only his CSLI\n\ninformation. What the affidavit is missing is an explanation for why police\n\nthought Gilton’s CSLI information was relevant, since there was no\n\nsuggestion—other than the cryptic information that multiple family members were\n\nlikely involved and that there was a second person in the car, who might or might\n\nnot have been a family member—that Gilton was in San Francisco rather than\n\nhome in Los Angeles. But we cannot say that no reasonable officer would have\n\nrelied on the warrant obtained from the Superior Court just two days into their\n\ninvestigation.\n\n Gilton points primarily to our holding in Grant to support his contention that\n\nthe affidavit was so lacking in indicia of probable cause that no reasonable officer\n\nwould depend on the warrant. See 682 F.3d at 836–41. In Grant, the police\n\nobtained a warrant to search Grant’s home nine months after the homicide. Id. at\n\n828. They had “no evidence suggesting that Grant was involved.” Id. at 832.\n\nRather, the police suspected Grant’s two sons. Police had no evidence that one of\n\nthe boys had visited his father after the homicide, id. at 833, and only scant\n\nevidence that the second son had visited his father some six months after the\n\nmurder (and two months prior to the search), id. at 833–34. There was no evidence\n\nthat the murder weapon was in Grant’s home. Id. In fact, the police found nothing\n\n\n 18\n\fin Grant’s home related to the murder, but charged him with being a felon in\n\npossession of other firearms. Id. We concluded that there was no “plausible\n\nconnection between Grant’s house and any evidence of the murder” and\n\ndetermined that “the magistrate so obviously erred in approving the warrant that\n\nthe officers executing it could not have relied on it in good faith.” Id. at 841\n\n(quotation marks and citation omitted). Gilton argues that as in Grant, the\n\naffidavit’s sparse references to him rendered the magistrate’s error “so obvious[]”\n\nthat the officers executing the warrant “could not have relied on it in good faith.”\n\nSee id.\n\n Grant is distinguishable from this case. Here, Gilton was more closely\n\nconnected to the suspected crime than the father in Grant. The affidavit here\n\nindicated that Gilton had contact with L.G. up until the day preceding the murder.\n\nThe police suspected that the murder was family related, stemming from Sneed’s\n\npromoting the prostitution of the under-aged L.G., and coordinated by at least two\n\nfamily members. Thus, rather than appearing as a last-ditch, needle-in-the-\n\nhaystack effort—as the search in Grant did—the conclusion that Gilton might have\n\nbeen involved the events leading to Sneed’s murder was not at all far-fetched. To\n\nhold that the police could not have relied in good faith on the magistrate’s\n\ndetermination here would be to “[p]enaliz[e] the officer for the magistrate’s error,\n\n\n 19\n\frather than his own,” and thus “cannot logically contribute to the deterrence of\n\nFourth Amendment violations.” See Leon, 468 U.S. at 921.\n\n This is particularly true in light of the fact that in 2012, no circuit court had\n\nyet held the Fourth Amendment applicable to CSLI data. See United States v.\n\nThompson, 866 F.3d 1149 (10th Cir. 2017), vacated 138 S. Ct. 2706 (2018);\n\nUnited States v. Stimler, 864 F.3d 253 (3d Cir. 2017), vacated in part by United\n\nStates v. Goldstein, 902 F.3d 411 (3d Cir. 2018); United States v. Graham, 824\n\nF.3d 421 (4th Cir. 2016) (en banc), abrogated by Carpenter, 138 S. Ct. 2206\n\n(2018); United States v. Carpenter, 819 F.3d 880 (6th Cir. 2016), rev’d 138 S. Ct.\n\n2206 (2018); United States v. Davis, 785 F.3d 498 (11th Cir. 2015) (en banc),\n\nabrogated by Carpenter, 138 S. Ct. 2206 (2018); In re Application of U.S. for\n\nHistorical Cell Site Data, 724 F.3d 600 (5th Cir. 2013), abrogated by Carpenter,\n\n138 S. Ct. 2206 (2018). In light of the prevailing belief in 2012 that CSLI data was\n\nnot protected by the Fourth Amendment, we conclude that there was no “willful”\n\nor “grossly negligent” error here where the officers nevertheless took the\n\nprecautionary step of seeking a warrant and provided ample factual background by\n\nwhich the magistrate could reach his own determination of the existence of\n\nprobable cause. Although we conclude that the magistrate’s determination was\n\nerroneous, we hold that the police here were not required to second-guess the\n\n\n 20\n\fdetermination of a neutral and detached magistrate. As such, we conclude that\n\napplication of the exclusionary rule to Gilton’s CSLI data would have no\n\n“appreciable deterrent” effect and is thus unwarranted.3\n\n III. CONCLUSION\n\n We REVERSE the district court’s order granting Gilton’s motion to\n\nsuppress.\n\n\n\n\n 3\n Having concluded that application of the good faith exception is warranted\nunder Leon, we do not reach the government’s arguments that the good faith\nexception is also warranted on the basis of Davis v. United States, 564 U.S. 229\n(2011) (upholding the good faith exception where officers reasonably relied on\nbinding appellate precedent) or Illinois v. Krull, 480 U.S. 340 (1987) (upholding\nthe good faith exception where officers reasonably relied on a statute).\n 21\n\f FILED\nUSA V. ANTONIO GILTON, No. 16-10109 MAR 4 2019\n MOLLY C. DWYER, CLERK\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. Antonio Gilton\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-04\\nOpinion type: majority\\nHas dissenting opinion: Yes\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFILED\\n FOR PUBLICATION\\n MAR 04 2019\\n UNITED STATES COURT OF APPEALS MOLLY C. DWYER, CLERK\\n U.S. COURT OF APPEALS\\n\\n\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 16-10109\\n\\n Plaintiff-Appellant, D.C. No.\\n 3:13-cr-00764-WHO-1\\nv.\\n\\nREGINALD ELMORE, AKA Fat Reg; OPINION\\nESAU FERDINAND, AKA Sauce,\\n\\n Defendants,\\n\\nBARRY GILTON, AKA Prell; ADRIAN\\nGORDON, AKA Tit; MONZELL\\nHARDING, Jr.; CHARLES HEARD,\\nAKA Cheese; LUPE MERCADO; PAUL\\nROBESON, AKA P World; ALFONZO\\nWILLIAMS, AKA Fonz, AKA Relly;\\nJAQUAIN YOUNG, AKA Loc,\\n\\n Defendants,\\n\\nand\\n\\nANTONIO GILTON, AKA TG, AKA\\nTone,\\n\\n Defendant-Appellee.\\n\\n\\n Appeal from the United States District Court\\n for the Northern District of California\\n William Horsley Orrick, District Judge, Presiding\\n\\f Argued and Submitted January 7, 2019\\n San Diego, California\\n\\nBefore: J. Clifford Wallace, M. Margaret McKeown, and Jay S. Bybee, Circuit\\nJudges.\\n\\n Opinion by Judge Bybee\\n\\nBYBEE, Circuit Judge:\\n\\n Following the June 2012 murder of Calvin Sneed, police obtained a warrant\\n\\nauthorizing the seizure of Antonio Gilton’s historical cell-site location information.\\n\\nGilton was subsequently charged with four counts relating to the murder of Sneed.\\n\\nHe moved to suppress the location data, arguing that the warrant issued without\\n\\nprobable cause and that the officers’ reliance on the warrant was not in good faith.\\n\\nThe district court granted Gilton’s motion, concluding that the warrant was so\\n\\ndeficient in indicia of probable cause that no officer could have relied on the\\n\\nwarrant in good faith.\\n\\n Although we agree with the district court that the warrant authorizing the\\n\\nseizure of Gilton’s location data was not supported by probable cause, we conclude\\n\\nthat the deficiencies were not so stark as to render the officers’ reliance on the\\n\\nwarrant “entirely unreasonable.” See United States v. Leon, 468 U.S. 897, 923\\n\\n(1984). We reverse.\\n\\n I. FACTS AND PROCEEDINGS\\n\\n\\n 2\\n\\f A\\n\\n At approximately 2:00 a.m. on June 4, 2012, officers of the San Francisco\\n\\nPolice Department responded to a report of shots fired in the area of Meade and Le\\n\\nConte Avenues, San Francisco. When the officers arrived, they found Calvin\\n\\nSneed slumped in the driver’s seat of a crashed Toyota Camry with a gunshot\\n\\nwound to his head. Sneed was later pronounced dead.\\n\\n Standing distraught next to the car was Sneed’s minor girlfriend, L.G. She\\n\\ntold the police that approximately eight months before the shooting, she had moved\\n\\nfrom San Francisco to Los Angeles for a “new start,” and that she had been staying\\n\\nwith her “elder brother,” Antonio Gilton (hereinafter “Gilton”).1 In Los Angeles,\\n\\nL.G. met and began dating Calvin Sneed. L.G. subsequently learned that Sneed\\n\\nwas pimping young women for prostitution in the Los Angeles area, and she began\\n\\nto advertise herself on various prostitution websites.\\n\\n L.G.’s friends and family eventually discovered what she was doing, and\\n\\nL.G.’s mother traveled to Los Angeles on May 31, 2012, to persuade L.G. to return\\n\\nto San Francisco. L.G. did not return with her mother. Three days later, however,\\n\\n\\n\\n\\n 1\\n On appeal, Antonio Gilton informs us that he is actually L.G.’s cousin, not\\nher older brother. All parties agree, however, that L.G. was referring to Antonio\\nGilton when she mentioned her “elder brother.”\\n 3\\n\\fL.G. and Sneed traveled together to San Francisco, where Sneed dropped L.G. off\\n\\nat her parents’ house around 4:00 p.m.\\n\\n L.G. stayed with her parents until approximately 12:15 a.m. on June 4, 2012,\\n\\nwhen she texted Sneed to come and pick her up. L.G. began to argue with her\\n\\nmother about returning to Los Angeles with Sneed, but her father merely told her\\n\\n“you grown” and instructed her “before you leave, turn the lights off” before he\\n\\nexited the room.\\n\\n Around 1:56 a.m., L.G. was waiting for Sneed outside her parents’ home and\\n\\nnoticed a silver SUV parked nearby with its lights on. As Sneed’s car arrived, the\\n\\nsilver SUV drove off, but as Sneed drove past where L.G. was standing and turned\\n\\nthe corner, L.G. saw the SUV reappear and accelerate towards Sneed’s vehicle.\\n\\nL.G. heard gunshots and saw a muzzle flash coming from the SUV. L.G. then\\n\\nheard a crash and ran to Sneed’s car, where she found him “slumped in the driver’s\\n\\nseat with a gunshot wound to his head.” L.G. cooperated with police and allowed\\n\\nthem to search her cell phone. During the search, the police identified and L.G.\\n\\nconfirmed cell phone numbers for her father, her mother, Gilton, and L.G.’s\\n\\nyounger brother.\\n\\n Later that day, the police received confidential information implicating\\n\\nL.G.’s father, Barry Gilton, and a second, unidentified individual in the murder.\\n\\n\\n 4\\n\\fAs a result, the police obtained historical cell-site location information (“CSLI”)\\n\\ndata for Barry Gilton’s phone pursuant to an exigent circumstances request to T-\\n\\nMobile.2 Although Barry Gilton told police that he had been at home in bed after\\n\\n12:15 a.m. on the night of the murder, the CSLI data indicated that between 12:49\\n\\na.m. and 2:19 a.m. that night, Barry Gilton’s cell phone traveled from near his\\n\\nhome to the Western Addition. The cell phone then returned to the vicinity of the\\n\\nGilton home around the time of the murder before traveling towards the northern\\n\\narea of the Mission after the shooting. The police also obtained video surveillance\\n\\nfrom a camera near the site of the murder that showed a light colored mid-size\\n\\nSUV believed to be the vehicle used in the shooting.\\n\\n Based on all the information set forth above, San Francisco Police\\n\\nDepartment (SFPD) Sergeant Gary Watts submitted a fourteen-page affidavit in\\n\\nsupport of a state search warrant. The application sought CSLI data for two cell\\n\\n\\n\\n 2\\n Cell sites usually consist of a set of radio antennas mounted on a tower,\\nalthough “they can also be found on light posts, flagpoles, church steeples, or the\\nsides of buildings.” Carpenter v. United States, 138 S. Ct. 2206, 2211 (2018).\\n“Each time [a] phone connects to a cell site, it generates a time-stamped record\\nknown as cell-site location information (CSLI).” Id. This CSLI data indicates the\\ngeneral geographic area in which the cell phone user was located when his or her\\nphone connected to the network. Because most smartphones tap into the wireless\\nnetwork “several times a minute whenever their signal is on . . . modern cell\\nphones generate increasingly vast amounts of increasingly precise CSLI.” Id. at\\n2211–12.\\n 5\\n\\fphone numbers, a phone associated with an unknown individual and the phone\\n\\nassociated with Gilton. Sergeant Watts laid out in some detail what the police\\n\\nknew about the crime, Sneed’s relationship with L.G., and information learned\\n\\nfrom the confidential informant. He related how L.G. was related to Gilton, that\\n\\nshe had been staying with Gilton in Los Angeles (where she met Sneed and began\\n\\nengaging in prostitution), that Gilton’s phone number was in her cell phone, and\\n\\nthat the murder was likely committed by a family member or members. Watts\\n\\naverred that “there appear[ed] to be probable cause to believe that the cell phone\\n\\nnumbers provided [would] tend to show . . . possible first-hand knowledge of those\\n\\npersons responsible for the shooting of . . . Calvin Sneed” and that “the cell-site\\n\\ntower locations used on the date and times listed could possibly lead to the proper\\n\\nidentity and the whereabouts of additional persons associated with this crime.”\\n\\nWatts also stated that he had “discussed the merits of the case with the District\\n\\nAttorney’s Office.”\\n\\n That same day, June 6, 2012, a judge of the Superior Court of California,\\n\\nCounty of San Francisco issued a warrant to Sprint for the seizure of cell phone\\n\\nrecords for Gilton’s phone number. The warrant identified three categories of\\n\\ninformation for seizure: (1) subscriber and billing information; (2) all incoming\\n\\nand outgoing calls and text messages from the period of May 1, 2012, to June 6,\\n\\n\\n 6\\n\\f2012; and (3) cell-site location information (CSLI). Only the third category—the\\n\\nCSLI data—is at issue in this appeal.\\n\\n B\\n\\n Gilton was indicted by a federal grand jury in 2013, on four counts related to\\n\\nSneed’s murder. In 2015, Gilton filed a motion to suppress the CSLI obtained by\\n\\nthe San Francisco police pursuant to the Sprint warrant. The district court granted\\n\\nGilton’s motion to suppress. The court concluded first that Gilton had a reasonable\\n\\nexpectation of privacy in the historical CSLI he sought to suppress. Accordingly,\\n\\nthe court determined that probable cause was required to obtain that CSLI from\\n\\nSprint. The court concluded, however, that “[t]he affidavit in support of the Sprint\\n\\nwarrant plainly failed to provide a substantial basis for concluding that there was\\n\\nprobable cause to search” because the affidavit “hardly mention[ed] Gilton” and\\n\\ndid not provide a substantial basis for inferring that Gilton was in the San\\n\\nFrancisco area at the time of the shooting. Even assuming that the warrant\\n\\nsupported an inference of a “family-based attack,” the district court concluded that\\n\\nthe “facts pointed to one particular family member being involved: [Barry] Gilton,\\n\\nnot [Antonio] Gilton.” Finally, the district court concluded that the “good faith”\\n\\nexception did not apply because “it was entirely unreasonable to believe that the\\n\\n\\n\\n\\n 7\\n\\faffidavit’s passing, innocuous references to A. Gilton established probable cause to\\n\\nobtain his cell phone data.”\\n\\n The government appealed the district court’s order granting the motion to\\n\\nsuppress. We originally heard argument in March 2017, but then withdrew the\\n\\ncase from submission pending the final decision of the Supreme Court in United\\n\\nStates v. Carpenter, 138 S. Ct. 2206 (2018). Following the Court’s resolution of\\n\\nCarpenter, we requested supplemental briefing and heard reargument in January\\n\\n2019. We agree with the district court that the police lacked probable cause to\\n\\nobtain Gilton’s CSLI data, but we disagree with the district court’s conclusion that\\n\\nthe good faith exception does not apply. Accordingly, we reverse.\\n\\n II. ANALYSIS\\n\\n In Carpenter, the Supreme Court held that “an individual maintains a\\n\\nlegitimate expectation of privacy in the record of his physical movements as\\n\\ncaptured through CSLI.” 138 S. Ct. at 2217. Consequently, Fourth Amendment\\n\\nprotections generally require the government to “obtain a warrant supported by\\n\\nprobable cause before acquiring [CSLI] records.” Id. at 2221.\\n\\n Here, the government did in fact obtain a warrant authorizing the acquisition\\n\\nof Gilton’s CSLI data. Our analysis is thus confined to the questions of whether\\n\\nthat warrant was supported by probable cause, and, if not, whether the search\\n\\n\\n 8\\n\\fshould nevertheless be upheld on the basis of the officers’ good faith reliance on\\n\\nthe warrant.\\n\\n A\\n\\n We turn first to the issue of whether the Sprint warrant was supported by\\n\\nprobable cause. Probable cause exists where the totality of the circumstances\\n\\nindicates a “fair probability that . . . evidence of a crime will be found in a\\n\\nparticular place.” Illinois v. Gates, 462 U.S. 213, 238 (1983). This standard does\\n\\nnot require the affidavit to establish that the evidence is in fact in the place to be\\n\\nsearched, or even that it is more likely than not to be there. United States v.\\n\\nFernandez, 388 F.3d 1199, 1254 (9th Cir. 2004), modified 425 F.3d 1248 (9th Cir.\\n\\n2005). Rather, the issuing judge “need only conclude that it would be reasonable\\n\\nto seek the evidence in the place indicated in the affidavit.” Id. (quotation\\n\\nomitted).\\n\\n In general, the issuing judge’s finding of probable cause is entitled to “great\\n\\ndeference.” United States v. Krupa, 658 F.3d 1174, 1177 (9th Cir. 2011).\\n\\n“[N]otwithstanding the deference that magistrates deserve,” however, we may\\n\\ndetermine that a warrant was invalid where “the magistrate’s probable-cause\\n\\ndetermination reflected an improper analysis of the totality of the circumstances.”\\n\\n\\n\\n\\n 9\\n\\fLeon, 468 U.S. at 915. In addition, the decision of the magistrate “cannot be a\\n\\nmere ratification of the bare conclusions of others.” Gates, 462 U.S. at 239.\\n\\n Here, police sought a warrant to obtain subscriber and billing information,\\n\\nincoming and outgoing calls, incoming and outgoing texts, and CSLI information\\n\\nfor phones belonging to Gilton and a second, unidentified person. The application\\n\\nwas supported by a lengthy affidavit describing the results of interviews, video\\n\\nsurveillance, and an exigent circumstances request to T-Mobile. The affidavit thus\\n\\nprovided the magistrate with a basis for determining the existence of probable\\n\\ncause, see id., and the magistrate’s finding of probable cause is appropriately\\n\\nentitled to deference, see Krupa, 658 F.3d at 1177.\\n\\n Although the affidavit explained in some detail the need for the information\\n\\nfrom the unidentified person’s phone, the affidavit mentions Gilton only three\\n\\ntimes. First, the affidavit records L.G.’s testimony that she “was staying with her\\n\\nelder brother in L.A.” Second, the affidavit indicates that one of the phone\\n\\nnumbers discovered in L.G.’s phone belonged to Gilton. Third, the affidavit states\\n\\nthat L.G. identified which number belonged to Gilton. As the district court pointed\\n\\nout, the affidavit does not indicate that police had reason to believe that Gilton\\n\\n“was . . . in or around San Francisco on or around June 4, 2012.”\\n\\n\\n\\n\\n 10\\n\\f The affidavit’s scant and innocuous references to Gilton do not establish a\\n\\n“fair probability” that evidence of the crime would be found in Gilton’s location\\n\\ndata. Rather, these facts merely indicate that L.G. had the phone number of a\\n\\nfamily member with whom she had lived in Los Angeles. Surely, it is common for\\n\\nan adolescent to store the phone number of the relative with whom she is living,\\n\\nand the mere existence of a familial connection between L.G. and Gilton is not\\n\\nsufficient to render it “reasonable” to search for evidence of the crime in Gilton’s\\n\\nlocation data. See United States v. Grant, 682 F.3d 827, 836–37 (9th Cir. 2012)\\n\\n(stating that an association “through family . . . affiliation” is insufficient to\\n\\nestablish probable cause).\\n\\n The government depends on two related inferences to support the finding of\\n\\nprobable cause. First, the government contends that the totality of the\\n\\ncircumstances supports an inference that “[t]he murder of Calvin Sneed was a\\n\\nfamily solution to a family problem.” Second, because Gilton is related to L.G.\\n\\nand because L.G. met Sneed while living with Gilton, the government asserts that\\n\\nthe magistrate could reasonably infer that Gilton was well aware of Sneed’s\\n\\nrelationship with L.G., that Gilton had reason to be upset with Sneed, and that he\\n\\nhad a motive to commit the crime. The government argues that these two\\n\\ninferences tie Gilton to the murder. We disagree.\\n\\n\\n 11\\n\\f First, we agree with the district court that the affidavit “pointed to one\\n\\nparticular family member being involved in the attack: B. Gilton, not A. Gilton.”\\n\\nAlthough the affidavit indicated that more than one person was likely involved in\\n\\nthe shooting and that the second person might have been a family member, nothing\\n\\nin the affidavit suggested that the identity of the second person was Gilton. Nor do\\n\\nwe agree with the government’s contention that a magistrate could reasonably\\n\\nbelieve that Gilton was the accomplice because of his supposed motive. As the\\n\\ndistrict court pointed out, there is no evidence in the affidavit that Gilton “had\\n\\ncommunicated with family members (or anyone else) about [L.G.’s] relationship\\n\\nwith Sneed,” although that surely was one of the reasons police wanted Gilton’s\\n\\ncall and text records. There is thus no basis in the affidavit to support the inference\\n\\nthat Gilton was upset with Sneed, that he had motive to commit the crime, and,\\n\\nimportantly, that he was in the San Francisco area on the night of the murder. We\\n\\nconclude, as did the district court, that the affidavit’s passing references to Gilton\\n\\nare insufficient to support a reasonable inference that evidence of a crime would be\\n\\nfound in his CSLI data.\\n\\n B\\n\\n Having determined that the warrant was not supported by probable cause,\\n\\nwe turn now to the question of whether the officers nevertheless relied in good\\n\\n\\n 12\\n\\ffaith on the warrant they obtained. The Supreme Court has made quite clear that\\n\\n“[t]he Fourth Amendment contains no provision expressly precluding the use of\\n\\nevidence obtained in violation of its commands.” Leon, 468 U.S. at 906. Because\\n\\n“[t]he wrong condemned by the Amendment is ‘fully accomplished’ by the\\n\\nunlawful search or seizure itself, . . . the exclusionary rule is neither intended nor\\n\\nable to ‘cure the invasion of the defendant’s rights which he has already suffered.’”\\n\\nId. (first quoting United States v. Calandra, 414 U.S. 338, 354 (1974); then\\n\\nquoting Stone v. Powell, 428 U.S. 465, 540 (1976)). As a result, “the use of fruits\\n\\nof a past unlawful search or seizure ‘work[s] no new Fourth Amendment wrong.’”\\n\\nId. (alteration in original) (quoting Calandra, 414 U.S. at 354).\\n\\n The exclusionary rule is thus not “a personal constitutional right of the party\\n\\naggrieved,” but rather “a judicially created remedy designed to safeguard Fourth\\n\\nAmendment rights generally through its deterrent effect.” Calandra, 414 U.S. at\\n\\n348. As such, the question of “[w]hether the exclusionary sanction is appropriately\\n\\nimposed in a particular case” is an entirely separate issue “from the question\\n\\nwhether the Fourth Amendment rights of the party seeking to invoke the rule were\\n\\nviolated by police conduct.” Leon, 468 U.S. at 906 (quoting Gates, 462 U.S. at\\n\\n223). Thus, having determined that the SFPD’s search of Gilton’s CSLI data\\n\\n\\n\\n\\n 13\\n\\fviolated his Fourth Amendment rights, we must address separately the question of\\n\\nwhether exclusion is an appropriate remedy.\\n\\n The question of application of the exclusionary rule is answered by weighing\\n\\n“[t]he substantial social costs exacted by the exclusionary rule” with the benefit of\\n\\nincreased deterrence of police misconduct. Id. at 906–07. “If . . . the exclusionary\\n\\nrule does not result in appreciable deterrence” of police misconduct, “then, clearly,\\n\\nits use in the instant situation is unwarranted.” United States v. Janis, 428 U.S.\\n\\n433, 454 (1976). To have any appreciable deterrent benefit, the exclusion of\\n\\nevidence “must alter the behavior of individual law enforcement officers or the\\n\\npolicies of their departments.” Leon, 468 U.S. at 918.\\n\\n In Leon, the Supreme Court took up the issue of whether the exclusionary\\n\\nrule should be applied in cases where the police seized evidence “in reasonable,\\n\\ngood-faith reliance on a search warrant that is subsequently held to be defective.”\\n\\nId. at 905. The Court reasoned that “[t]he deterrent purpose of the exclusionary\\n\\nrule necessarily assumes that the police have engaged in willful, or at the very least\\n\\nnegligent, conduct which has deprived the defendant of some right,” and that\\n\\n“[w]here the official action was pursued in complete good faith . . . the deterrence\\n\\nrationale loses much of its force.” Id. at 919 (quoting United States v. Peltier, 422\\n\\nU.S. 531, 539 (1975)). In such cases, “excluding the evidence will not further the\\n\\n\\n 14\\n\\fends of the exclusionary rule in any appreciable way.” Id. at 920 (quoting Stone,\\n\\n428 U.S. at 539–40).\\n\\n The Court emphasized that “[t]his is particularly true . . . when an officer\\n\\nacting with objective good faith has obtained a search warrant from a judge or\\n\\nmagistrate and acted within its scope.” Id. Because “[i]t is the magistrate’s\\n\\nresponsibility to determine whether the officer’s allegations establish probable\\n\\ncause[,] . . . [i]n the ordinary case, an officer cannot be expected to question the\\n\\nmagistrate’s probable cause determination.” Id. at 921. As a result, “[i]n most\\n\\nsuch cases, there is no police illegality” and “[p]enalizing the officer for the\\n\\nmagistrate’s error, rather than his own, cannot logically contribute to the deterrence\\n\\nof Fourth Amendment violations.” Id. at 920–21. The Court concluded that “the\\n\\nmarginal or nonexistent benefits produced by suppressing evidence obtained in\\n\\nobjectively reasonable reliance on a subsequently invalidated search warrant\\n\\ncannot justify the substantial costs of exclusion.” Id. at 922.\\n\\n The existence of a search warrant does not automatically preclude\\n\\napplication of the exclusionary rule, however. Rather, the Court instructed that we\\n\\nshould conduct our analysis “on a case-by-case basis,” id. at 918, and recognized\\n\\nfour exceptions to the exception where suppression “remains an appropriate\\n\\nremedy,” id. at 923. First, in cases where “the magistrate or judge . . . was misled\\n\\n\\n 15\\n\\fby information in an affidavit that the affiant knew was false or would have known\\n\\nwas false except for his reckless disregard of the truth.” Id. Second, “in cases\\n\\nwhere the issuing magistrate wholly abandoned his judicial role” by acting as “an\\n\\nadjunct law enforcement officer” or mere “rubber stamp” for the police. Id. at 914,\\n\\n923 (citing Lo-Ji Sales, Inc. v. New York, 442 U.S. 319 (1979)). Third, in cases\\n\\nwhere “an affidavit [is] so lacking in indicia of probable cause as to render official\\n\\nbelief in its existence entirely unreasonable.” Id. at 923 (citation omitted). Finally,\\n\\nin cases where the warrant is “so facially deficient—i.e., in failing to particularize\\n\\nthe place to be searched or the things to be seized—that the executing officers\\n\\ncannot reasonably presume it to be valid.” Id.\\n\\n Here, the parties do not urge and the evidence does not support a conclusion\\n\\nthat the officers wilfully mislead the magistrate, that the magistrate wholly\\n\\nabdicated his role as a neutral and detached judicial officer, or that the warrant was\\n\\nfacially deficient. Thus, the sole question before us is whether Officer Watts’s\\n\\naffidavit was “so lacking in indicia of probable cause” that no reasonable officer\\n\\nwould believe that probable cause existed to search Gilton’s CSLI data. We\\n\\nconclude that the deficiencies in probable cause discussed in subpart II.A., supra,\\n\\nare not so stark as to render official belief in the existence of probable cause\\n\\n“entirely unreasonable.”\\n\\n\\n 16\\n\\f First, we observe that the application for the warrant was issued only two\\n\\ndays after the murder. Second, the affidavit in support of the application was\\n\\nlengthy and laid out for the magistrate what the police knew about the murder and\\n\\nhow they had obtained the information they had, including the results of interviews\\n\\nwith family members and witnesses, video surveillance from a camera on a house\\n\\nnear the murder scene, information received from an anonymous caller who may\\n\\nhave had first-hand facts, and cell phone information obtained on an emergency\\n\\nbasis from the T-Mobile Law Enforcement Center. Third, the police suspected that\\n\\nboth L.G.’s father, Barry Gilton, and a female family member had been involved in\\n\\ninstigating Sneed’s murder. The police also had reason to believe Barry Gilton and\\n\\nat least one other person were in the car when the shooting occurred. Fourth, the\\n\\naffidavit indicated that police had discussed the case with the District Attorney’s\\n\\noffice.\\n\\n For the reasons we have previously explained, all of these facts did not add\\n\\nup to probable cause to obtain Gilton’s CSLI data. The police likely had probable\\n\\ncause to obtain the other information they sought from Gilton’s cellphone—his\\n\\nincoming and outgoing calls and texts—because L.G. had been living with Gilton\\n\\nin Los Angeles when she was dating Sneed, and Gilton may well have been in\\n\\ncontact with L.G.’s family in San Francisco about Sneed. Gilton does not\\n\\n\\n 17\\n\\fchallenge the scope of the warrant with respect to his calls and texts, only his CSLI\\n\\ninformation. What the affidavit is missing is an explanation for why police\\n\\nthought Gilton’s CSLI information was relevant, since there was no\\n\\nsuggestion—other than the cryptic information that multiple family members were\\n\\nlikely involved and that there was a second person in the car, who might or might\\n\\nnot have been a family member—that Gilton was in San Francisco rather than\\n\\nhome in Los Angeles. But we cannot say that no reasonable officer would have\\n\\nrelied on the warrant obtained from the Superior Court just two days into their\\n\\ninvestigation.\\n\\n Gilton points primarily to our holding in Grant to support his contention that\\n\\nthe affidavit was so lacking in indicia of probable cause that no reasonable officer\\n\\nwould depend on the warrant. See 682 F.3d at 836–41. In Grant, the police\\n\\nobtained a warrant to search Grant’s home nine months after the homicide. Id. at\\n\\n828. They had “no evidence suggesting that Grant was involved.” Id. at 832.\\n\\nRather, the police suspected Grant’s two sons. Police had no evidence that one of\\n\\nthe boys had visited his father after the homicide, id. at 833, and only scant\\n\\nevidence that the second son had visited his father some six months after the\\n\\nmurder (and two months prior to the search), id. at 833–34. There was no evidence\\n\\nthat the murder weapon was in Grant’s home. Id. In fact, the police found nothing\\n\\n\\n 18\\n\\fin Grant’s home related to the murder, but charged him with being a felon in\\n\\npossession of other firearms. Id. We concluded that there was no “plausible\\n\\nconnection between Grant’s house and any evidence of the murder” and\\n\\ndetermined that “the magistrate so obviously erred in approving the warrant that\\n\\nthe officers executing it could not have relied on it in good faith.” Id. at 841\\n\\n(quotation marks and citation omitted). Gilton argues that as in Grant, the\\n\\naffidavit’s sparse references to him rendered the magistrate’s error “so obvious[]”\\n\\nthat the officers executing the warrant “could not have relied on it in good faith.”\\n\\nSee id.\\n\\n Grant is distinguishable from this case. Here, Gilton was more closely\\n\\nconnected to the suspected crime than the father in Grant. The affidavit here\\n\\nindicated that Gilton had contact with L.G. up until the day preceding the murder.\\n\\nThe police suspected that the murder was family related, stemming from Sneed’s\\n\\npromoting the prostitution of the under-aged L.G., and coordinated by at least two\\n\\nfamily members. Thus, rather than appearing as a last-ditch, needle-in-the-\\n\\nhaystack effort—as the search in Grant did—the conclusion that Gilton might have\\n\\nbeen involved the events leading to Sneed’s murder was not at all far-fetched. To\\n\\nhold that the police could not have relied in good faith on the magistrate’s\\n\\ndetermination here would be to “[p]enaliz[e] the officer for the magistrate’s error,\\n\\n\\n 19\\n\\frather than his own,” and thus “cannot logically contribute to the deterrence of\\n\\nFourth Amendment violations.” See Leon, 468 U.S. at 921.\\n\\n This is particularly true in light of the fact that in 2012, no circuit court had\\n\\nyet held the Fourth Amendment applicable to CSLI data. See United States v.\\n\\nThompson, 866 F.3d 1149 (10th Cir. 2017), vacated 138 S. Ct. 2706 (2018);\\n\\nUnited States v. Stimler, 864 F.3d 253 (3d Cir. 2017), vacated in part by United\\n\\nStates v. Goldstein, 902 F.3d 411 (3d Cir. 2018); United States v. Graham, 824\\n\\nF.3d 421 (4th Cir. 2016) (en banc), abrogated by Carpenter, 138 S. Ct. 2206\\n\\n(2018); United States v. Carpenter, 819 F.3d 880 (6th Cir. 2016), rev’d 138 S. Ct.\\n\\n2206 (2018); United States v. Davis, 785 F.3d 498 (11th Cir. 2015) (en banc),\\n\\nabrogated by Carpenter, 138 S. Ct. 2206 (2018); In re Application of U.S. for\\n\\nHistorical Cell Site Data, 724 F.3d 600 (5th Cir. 2013), abrogated by Carpenter,\\n\\n138 S. Ct. 2206 (2018). In light of the prevailing belief in 2012 that CSLI data was\\n\\nnot protected by the Fourth Amendment, we conclude that there was no “willful”\\n\\nor “grossly negligent” error here where the officers nevertheless took the\\n\\nprecautionary step of seeking a warrant and provided ample factual background by\\n\\nwhich the magistrate could reach his own determination of the existence of\\n\\nprobable cause. Although we conclude that the magistrate’s determination was\\n\\nerroneous, we hold that the police here were not required to second-guess the\\n\\n\\n 20\\n\\fdetermination of a neutral and detached magistrate. As such, we conclude that\\n\\napplication of the exclusionary rule to Gilton’s CSLI data would have no\\n\\n“appreciable deterrent” effect and is thus unwarranted.3\\n\\n III. CONCLUSION\\n\\n We REVERSE the district court’s order granting Gilton’s motion to\\n\\nsuppress.\\n\\n\\n\\n\\n 3\\n Having concluded that application of the good faith exception is warranted\\nunder Leon, we do not reach the government’s arguments that the good faith\\nexception is also warranted on the basis of Davis v. United States, 564 U.S. 229\\n(2011) (upholding the good faith exception where officers reasonably relied on\\nbinding appellate precedent) or Illinois v. Krull, 480 U.S. 340 (1987) (upholding\\nthe good faith exception where officers reasonably relied on a statute).\\n 21\\n\\f FILED\\nUSA V. ANTONIO GILTON, No. 16-10109 MAR 4 2019\\n MOLLY C. DWYER, CLERK\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1271,"cells":{"id":{"kind":"number","value":4595923,"string":"4,595,923"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Christopher Spreitz v. Charles Ryan"},"decision_date":{"kind":"string","value":"2019-03-04"},"docket_number":{"kind":"string","value":"09-99006"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Richard A. Paez, Marsha S. Berzon, and Richard C. Tallman, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PAEZ, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nCHRISTOPHER J. SPREITZ, No. 09-99006\\n Petitioner-Appellant,\\n D.C. No.\\n v. 4:02-CV-00121-JMR\\n\\nCHARLES L. RYAN, Warden,\\n Respondent-Appellee. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n John M. Roll, District Judge, Presiding\\n\\n Argued July 11, 2013\\n Submitted March 4, 2019\\n San Francisco, California\\n\\n Filed March 4, 2019\\n\\n Before: Richard A. Paez, Marsha S. Berzon,\\n and Richard C. Tallman, Circuit Judges.\\n\\n Opinion by Judge Paez;\\n Dissent by Judge Tallman\\n\\n2 SPREITZ V. RYAN\\n\\n SUMMARY *\\n\\n\\n Habeas Corpus / Death Penalty\\n\\n The panel reversed the district court’s denial of habeas\\ncorpus relief with respect to Christopher J. Spreitz’s death\\nsentence, and remanded, in a case in which Spreitz argued\\nthat the Arizona Supreme Court violated Eddings v.\\nOklahoma, 455 U.S. 104 (1982), by refusing to consider, as\\na matter of law, mitigating evidence of Spreitz’s\\nlongstanding alcohol and substance abuse on the ground that\\nhe did not establish a causal connection between this\\nmitigating evidence and the crime.\\n\\n The panel held that the district court erred in concluding\\nthat Spreitz’s claim that the Arizona Supreme Court violated\\nEddings is procedurally defaulted. The panel explained that\\nthe first opportunity Spreitz had to raise that claim was\\nbefore the post-conviction-relief (PCR) court, at which time\\nhe did so.\\n\\n Because the decision of the PCR court – which first\\ndeclared the claim waived, but proceeded to adjudicate the\\nclaim on the merits – was contrary to clearly established\\nSupreme Court precedent, the panel accorded that decision\\nno deference and reviewed Spreitz’s Eddings claim de novo.\\nThe panel concluded that the Arizona Supreme Court\\nviolated Eddings by impermissibly requiring that Spreitz\\nestablish a causal connection between his longstanding\\nsubstance abuse and the murder before considering and\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n SPREITZ V. RYAN 3\\n\\nweighing the evidence as a nonstatutory mitigating factor.\\nThe panel concluded that the error was not harmless.\\n\\n The panel affirmed the district court’s judgment denying\\nrelief with respect to Spreitz’s conviction in a concurrently\\nfiled memorandum disposition.\\n\\n Dissenting, Judge Tallman wrote that the record does not\\nestablish that either the sentencing court or the Arizona\\nSupreme Court unconstitutionally refused to consider\\nrelevant mitigating evidence; and that even if the Arizona\\ncourts did violate Eddings, Spreitz cannot show that this\\nerror had a “substantial and injurious effect or influence” on\\nhis ultimate sentence.\\n\\n\\n COUNSEL\\n\\nTimothy M. Gabrielsen (argued), Assistant Federal Public\\nDefender; John M. Sands, Federal Public Defender; Office\\nof the Federal Public Defender, Tucson, Arizona; Susan B.\\nFox and Sean Bruner, Law Office of Sean Bruner Ltd.,\\nTucson, Arizona; for Petitioner-Appellant.\\n\\nLacey Stover Gard (argued) and Jeffrey A. Zick, Section\\nChief Counsel; Kent E. Cattani, Chief Counsel, Criminal\\nAppeals/Capital Litigation Section; Mark Brnovich,\\nAttorney General; Office of the Attorney General, Tucson,\\nArizona; for Respondents-Appellees.\\n\\n4 SPREITZ V. RYAN\\n\\n OPINION\\n\\nPAEZ, Circuit Judge:\\n\\n In 1994, an Arizona jury convicted Christopher J. Spreitz\\n(“Spreitz”) of first-degree murder. The victim was thirty-\\nnine year old Ruby Reid (“Reid”). Finding that the cruelty\\nof the murder outweighed any mitigating circumstances, the\\ntrial judge sentenced Spreitz to death. Spreitz appeals the\\ndistrict court’s denial of his petition for a writ of habeas\\ncorpus challenging his conviction and sentence. We affirm\\nthe district court’s judgment with respect to Spreitz’s\\nconviction, 1 and reverse with respect to his sentence. 2\\n\\n In challenging his sentence, Spreitz argues that the\\nArizona Supreme Court unconstitutionally affirmed his\\ndeath sentence by failing to consider mitigating evidence of\\nhis longstanding alcohol and substance abuse. He contends\\nthat the state court refused to consider, as a matter of law,\\nthis evidence in mitigation because he did not establish a\\ncausal connection between the crime and his long-term\\nalcohol and substance abuse. In Eddings v. Oklahoma,\\n455 U.S. 104, 110 (1982), the Supreme Court held that under\\nboth the Eighth and Fourteenth Amendments, a sentencer in\\na capital case may not “refuse to consider, as a matter of law,\\n\\n 1\\n We affirm the judgment with respect to Spreitz’s conviction in a\\nconcurrently filed memorandum disposition.\\n\\n 2\\n On February 13, 2015, we vacated submission of Spreitz’s case\\npending final resolution of the en banc proceedings in McKinney v. Ryan,\\n730 F.3d 903 (9th Cir. 2013). In December 2015, the en banc court\\nissued an opinion in McKinney. 813 F.3d 798 (9th Circ. 2015) (en banc).\\nAfter the Supreme Court denied the State’s petition for a writ of\\ncertiorari, Ryan v. McKinney, 137 S. Ct. 39 (2016) (mem), we ordered\\nsupplemental briefing on the applicability of McKinney to Spreitz’s case.\\n\\n SPREITZ V. RYAN 5\\n\\nany relevant mitigating evidence” offered by the defendant.\\nId. at 114. Although a sentencer “may determine the weight\\nto be given relevant mitigating evidence . . . they may not\\ngive it no weight by excluding such evidence from their\\nconsideration.” Id. at 114–15 (footnote omitted). In\\ninterpreting and applying Eddings, the Supreme Court has\\nexplained that “full consideration of evidence that mitigates\\nagainst the death penalty is essential if the [sentencer] is to\\ngive a reasoned moral response to the defendant’s\\nbackground, character, and crime.” Penry v. Lynaugh\\n(Penry I), 492 U.S. 302, 328 (1989), abrogated on other\\ngrounds by Atkins v. Virginia, 536 U.S. 304 (2002) (internal\\nquotation marks and citation omitted). Moreover, the\\nSupreme Court has been clear: requiring a defendant to\\nprove a causal nexus between his mitigating evidence and\\nthe crime is “a test we never countenanced and now have\\nunequivocally rejected.” Smith v. Texas, 543 U.S. 37, 45\\n(2004) (per curiam).\\n\\n At the time of Spreitz’s sentencing, Arizona Revised\\nStatute Annotated § 13-703(G)(1994) 3 listed five mitigating\\nfactors, and Arizona case law additionally recognized\\nnonstatutory mitigating factors, including, for example, a\\ndefendant’s difficult family background or mental condition\\nnot severe enough to qualify as a statutory mitigating factor.\\nIn an en banc decision of our court, McKinney v. Ryan,\\n813 F.3d 798 (9th Cir. 2015), cert denied, 137 S. Ct. 39\\n(2016) (mem), we explained:\\n\\n For a period of a little over 15 years in capital\\n cases, in clear violation of Eddings, the\\n\\n 3\\n Arizona has since revised its death penalty sentencing scheme. All\\nreferences to Arizona’s Revised Statute Annotated are to those\\nprovisions in effect at the time of Spreitz’s sentencing.\\n\\n6 SPREITZ V. RYAN\\n\\n Supreme Court of Arizona articulated and\\n applied a “causal nexus” test for nonstatutory\\n mitigation that forbade as a matter of law\\n giving weight to mitigating evidence, such as\\n family background or mental condition,\\n unless the background or mental condition\\n was causally connected to the crime.\\n\\nId. at 802. As a result, we held in McKinney that\\n“[a]pplication of the causal nexus test to nonstatutory\\nmitigating factors violated Eddings, for it resulted in Arizona\\ncourts being entirely forbidden, as a matter of state law, to\\ntreat as a mitigating factor a family background or a mental\\ncondition that was not causally connected to a defendant’s\\ncrime.” Id. Spreitz argues that the Arizona Supreme Court\\napplied its causal nexus test in his case, refusing to consider\\nevidence of his long-term substance and alcohol abuse\\nbecause he did not adequately establish a causal connection\\nbetween that history of abuse and his crime.\\n\\n As in McKinney, “the precise question before us is\\nwhether the Arizona Supreme Court applied its\\nunconstitutional causal nexus test in affirming [Spreitz]’s\\ndeath sentence on de novo review.” 4 Id. at 804 (emphasis\\n\\n\\n 4\\n The Arizona Supreme Court is required by statute to undertake an\\nindependent review of a death sentence. Ariz. Rev. Stat. § 13-703.01(A).\\nIn conducting that review for crimes that occurred before 2002, the court\\n“independently review[s] the trial court’s findings of aggravation and\\nmitigation and the propriety of the death sentence. In doing so, [the\\ncourt] review[s] the record de novo, considering the quality and the\\nstrength, not simply the number, of aggravating and mitigating factors.”\\nState v. Lynch, 357 P.3d 119, 141 (Ariz. 2015) (internal quotation marks\\nand citations omitted), rev’d on other grounds, Lynch v. Arizona, 136 S.\\nCt. 1818 (2016).\\n\\n SPREITZ V. RYAN 7\\n\\nand internal quotation marks omitted). For the reasons that\\nfollow, we conclude that it did.\\n\\n I.\\n\\n A. Spreitz’s Crimes, Conviction, and Sentence\\n\\n On May 25, 1989, the police arrested Spreitz after\\ndiscovering Ruby Reid’s body in the desert. Upon\\nquestioning, Spreitz confessed to murdering Reid. We\\nbriefly provide the facts of the murder.\\n\\n On the evening of May 18, after drinking heavily and\\nbeing rejected by the woman he was dating, Spreitz “picked\\nup” Reid at a convenience store. State v. Spreitz (Spreitz I),\\n945 P.2d 1260, 1264–65 (Ariz. 1997). 5 In his confession,\\nSpreitz claimed that Reid voluntarily left with him and that\\nhis understanding was that they would have sex later that\\nevening. Id. at 1265. Spreitz further claimed that he drove\\nher out to the desert, where Reid decided she no longer\\nwanted to have sex. Id. The two fought as a result. Id.\\nSpreitz explained that Reid slapped him and that he\\nresponded by punching her in the mouth. Id. Spreitz then\\nsexually assaulted Reid—“remov[ing] her clothing and\\nha[ving] vaginal intercourse with her.” Id. Spreitz also\\nrecounted that he hit Reid in the head multiple times with a\\nrock “to make her stop yelling.” Id. He explained that he\\nleft Reid without knowing whether she was alive or dead.\\nId.\\n\\n Shortly after leaving Reid in the desert, Spreitz was\\nstopped by a Tucson police department officer. Id. at 1264.\\n\\n 5\\n Our recitation of the facts is adopted from the Arizona Supreme\\nCourt’s opinion affirming Spreitz’s conviction and death sentence.\\n\\n8 SPREITZ V. RYAN\\n\\nThe officer observed that Spreitz had a ripped shirt, smelled\\nof feces, and appeared to be covered in blood and fecal\\nmatter. Id. In addition, when detectives later searched\\nSpreitz’s car, they found blood spatter in the trunk, some of\\nwhich was inconsistent with Spreitz’s blood characteristics.\\nId. at 1265.\\n\\n On Monday morning, May 22, Reid’s naked and\\ndecomposing body was discovered on the outskirts of\\nTucson. Id. At trial, “the medical examiner testified that,\\ndue to the advanced state of decomposition, he could not\\ndetermine the full extent and nature of [Reid]’s injuries.” Id.\\nEven so, he was able to observe “bruising on the legs, arms,\\nand back; bruising and abrasions on the buttocks; several\\nbroken ribs; internal bleeding; a broken jaw; several head\\nlacerations; and a skull fracture where the skull had been\\n‘shoved in.’” Id. The medical examiner concluded that Reid\\nhad been killed by “blunt-force trauma to the head.” Id.\\n\\n In addition to finding Reid’s body “[a]t the scene, police\\ndetectives observed tire tracks leading back to the pavement,\\noil stains in the dirt, footprints, and drag marks in the dirt\\nleading away from the body. They also found feces-stained\\npants, tennis shoes, socks, a used tampon, and a torn\\nbrassiere. Two blood-stained rocks lay next to the body.”\\nId. A few days later, police arrested Spreitz. Id.\\n\\n On June 2, 1989, a grand jury indicted Spreitz for first-\\ndegree murder, Ariz. Rev. Stat. Ann. §§ 13-1105, 13-703;\\nsexual assault, Ariz. Rev. Stat. Ann. §§ 13-406(A) & (B);\\nand kidnapping, Ariz. Rev. Stat. Ann. §§ 13-304(A)(3) &\\n(B). Spreitz I, 945 P.2d at 1265. After five years of pre-trial\\nproceedings mostly regarding the admissibility of DNA\\nevidence, a seven-day jury trial began on August 9, 1994.\\nId. at 1266. After the conclusion of the trial, the jury\\nreturned guilty verdicts on all three counts: first-degree\\n\\n SPREITZ V. RYAN 9\\n\\nmurder (both premeditated and felony murder), sexual\\nassault, and kidnapping. Id.\\n\\n Prior to both his aggravation-mitigation and sentencing\\nhearings before the trial judge, Spreitz submitted evidence\\nand a memorandum in support of certain mitigating\\ncircumstances. As noted earlier, at the time of Spreitz’s\\nsentencing, Arizona’s death penalty statutes provided a list\\nof five specific mitigating factors; Arizona case law\\nrecognized nonstatutory mitigating factors as well. See Ariz.\\nRev. Stat. Ann. § 13-703(G); McKinney, 813 F.3d at 802.\\nSpreitz argued as nonstatutory mitigating factors: “(1) his\\ndysfunctional family life and lack of socialization; (2) a\\nhistory of alcohol and drug abuse; (3) his expressions of\\nremorse; [(4)] his good behavior while incarcerated; [(5)] his\\nlack of adult convictions; [and (6)] no prior record of violent\\ntendencies.” Spreitz I, 945 P.2d at 1279. Spreitz argued as\\nstatutory mitigating factors: (1) his age at the time of the\\nmurder, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and (2) that\\nhis “capacity to appreciate the wrongfulness of his conduct\\nor to conform his conduct to the requirements of law was\\nsignificantly impaired [due to alcohol use], but not so\\nimpaired as to constitute a defense to prosecution,” id.\\n§ 13-703(G)(1); see Spreitz I, 945 P.2d at 1279.\\n\\n Spreitz provided evidence of and argued for all the\\nforegoing mitigating circumstances but focused heavily on a\\ncombination of his relationship with his mother and his long\\nhistory of alcohol and substance abuse. To that end, Spreitz\\nsubmitted a written report by and presented testimony from\\nan examining psychologist, Dr. Todd Flynn, Ph.D. After\\nconducting interviews and research, Dr. Flynn concluded\\nthat Spreitz’s longstanding alcohol and substance abuse\\nshould be considered as both a statutory and nonstatutory\\nmitigating factor. In his report, which was admitted into\\n\\n10 SPREITZ V. RYAN\\n\\nevidence at the aggravation and mitigation hearing, Dr.\\nFlynn repeatedly emphasized Spreitz’s longstanding\\nsubstance abuse 6:\\n\\n By age twelve or thirteen, Chris Spreitz\\n began drinking alcohol and smoking\\n marijuana. By age 15, he drank steadily on\\n weekends and would have a shot of vodka\\n before school.\\n\\n The collateral information shows that the\\n alcohol abuse continued to intensify after he\\n left home. A variety of persons . . . described\\n him as a heavy drinker. This includes a\\n second cousin, Scott [Jouett], who saw him to\\n be intoxicated, “a majority of the time,” when\\n he was visiting Santa Barbara a week before\\n the current offense. To the interviewing\\n investigator, Mr. [Jouett] also described,\\n “several different occasions when Chris has\\n blackouts,” while drinking alcohol.\\n\\n It appears completely clear from the available\\n information that Chris Spreitz had a long-\\n standing problem with alcohol which\\n probably reached the level of physical\\n dependence. He described himself as\\n drinking in the morning as early as age 15.\\n Virtually everyone else who spent much time\\n with him described him as a heavy drinker.\\n\\n\\n\\n 6\\n Dr. Flynn testified consistently with his report. For ease of\\nreference, we refer only to his report.\\n\\n SPREITZ V. RYAN 11\\n\\nUltimately, Dr. Flynn concluded that Spreitz’s alcohol\\nabuse, childhood home life, and stunted development,\\ncombined with rejection by his girlfriend on the day of the\\ncrime, and intoxication at the time of the crime, led to the\\nmurder.\\n\\n At the end of his report, Dr. Flynn summarized his\\nfindings and opinions with respect to both statutory and\\nnonstatutory mitigating factors. He opined that both\\nSpreitz’s age, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and\\nimpaired capacity to appreciate the wrongfulness of his\\nconduct or conform his conduct to the law, Ariz. Rev. Stat.\\nAnn. § 13-703(G)(1), were mitigating statutory factors. As\\nto § 13-703(G)(5), specifically, Dr. Flynn opined that a\\ncombination of a disturbing upbringing in a “pathogenic,\\nemotionally neglectful home environment,” and “[y]ears of\\nalcoholism intoxication” combined to cause major deficits in\\nSpreitz’s social and emotional development. As to\\n§ 13-703(G)(1), Dr. Flynn offered a similar conclusion. In\\nlight of Spreitz’s “history of alcoholism . . . , a significant\\nbut unknown degree of alcohol intoxication is likely” on the\\nnight of the crime. In addition, Spreitz’s “history strongly\\nsuggests years of early experiences likely to have caused a\\nbuild-up of pent-up angry, aggressive feelings toward\\nwomen generally (and older women especially) which may\\nhave burst forth with uncontrollable intensity with or without\\nalcohol intoxication.” Dr. Flynn concluded that Spreitz’s\\nintoxication on the night of the crime coupled with his early\\nchildhood experiences “likely . . . contributed to an\\nuncontrollable outburst of aggression” and inability to\\ncontrol his conduct.\\n\\n Dr. Flynn also concluded that certain nonstatutory\\nmitigating factors were present. The nonstatutory mitigating\\nfactors included, in his opinion: Spreitz’s low potential for\\n\\n12 SPREITZ V. RYAN\\n\\nfuture violence, the “failure of [Spreitz’s] parents to provide\\ntreatment for alcohol abuse in [his] teenage years,” and the\\n“emotionally deprived, physically punitive home\\nenvironment” of Spreitz’s upbringing. Dr. Flynn also\\nemphasized that even if both statutory mitigating factors\\nfailed to satisfy the statutory threshold, they may still “be\\nappropriately considered . . . as nonstatutory mitigation.”\\n\\n On November 28, 1994, the trial court conducted an\\naggravation-mitigation hearing. At the hearing, Spreitz\\ncalled three mitigation witnesses: Dr. Flynn and two\\ncorrectional officers from the Pima County jail, where\\nSpreitz was incarcerated. The State did not offer any\\nwitnesses or evidence at the hearing. Dr. Flynn testified\\nconsistently with his report, as detailed above.\\n\\n On the same date, a probation officer filed a Pre-\\nSentence Report, which concluded:\\n\\n It appears the defendant became involved in\\n the senseless commission of the instant\\n offense due to his alcohol and drug abuse.\\n After five years in custody, he now admits his\\n substance abuse problem; however, this does\\n not condone his involvement in the offense.\\n It is unfortunate the victim died before the\\n defendant had his revelation.\\n\\nPrior to the sentencing hearing, Spreitz also submitted\\nseveral letters from friends, family, and jail personnel.\\n\\n On December 21, 1994, the trial court conducted a\\nsentencing hearing. Spreitz, Spreitz’s counsel, the victim’s\\n\\n SPREITZ V. RYAN 13\\n\\nsister, and the prosecutor each addressed the court. 7 The\\nprosecutor first disputed the mitigating evidence presented\\nby Spreitz at the mitigation-aggravation hearing. With\\nrespect to Spreitz’s alcohol abuse, the prosecutor argued that\\nSpreitz’s intoxication at the time of the crime did not meet\\nthe statutory definition for mitigation under § 13-703(G)(1)\\nbecause his conduct both during the murder and afterwards\\ndemonstrated that “he knew what he was doing.” The\\nprosecutor emphasized that the evidence revealed that Reid\\nhad been forcibly abducted and stuffed into the trunk of a\\ncar, thus forcing her to spend time “contemplat[ing] the\\nuncertainty of her fate.” The prosecutor argued that the\\nevidence revealed signs of struggle and that the serious\\ninjuries to Reid’s body belied any notion that Spreitz did not\\nknow what he was doing. The prosecutor further argued that\\nthe fact that Reid defecated on herself revealed that she was\\nterrified prior to her murder. In sum, the prosecutor argued\\nthat given both the heinous nature of the crime and the\\nmanner in which Reid suffered, a finding in aggravation that\\nReid was murdered in an “especially cruel manner” was\\nwarranted.\\n\\n After a short recess, the sentencing judge rendered oral\\nfindings addressing the aggravation and mitigation issues.\\nHe then imposed a sentence of death. 8 Following the\\n\\n\\n\\n 7\\n The trial court did not entertain argument at the conclusion of the\\naggravation-mitigation hearing. As a result, counsel addressed the\\naggravation-mitigation evidence and sentence options at the subsequent\\nsentencing hearing.\\n 8\\n Ring v. Arizona, 536 U.S. 548 (2012), the Supreme Court decision\\nholding judge-sentencing in capital cases unconstitutional, had yet to be\\ndecided.\\n\\n14 SPREITZ V. RYAN\\n\\nhearing, the sentencing judge filed a judgment setting forth\\nwritten findings.\\n\\n The judge first found one aggravating circumstance—\\nthat the offense was committed in an “especially cruel\\nmanner.” Ariz. Rev. Stat. Ann. § 13-703(F)(6). The judge\\nproceeded to review the “many factors [that were] submitted\\nin mitigation,” ultimately finding five mitigating\\ncircumstances of various levels of significance:\\n\\n [1] A mitigating circumstance defense\\n submitted was that the defendant had an\\n extremely disruptive childhood . . . . The\\n court finds the home was sub-normal, not\\n even a minimally healthy one for developing\\n children; and it is obvious the defendant\\n suffered a disruptive middle childhood—had\\n a punitive, controlling, cold mother, who he\\n could not please, no matter what he did.\\n\\n The defendant in his life turned to substance\\n abuse—alcohol and some suggestion he was\\n using cocaine and other drugs. However, the\\n court does not find such is a mitigating\\n circumstance that impaired his ability to\\n make a judgment on whether he was acting\\n rightfully or wrongfully in the death of the\\n victim.\\n\\n The defendant’s history of intoxication is\\n longstanding. He had been abusing\\n substances for close to ten years of his life at\\n the time of this offense when he was twenty-\\n two. Again, the court does not believe that\\n the substance abuse or intoxication impaired\\n\\n SPREITZ V. RYAN 15\\n\\n the defendant’s ability and capacity to\\n appreciate the wrongfulness of his conduct to\\n any significant degree. . . . The court does not\\n believe intoxication is any sort of mitigating\\n circumstance.\\n\\n ...\\n\\n [2] The court acknowledges that the\\n defendant has begun a process of\\n improvement and emotional growth while\\n confined at the Pima County Jail, where he\\n has taken part in education programs.\\n Correctional officers have testified he was a\\n prisoner who caused no problems.\\n\\n [3] The age of the defendant at the time of the\\n offense (twenty-two) is not a mitigating\\n circumstance in and of itself. Immaturity\\n probably is, but the court does not believe\\n immaturity was a significant mitigating\\n circumstance.\\n\\n [4] The court finds that Mr. Spreitz has no\\n criminal history of a felony nature—there is\\n no history or propensity for acts of violence.\\n\\n [5] The court believes the defendant is\\n capable of being rehabilitated. The court\\n does not know whether he has a good\\n prognosis for the future, but the court\\n believes he can be rehabilitated.\\n\\nIn conclusion, the judge found that “the mitigating\\ncircumstances [were] not sufficient to balance the\\n\\n16 SPREITZ V. RYAN\\n\\naggravating circumstances, nor [were] they sufficiently\\nsubstantial to call for leniency.” He thus imposed a death\\nsentence as to the first-degree murder conviction.\\n\\n B. Spreitz’s Appeal and Post-Conviction Proceedings\\n\\n Spreitz appealed his conviction and sentence to the\\nArizona Supreme Court with the assistance of new counsel.\\nThe Arizona Supreme Court affirmed Spreitz’s conviction\\nand sentence after conducting its own independent review of\\nthe record. 9 In reviewing the mitigating evidence, the court\\nfound four of the five mitigating circumstances that the\\nsentencing judge had found, declining to find Spreitz’s good\\nbehavior while in jail awaiting sentencing as a mitigating\\ncircumstance. Spreitz I, 945 P.2d at 1280–81. The court\\nexplained:\\n\\n [1] “We agree with the sentencing judge that\\n defendant’s upbringing was subnormal.” Id.\\n at 1280.\\n\\n [2] “We also find that the sentencing judge\\n . . . properly found that his emotional\\n immaturity was not a significant mitigating\\n factor.” Id. at 1281.\\n\\n [3] “We agree that the record supports the\\n sentencing judge’s findings that defendant\\n had no previous adult felony convictions, no\\n prior acts of violence, and\\n\\n\\n 9\\n See note 4, supra. Because the Arizona Supreme Court conducted\\na de novo review, we focus on its analysis, rather than, as the dissent\\ndoes in part, the analysis by the trial court.\\n\\n SPREITZ V. RYAN 17\\n\\n [4] that the defendant is capable of\\n rehabilitation.” Id.\\n\\nThe Arizona Supreme Court also added that it considered\\nremorse as an additional nonstatutory mitigating factor. Id.\\n\\n The court addressed Spreitz’s history of alcohol and\\nsubstance abuse as follows:\\n\\n The record demonstrates defendant’s\\n longtime substance abuse problems. We\\n note, however, that defendant’s general\\n problems with substance abuse are not\\n essential to our decision here. We therefore\\n decline to conclude that defendant was\\n impaired by alcohol consumption to an extent\\n that it interfered with his “capacity to\\n appreciate the wrongfulness of his conduct or\\n to conform his conduct to the requirements of\\n the law.” A[riz]. R[ev]. S[tat]. [Ann.]\\n § 13-703(G)(1); see also State v. Medrano,\\n 185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)\\n (citing [State v.] Stokley, 182 Ariz. [505,]\\n 520, 898 P.2d [454,] 469 [1995]).\\n\\nId. at 1280–81. In the only other portion of its opinion\\naddressing Spreitz’s history of alcohol and substance abuse,\\nthe court said:\\n\\n The sentencing judge found that defendant’s\\n ability to appreciate the wrongfulness of his\\n conduct was not impaired on the night of the\\n murder to any significant extent by substance\\n abuse, emotional disorders, situational stress,\\n or by a combination of these. Our review of\\n\\n18 SPREITZ V. RYAN\\n\\n the record convinces us that the trial court’s\\n finding was proper.\\n\\nId. at 1281.\\n\\n The Arizona Supreme Court agreed with the trial court’s\\naggravation finding that the murder was done in an\\nespecially cruel manner, id. at 1279, and then proceeded to\\nreweigh the applicable aggravating and mitigating factors,\\nfinding that “the aggravating circumstance of especial\\ncruelty in defendant’s murder of Ruby Reid outweigh[ed] all\\nfactors mitigating in favor of leniency.” Id. at 1282. It\\naffirmed his death sentence. Id. at 1283. Spreitz filed a\\npetition for a writ of certiorari, which the United States\\nSupreme Court denied. Spreitz v. Arizona, 523 U.S. 1027\\n(1998) (mem).\\n\\n In March 2000, represented again by new counsel,\\nSpreitz filed a petition for post-conviction relief in the\\nArizona Superior Court (“PCR court”). See Ariz. R. Crim.\\nP. 32.1. Spreitz alleged, inter alia, that the sentencing judge\\nand the Arizona Supreme Court committed error by failing\\nto consider his history of alcohol and substance abuse as a\\nnonstatutory mitigating factor apart from its causal\\nconnection to the murder—i.e., Eddings error. He also\\nalleged that his appellate counsel was ineffective for failing\\nto raise the sentencing judge’s Eddings error on appeal to the\\nArizona Supreme Court. Spreitz additionally raised several\\nnew claims of trial-counsel-ineffectiveness.\\n\\n The PCR court denied all of Spreitz’s claims and entered\\nan order dismissing his petition for post-conviction relief. In\\nPart 3 of its order, the PCR court discussed Spreitz’s claims\\nof “nexus-error”—that both the trial court and the Arizona\\nSupreme Court erred when they failed to consider Spreitz’s\\nlongstanding alcohol and substance abuse on the basis that\\n\\n SPREITZ V. RYAN 19\\n\\nSpreitz had failed to establish a causal nexus between the\\nlong-term substance abuse and the murder. 10 The PCR court\\ndismissed the claim 11 as waived, reasoning that Spreitz had\\nfailed to raise the issue on direct appeal. Nonetheless, the\\nPCR court addressed the merits of the nexus-error claim in\\nthe course of analyzing Spreitz’s argument that his appellate\\ncounsel was ineffective for failing to raise it. In doing so,\\nthe PCR court explained that the claim failed because:\\n\\n [I]t must be demonstrated, under A[riz].\\n R[ev]. S[tat]. [Ann.] § 13-703(G)(l), that\\n there is a causal link between the history of\\n alcohol or substance abuse and the offense\\n itself. E.g., State v. Stokley, [] 182 Ariz.\\n [505,] 523 [898 P.2d 454, 472 (Ariz. 1995)].\\n Without some basis for explaining or\\n defining the individual’s behavior at the time\\n of the offense, the Petitioner’s history of\\n alcohol or substance abuse would be\\n inconsequential (which is exactly what the\\n trial court and Supreme Court concluded).\\n State v. Kayer, 194 Ariz. 423, 984 P.2d 81\\n ([Ariz.] 1999).\\n\\n At times, the court can and should consider\\n an individual’s long-term alcoholism and\\n\\n 10\\n For ease of reference, we refer to the Arizona courts’ alleged error\\nas “nexus-error.”\\n\\n 11\\n Although Spreitz raised, and the PCR court recognized, two\\ndistinct claims—one with respect to the sentencing court and one with\\nrespect to the Arizona Supreme Court—the PCR court’s analysis treats\\nthe claims as one and the same and refers to them in the singular. To\\nreflect accurately the PCR court’s discussion, our summary here likewise\\nrefers to a singular “claim.”\\n\\n20 SPREITZ V. RYAN\\n\\n substance abuse, usually in conjunction with\\n other factors or diagnosis, as non-statutory\\n mitigation. However, the impact or effect of\\n the alcoholism or substance abuse must be\\n substantial and of such severity that it\\n provides a sufficient basis for explaining the\\n defendant’s conduct, character or ability to\\n control his behavior at the time of the offense.\\n [Citations omitted].\\n\\n As previously discussed, there is no evidence\\n in Petitioner’s case to suggest that he suffered\\n any long-term effects from his alcohol or\\n drug abuse that precluded him from\\n controlling his behavior. Petitioner did not\\n suffer from any cognitive or emotional\\n deficits that rendered him incapable of\\n controlling his conduct. Therefore, the trial\\n court did not err in failing to find Petitioner’s\\n history of alcohol or substance abuse as a\\n separate, non-statutory mitigating factor.\\n [Citation omitted].\\n\\n Spreitz filed a petition for review in the Arizona\\nSupreme Court challenging the PCR court’s judgment. The\\nArizona Supreme Court summarily affirmed the PCR court’s\\nmerits determination with respect to Spreitz’s Eddings\\nclaims. State v. Spreitz (Spreitz II), 39 P.3d 525, 527 (Ariz.\\n2002).\\n\\n Spreitz filed his federal habeas petition in February\\n2003. He alleged in claim seven that the sentencing judge\\nand the Arizona Supreme Court had both committed nexus-\\nerror with respect to his long-time alcohol and substance\\nabuse, and that appellate counsel was ineffective for failing\\n\\n SPREITZ V. RYAN 21\\n\\nto argue the sentencing judge’s nexus-error on direct appeal\\nto the Arizona Supreme Court.\\n\\n The district court denied all of Spreitz’s claims. With\\nrespect to the alleged nexus-error by both the sentencing\\ncourt and the Arizona Supreme Court, the district court\\nfound the claims procedurally barred, relying on the PCR\\ncourt’s determination that Spreitz had waived the claims\\nbecause he could have raised them on direct appeal but failed\\nto do so. The district court concluded that Spreitz had\\nproperly exhausted his ineffective assistance of appellate\\ncounsel claim, and thus addressed the alleged nexus-error in\\nthat context, ultimately concluding it was not meritorious.\\nAccordingly, the district court concluded that “appellate\\ncounsel’s failure to raise this issue on appeal does not\\nconstitute ineffectiveness.”\\n\\n Spreitz timely appealed.\\n\\n II. Jurisdiction and Standard of Review\\n\\n We have jurisdiction pursuant to 28 U.S.C. §§ 1291\\nand 2253(a), and review de novo the district court’s denial\\nof a writ of habeas corpus. Poyson v. Ryan, 879 F.3d 875,\\n887 (9th Cir. 2018). Because Spreitz filed his federal habeas\\npetition after April 24, 1996, he must satisfy the\\nAntiterrorism and Effective Death Penalty Act (“AEDPA”).\\nUnder AEDPA, we may not grant habeas relief unless the\\nstate’s adjudication of Spreitz’s claim (1) “was contrary to\\n. . . clearly established federal law[] as determined by the\\nSupreme Court,” (2) “involved an unreasonable application\\nof” such law, or (3) “was based on an unreasonable\\ndetermination of the facts in light of the evidence presented\\nin the State court proceeding.” (28 U.S.C. § 2254(d)). “In\\nmaking this determination, we look to the last state court\\ndecision to address the claim,” White v. Ryan, 895 F.3d 641,\\n\\n22 SPREITZ V. RYAN\\n\\n665 (9th Cir. 2018) (citing Wilson v. Sellers, 138 S. Ct. 1188,\\n1192 (2018)), which for Spreitz’s nexus-error claim is that\\nof the PCR court.\\n\\n Spreitz argues that the PCR court’s decision was\\n“contrary to” Eddings v. Oklahoma, 455 U.S. 104 (1982).\\n“A state court’s decision is contrary to clearly established\\nfederal law if it ‘applies a rule that contradicts the governing\\nlaw set forth in [U.S. Supreme Court] cases’ or arrives at a\\ndifferent result in a case that ‘confronts a set of facts that are\\nmaterially indistinguishable from a decision of [the\\nSupreme] Court.’” Castellanos v. Small, 766 F.3d 1137,\\n1146 (9th Cir. 2014) (alteration in original) (quoting\\nWilliams v. Taylor, 529 U.S. 362, 405–06 (2000)). “If the\\nstate court applies a legal standard that contradicts clearly\\nestablished [Supreme Court] law, we review de novo the\\napplicant’s claims, applying the correct legal standard to\\ndetermine whether the applicant is entitled to relief.” Id.\\n(citation omitted).\\n\\n III. Discussion\\n\\n As discussed supra, the precise question we must decide\\nis whether the Arizona Supreme Court applied its\\nunconstitutional causal nexus test in violation of Eddings\\nwhen it affirmed Spreitz’s death sentence. To answer that\\nquestion, we must first determine whether the claim is\\nproperly before us. After concluding that Spreitz’s claim\\ncould not have been procedurally defaulted, we turn to the\\nlevel of deference we must accord the PCR court’s ruling on\\nSpreitz’s Eddings claim under AEDPA. Because the PCR\\ncourt’s decision was contrary to clearly established Supreme\\nCourt precedent, we accord that decision no deference and\\nreview Spreitz’s Eddings claim de novo. We conclude that\\nthe Arizona Supreme Court violated Eddings by\\nimpermissibly requiring that Spreitz establish a causal\\n\\n SPREITZ V. RYAN 23\\n\\nconnection between his longstanding substance abuse and\\nthe murder before considering and weighing the evidence as\\na nonstatutory mitigating factor. Finally, having determined\\nthat the Arizona Supreme Court applied an unconstitutional\\ncausal nexus test in its sentencing procedure, we turn to\\nwhether the error was harmless under Brecht v. Abrahamson,\\n507 U.S. 619 (1993). We conclude that the error was not\\nharmless and therefore reverse the district court’s judgment\\nwith respect to Spreitz’s sentence.\\n\\n A.\\n\\n Under the doctrine of procedural default, a federal\\nhabeas court “will not review the merits of claims, including\\nconstitutional claims, that a state court declined to hear\\nbecause the prisoner failed to abide by a state procedural\\nrule.” Martinez v. Ryan, 566 U.S. 1, 9 (2012). “We review\\nde novo a district court’s conclusion that a claim is\\nprocedurally defaulted.” Cooper v. Neven, 641 F.3d 322,\\n326 (9th Cir. 2011).\\n\\n Here, the district court concluded that both of Spreitz’s\\nEddings claims—that the sentencing court and the Arizona\\nSupreme Court applied an unconstitutional causal nexus\\ntest—were procedurally defaulted, because the PCR court\\nfound each claim waived for failure to raise them on direct\\nappeal. This conclusion, however, is erroneous with respect\\nto Spreitz’s claim that the Arizona Supreme Court violated\\nEddings. After all, “[t]he Arizona Supreme Court reviews\\ncapital sentences de novo, making its own determination of\\nwhat constitute[s] legally relevant aggravating and\\nmitigating factors, and then weighing those factors\\nindependently. Ariz. Rev. Stat. Ann. § 13-755. The Arizona\\nSupreme Court ‘conducts a thorough and independent\\nreview of the record and of the aggravating and mitigating\\nevidence to determine whether the sentence is justified.’”\\n\\n24 SPREITZ V. RYAN\\n\\nMcKinney, 813 F.3d at 819 (emphasis removed) (quoting\\nState v. McKinney, 917 P.2d 1214, 1225 (Ariz. 1996)).\\nSpreitz could not have raised on direct appeal his claim that\\nthe Arizona Supreme Court violated Eddings in performing\\nits de novo review of Spreitz’s death sentence. The first\\nopportunity he had to raise that claim was before the PCR\\ncourt, at which time he did so. Thus, Spreitz’s claim is not\\nprocedurally defaulted and is properly before us. 12\\n\\n B.\\n\\n Because Spreitz’s claim that the Arizona Supreme Court\\napplied an unconstitutional causal nexus test to his\\nnonstatutory mitigating evidence of long-time alcohol and\\n 12\\n The district court declined to issue a certificate of appealability\\n(“COA”) on Spreitz’s substantive Eddings claim, which he raised as\\nclaim 1.7 on pages 108 through 111 in his federal habeas petition.\\nAlthough Spreitz did not present his Eddings claim as a separate\\nuncertified issue, we may grant a COA regarding any uncertified issue\\ndiscussed in a petitioner’s opening brief. See 9th Cir. R. 22-1(e).\\nSpreitz’s opening brief discusses the Arizona courts’ Eddings errors. We\\ntherefore may exercise our discretion to expand the COA to encompass\\nthe Eddings claim.\\n\\n To obtain a COA under 28 U.S.C. § 2253(c), Spreitz must\\ndemonstrate that “reasonable jurists could debate whether . . . the\\npetition should have been resolved in a different manner or that the issues\\npresented were adequate to deserve encouragement to proceed further.”\\nSlack v. McDaniel, 529 U.S. 473, 484 (2000). Reasonable jurists could\\ndebate the merits of Spreitz’s Eddings claim. We thus exercise our\\ndiscretion and expand the COA to encompass Spreitz’s substantive\\nEddings claim with respect to the Arizona Supreme Court. We\\nacknowledge that under Ninth Circuit Rule 22-1(f), the State would\\nnormally be afforded an opportunity to submit supplemental briefing on\\nthe uncertified Eddings issue. Here, however, the Eddings issue has been\\nexhaustively briefed, in the context of Spreitz’s appellate counsel’s\\nalleged ineffectiveness before the Arizona Supreme Court. Additional\\nbriefing is unwarranted.\\n\\n SPREITZ V. RYAN 25\\n\\nsubstance abuse is properly before us, we next determine the\\nlevel of deference to accord the PCR court’s adjudication of\\nthat claim. As discussed above, although the PCR court first\\ndeclared the claim waived, it proceeded to adjudicate the\\nclaim on the merits. In denying the claim, the PCR court\\nexplained:\\n\\n Without some basis for explaining or\\n defining the individual’s behavior at the time\\n of the offense, [Spreitz]’s history of alcohol\\n or substance abuse would be inconsequential\\n (which is exactly what the trial court and\\n Supreme Court concluded).\\n\\n At times, the court can and should consider\\n an individual’s long-term alcoholism and\\n substance abuse, usually in conjunction with\\n other factors or diagnosis, as non-statutory\\n mitigation. However, the impact or effect of\\n the alcoholism or substance abuse must be\\n substantial and of such severity that it\\n provides a sufficient basis for explaining the\\n defendant’s conduct, character, or ability to\\n control his behavior at the time of the offense.\\n\\n ....\\n\\n As previously discussed, there is no evidence\\n in [Spreitz]’s case to suggest that he suffered\\n any long-term effects from his alcohol or\\n drug abuse that precluded him, from\\n controlling his behavior. Petitioner did not\\n suffer from any cognitive or emotional\\n deficits that rendered him incapable of\\n controlling his conduct. Therefore, the trial\\n\\n26 SPREITZ V. RYAN\\n\\n court did not err in failing to find [Spreitz]’s\\n history of alcohol or substance abuse as a\\n separate, non-statutory mitigating factor.\\n\\n In other words, the PCR court concluded that Spreitz’s\\nlongstanding alcohol and substance abuse could only be\\nconsidered a nonstatutory mitigating factor if it “provide[d]\\na sufficient basis for explaining the defendant’s conduct,\\ncharacter, or ability to control his behavior at the time of the\\noffense.” This discussion both accurately described the\\ncausal nexus test and approved of it.\\n\\n Such reasoning is contrary to clearly established federal\\nlaw. As explained in McKinney, “the causal nexus test\\nclearly violates Eddings.” 813 F.3d at 810. Therefore,\\n“[b]ecause the state court used the wrong standard, we need\\nnot defer to that decision.” Hardy v. Chappell, 849 F.3d 803,\\n820 (9th Cir. 2016).\\n\\n C.\\n\\n Because the PCR court’s decision was “contrary to”\\nclearly established law, we review de novo the merits of\\nSpreitz’s Eddings claim. 28 U.S.C. § 2254(d)(1). We\\ntherefore consider whether the Arizona Supreme Court\\napplied its unconstitutional causal nexus test in reviewing\\nand affirming Spreitz’s death sentence. In so doing, “we\\nlook only to the decision of th[e Arizona Supreme Court],\\n. . . only [considering] . . . the decision of the sentencing\\njudge . . . to the degree it was adopted or substantially\\nincorporated by the Arizona Supreme Court.” McKinney,\\n813 F.3d at 819.\\n\\n SPREITZ V. RYAN 27\\n\\n 1.\\n\\n Our decision in McKinney frames our consideration of\\nthe Arizona Supreme Court’s decision and Spreitz’s claim.\\nMcKinney makes clear that the Arizona Supreme Court\\nconsistently articulated and applied its unconstitutional\\ncausal nexus test during the period in which Spreitz’s death\\nsentence was litigated in the trial court and reviewed by the\\nArizona Supreme Court. Id. at 824. Moreover, McKinney\\nmakes plain that although on habeas review, we generally\\napply a “presumption that state courts know and follow the\\nlaw,” “the Arizona Supreme Court’s consistent articulation\\nand application of its causal nexus test . . . make such a\\ncourse impossible.” Id. at 803 (internal quotation marks and\\ncitation omitted). In appeals heard from the late 1980s until\\nat least 2002, see State v. Canez, 42 P.3d 564, 594 (Ariz.\\n2002), it was unmistakably clear that “the Arizona Supreme\\nCourt did not know and follow federal law.” Id. (internal\\nquotation marks omitted). 13 Therefore, just as in McKinney\\nand subsequent cases applying McKinney, see, e.g., Poyson,\\n879 F.3d at 889, “the presumption” that the Arizona\\nSupreme Court knew and followed the law “is rebutted” here\\nas well. McKinney, 813 F.3d at 804.\\n\\n That said, McKinney does not dispose of Spreitz’s claim.\\nWe must still review the Arizona Supreme Court’s decision\\nto determine whether it did, in fact, apply its unconstitutional\\ncausal nexus test in Spreitz’s case. In doing so, we are\\nmindful of the fact that at the time the Arizona Supreme\\nCourt decided Spreitz’s appeal, “if there is to be a\\npresumption, it is that the Arizona Supreme Court violated\\n\\n\\n 13\\n In 2005, the Arizona Supreme Court repudiated its earlier nexus\\ntest. See State v. Anderson, 111 P.3d 369, 391–92 (Ariz. 2005).\\n\\n28 SPREITZ V. RYAN\\n\\nthe dictates of Lockett and Eddings.” Greenway v. Ryan,\\n866 F.3d 1094, 1095 (9th Cir. 2017).\\n\\n 2.\\n\\n Against this backdrop, we are convinced that the Arizona\\nSupreme Court applied its causal nexus test with respect to\\nSpreitz’s evidence of longstanding alcohol and substance\\nabuse in violation of Eddings.\\n\\n We quote again the Arizona Supreme Court’s discussion\\nof Spreitz’s history of alcohol and substance abuse:\\n\\n The record demonstrates defendant’s\\n longtime substance abuse problems. We\\n note, however, that defendant’s general\\n problems with substance abuse are not\\n essential to our decision here. We therefore\\n decline to conclude that defendant was\\n impaired by alcohol consumption to an extent\\n that it interfered with his “capacity to\\n appreciate the wrongfulness of his conduct or\\n to conform his conduct to the requirements of\\n the law.” A[riz]. R[ev]. S[tat]. [Ann.] § 13-\\n 703(G)(1); see also State v. Medrano,\\n 185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)\\n (citing Stokley, 182 Ariz. at 520, 898 P.2d at\\n 469).\\n\\n ....\\n\\n The sentencing judge found that defendant’s\\n ability to appreciate the wrongfulness of his\\n conduct was not impaired on the night of the\\n murder to any significant extent by substance\\n abuse, emotional disorders, situational stress,\\n\\n SPREITZ V. RYAN 29\\n\\n or by a combination of these. Our review of\\n the record convinces us that the trial court’s\\n finding was proper.\\n\\nSpreitz I, 945 P.2d at 1280–81. This discussion\\ndemonstrates that the Arizona Supreme Court primarily\\nconcluded that Spreitz failed to show statutory mitigation\\nunder Ariz. Rev. Stat. Ann. § 13-703(G)(1) based on his\\nintoxication on the night of the murder. “When applied\\nsolely in the context of statutory mitigation under\\n§ 13-703(G)(1), the causal nexus test does not violate\\nEddings.” McKinney, 813 F.3d at 810.\\n\\n The plain text of the Arizona Supreme Court’s decision,\\nhowever, also clearly demonstrates that the court did not\\nsolely apply its causal nexus test to Spreitz’s evidence of\\nstatutory mitigation. The court initially recognized Spreitz’s\\n“general problems with substance abuse”—evidence that\\nshould have been relevant as nonstatutory mitigation—but\\nconcluded those problems were “not essential to [its]\\ndecision” because they did not show Spreitz was “impaired\\non the night of the murder.” Spreitz I, 945 P.2d at 1280–81.\\nIn other words, the court did not acknowledge the relevance\\nof longtime substance abuse as nonstatutory mitigation in the\\nabsence of a causal relationship to the crime. Instead, the\\ncourt held that because there was no causal relationship,\\nSpreitz’s long-term alcohol abuse was not of significance—\\ni.e. “not essential”—to the court’s overall determination of\\neither statutory or nonstatutory mitigation.\\n\\n The Supreme Court, however, has been clear that “‘full\\nconsideration of evidence that mitigates against the death\\npenalty is essential if the sentencer is to give a reasoned\\nmoral response to the defendant’s background, character,\\nand crime.’” McKinney, 813 F.3d at 812 (first emphasis\\n\\n30 SPREITZ V. RYAN\\n\\nadded) (quoting Penry I, 492 U.S. at 319, 328).\\nAccordingly, “Eddings requires that ‘[t]he sentencer must\\nalso be able to consider and give effect’” to any relevant\\nmitigation evidence proffered by the defendant, which\\nincludes evidence going toward nonstatutory mitigation. Id.\\n(quoting Penry I, 492 U.S. at 319, 328). The Arizona\\nSupreme Court’s refusal to consider Spreitz’s alcohol and\\nsubstance abuse beyond its connection, if any, to Reid’s\\nmurder constitutes application of its unconstitutional causal\\nnexus test to relevant nonstatutory mitigation evidence.\\n\\n Our understanding of the Arizona Supreme Court’s\\nopinion is bolstered by the fact that in its discussion of\\nSpreitz’s longstanding alcohol and substance abuse, the\\ncourt cited to a prior decision that likewise applied the\\nunconstitutional causal nexus test, State v. Medrano,\\n914 P.2d 255 (Ariz. 1996). In McKinney, we explained that\\nin Medrano, the Arizona Supreme Court upheld the\\nsentencing judge’s determination that the defendant’s\\ncocaine use failed as a nonstatutory mitigating circumstance\\nbecause he did not show that it contributed to his conduct on\\nthe night of the murder. McKinney, 813 F.3d at 825–26. In\\nso doing, “[t]he Arizona Supreme Court applied the causal\\nnexus test.” Id. at 825. Although the Arizona Supreme\\nCourt’s citation to Medrano is not dispositive, it certainly\\ncorroborates our understanding of the court’s opinion. See\\nalso Poyson, 879 F.3d at 890 (emphasizing, among other\\nconsiderations, that the Arizona Supreme Court cited a\\npassage from a case specifically identified in McKinney as\\napplying the unconstitutional causal nexus test in order to\\nconclude that the state court also applied that test to Poyson’s\\nevidence). Moreover, as we explained in McKinney, “the\\nArizona Supreme Court applied its unconstitutional nexus\\ntest consistently” for fifteen years because it had “a mistaken\\nunderstanding of Eddings.” 813 F.3d at 826. It is thus\\n\\n SPREITZ V. RYAN 31\\n\\nentirely logical that in citing the exact portion of Medrano\\nwhere it previously applied its causal nexus test, the Arizona\\nSupreme Court was applying the same test in Spreitz’s case.\\nSee id. at 821; see also Poyson, 879 F.3d at 890.\\n\\n We find further support for our understanding of the\\nArizona Supreme Court’s unconstitutional treatment of\\nSpreitz’s alcohol and substance abuse when we compare it\\nwith the manner in which the court discussed other\\nmitigating factors. As detailed earlier, in evaluating the\\nevidence of Spreitz’s general history of alcohol and\\nsubstance abuse, the court stated that this evidence was “not\\nessential” to its decision. Spreitz I, 945 P.2d at 1280. By\\ncontrast, when the court considered other mitigating factors,\\nit specifically discussed the weight it would give those\\nfactors, as opposed to simply dismissing them outright. For\\nexample, when discussing Spreitz’s subnormal upbringing,\\nthe court stated, “Although we recognize defendant’s\\nupbringing as a mitigating circumstance, we accord it little\\nweight.” Id. at 1280 (emphasis added). As another example,\\nafter “finding that since his arrest, [Spreitz] has\\ndemonstrated remorse,” the court stated that his “remorse for\\nhis actions does little to counterbalance especial cruelty as a\\nserious aggravating circumstance.” Id. at 1281 (emphasis\\nadded). And finally, when discussing the sentencing judge’s\\ndetermination regarding Spreitz’s emotional immaturity, the\\ncourt stated, “We also find that the sentencing judge . . .\\nproperly found that his emotional immaturity was not a\\nsignificant mitigating factor.” Id. (emphasis added). In\\nother words, when the Arizona Supreme Court wanted to\\nassign weight to a given factor, it said so. This pattern stands\\nin stark contrast to the court’s conclusion that anything\\nrelated to Spreitz’s long-term alcohol or substance abuse that\\nfell short of the statutory definition could not serve as\\nmitigation. Id. at 1280–81.\\n\\n32 SPREITZ V. RYAN\\n\\n Finally, as also discussed earlier, the PCR court\\ndistinctly interpreted the Arizona Supreme Court’s decision\\nas having applied a causal nexus test. The PCR court\\nexplained that “[w]ithout some basis for explaining or\\ndefining the individual’s behavior at the time of the offense,\\n[Spreitz]’s history of alcohol or substance abuse would be\\ninconsequential (which is exactly what the trial court and\\nSupreme Court concluded).” Moreover, in articulating how\\nthe Arizona Supreme Court had applied its causal nexus test\\nin Spreitz’s direct appeal, the PCR court, as had the Arizona\\nSupreme Court, cited to a decision of the Arizona Supreme\\nCourt indisputably applying an unconstitutional nexus test,\\nState v. Kayer, 984 P.2d 31 (Ariz. 1999). See McKinney,\\n813 F.3d at 816 (explaining that Kayer held “that the\\ndefendant’s mental impairment ‘was not established as a\\nnonstatutory mitigating factor’ in part because ‘defendant\\noffered no evidence to show the requisite causal nexus that\\nmental impairment affected his judgment or his actions at the\\ntime of the murder’” (quoting Kayer, 984 P.2d at 46)). The\\nfact that the PCR court interpreted the Arizona Supreme\\nCourt’s opinion as we do—and then itself applied the\\nconstitutionally erroneous nexus test—lends further support\\nto our reading.\\n\\n We recognize that the Arizona Supreme Court’s decision\\non the particular issue before us was framed by language that\\nmight, in a decisional vacuum, suggest the court knew,\\nunderstood, and applied the law with respect to Eddings.\\nThe court acknowledged that Spreitz argued that his history\\nof alcohol and drug abuse served as both a statutory and\\nnonstatutory mitigating factor. Spreitz I, 945 P.2d at 1279.\\nIt then stated that it “must consider any aspect of the\\ndefendant’s character or record . . . relevant to determining\\nwhether the death penalty should be imposed.” Id. And\\nfinally, in reaching the conclusion to affirm Spreitz’s death\\n\\n SPREITZ V. RYAN 33\\n\\nsentence, the court explained that it had “examin[ed] the\\nentire record and reweigh[ed] the applicable aggravating and\\nmitigating factors.” Id. at 1281–82.\\n\\n The Arizona Supreme Court’s acknowledgment of what\\nSpreitz argued does not indicate that it relied on or accepted\\nthat argument as correct. Further, we cannot, given\\nMcKinney, Poyson, and Greenway, proceed in a decisional\\nvacuum.\\n\\n Considered in light of the baseline that the Arizona\\nSupreme Court was consistently applying its\\nunconstitutional causal nexus test in death penalty cases at\\nthe time Spreitz’s appeal was decided, as well as its citations\\nto cases held in McKinney to be applying that standard, we\\nare confident that the manner in which the Arizona Supreme\\nCourt disposed of Spreitz’s evidence of his longstanding\\nalcohol and substance abuse confirms that it applied the\\ncausal nexus test it was using during that period.\\n\\n Styers v. Schriro, 547 F.3d 1026 (9th Cir. 2008) supports\\nthis conclusion. Styers stated that the mere fact that “the\\nArizona Supreme Court stated that it had ‘considered all of\\nthe proffered mitigation,’” is not sufficient to contradict the\\nfact that “its analysis prior to this statement indicated\\notherwise.” Id. at 1035 (internal citation omitted). In Styers,\\nnotwithstanding assertions to the contrary, the Arizona\\nSupreme Court refused to consider evidence of the\\ndefendant’s Post Traumatic Stress Disorder (“PTSD”)\\nbecause he had failed to connect it causally to his crimes. Id.\\nHere, too, the Arizona Supreme Court’s actual analysis of\\nSpreitz’s longstanding alcohol and substance abuse reveals\\nthat it refused to consider this particular aspect of Spreitz’s\\ncharacter free from any connection to the crime. Rather, the\\ncourt’s discussion demonstrates that in evaluating Spreitz’s\\nlong-term alcohol and substance abuse, it applied a causal\\n\\n34 SPREITZ V. RYAN\\n\\nnexus test in violation of Eddings and therefore refused, as a\\nmatter of law, to consider it in assessing Spreitz’s evidence\\nof nonstatutory mitigating circumstances.\\n\\n To the extent the state argues the Arizona Supreme\\nCourt’s opinion is ambiguous as to whether it applied an\\nunconstitutional causal nexus test or not—and we believe\\nthat it is not—McKinney requires resolving that ambiguity\\nin favor of Spreitz. McKinney held it clear that the Arizona\\nSupreme Court “consistently articulated and applied its\\ncausal nexus test” at the time the trial court imposed\\nSpreitz’s death sentence and the Arizona Supreme Court\\nreviewed and affirmed that sentence. McKinney, 813 F.3d\\nat 803 (emphasis in original). Again, “if there is to be a\\npresumption, it is that the Arizona Supreme Court violated\\nthe dictates of Lockett and Eddings.” Greenway, 866 at\\n1095.\\n\\n In sum, because (1) the Arizona Supreme Court clearly\\nstated that Spreitz’s longstanding alcohol and substance\\nabuse was “not essential” to its decision, which was in line\\nwith the court’s consistent practice of applying a causal\\nnexus test; (2) the Arizona Supreme Court cited to Medrano,\\nin which it had previously articulated a causal nexus test;\\n(3) the Arizona Supreme Court carefully laid out the weight\\nit would accord other mitigating factors as opposed to simply\\nignoring them; and (4) the PCR court likewise interpreted\\nthe Arizona Supreme Court’s opinion as applying a causal\\nnexus test, we conclude that the Arizona Supreme Court\\nheld, as a matter of law, that Spreitz’s long-term alcohol and\\nsubstance abuse was not a nonstatutory mitigating factor and\\ntherefore refused to consider it. This holding was in clear\\nviolation of Eddings.\\n\\n SPREITZ V. RYAN 35\\n\\n 3.\\n\\n In light of our conclusion that the Arizona Supreme\\nCourt failed to comply with Eddings, we must next decide\\nwhether Spreitz was prejudiced by this error. On federal\\nhabeas review, relief is warranted if an error “had substantial\\nand injurious effect or influence” on the challenged decision.\\nSee Brecht v. Abrahamson, 507 U.S. 619, 623 (1993). The\\npetitioner is “not entitled to habeas relief” unless he can\\nestablish that the error “resulted in actual prejudice.” Davis\\nv. Ayala, 135 S. Ct. 2187, 2197 (2015) (internal quotation\\nmarks omitted). Here, because we are “in grave doubt about\\nwhether [the Arizona Supreme Court’s] error of federal law\\nhad substantial and injurious effect or influence in\\ndetermining [whether to affirm Spreitz’s death sentence],\\nthat error is not harmless. And, [Spreitz] must win.”\\nMcKinney, 813 F.3d at 822 (quoting O’Neal v. McAninch,\\n513 U.S. 432, 435 (1995)).\\n\\n We previously have required resentencing based on\\nprejudicial Eddings error where the Arizona Supreme Court\\nrefused to consider, as a matter of law, mitigation evidence\\n“central to” the petitioner’s “plea for leniency.” See id. at\\n823; see also Coleman v. Calderon, 210 F.3d 1047, 1051\\n(9th Cir. 2000) (concluding that a penalty-phase jury\\ninstruction error was not harmless under Brecht because “it\\nundermined the very core of Coleman’s plea for life”). 14 For\\nthe reasons that follow, it is clear that Spreitz’s long-term\\n\\n 14\\n We do not suggest that Brecht prejudice may only be established\\nif inappropriately considered mitigating evidence was at “the very core\\nof [a defendant’s] plea for life.” Coleman, 210 F.3d at 1051. It is\\nconceivable that other forms of Eddings error could have “a substantial\\nand injurious effect” on a defendant’s sentence if they would be\\nsignificant to the sentencer, despite the defendant’s failure to rely on\\nthem as central to his or her argument. Brecht, 507 U.S. at 623.\\n\\n36 SPREITZ V. RYAN\\n\\nalcohol and substance abuse was important mitigation\\nevidence, and was so closely related as to become central to\\nhis plea for leniency. The Arizona Supreme Court’s refusal\\nto consider that evidence had a substantial and injurious\\neffect on its sentencing determination that requires\\nresentencing.\\n\\n While a defendant’s evidence of longstanding alcohol\\nand/or substance abuse can be probative mitigation evidence\\non its own, a “history of substance abuse [is] substantially\\nmitigating when . . . combined with other mitigating\\nevidence.” Henry v. Ryan, 720 F.3d 1073, 1090 (9th Cir.\\n2013) (emphasis added) (listing cases and holding that the\\ndefendant’s alcoholism “would not have had a substantial\\nand injurious effect” because it “st[ood] alone, was similar\\nto evidence already considered by the sentencing courts, and\\nwas of limited probative value”) (footnote omitted). 15 Here,\\nSpreitz’s evidence regarding his history of alcohol and\\nsubstance abuse—spanning nearly half his life by the time\\nhe committed the crime at the age twenty-two—was the\\n\\n 15\\n The dissent makes much of Henry, suggesting it dictates the\\noutcome of our prejudice analysis. Dissent at 29. Not so. In that case,\\nthe state court had found that Henry’s intoxication on the day of the\\nmurder did constitute a statutory mitigating factor but was not\\nsufficiently substantial to outweigh the two aggravating factors. Henry,\\n720 F.3d at 1090. Henry argued that the state court erred by failing to\\nconsider evidence of his “historical alcoholism” in addition to his\\nintoxication on the day of the crime. Id. We concluded that any error\\nwas harmless because the history of substance abuse “was similar to\\nevidence already considered by the sentencing courts and was of limited\\nprobative value.” Id. In other words, because the state court had found\\ninsufficient evidence of alcohol use that was tied to the murder, we could\\nnot conclude that general alcohol use that was not tied to the murder\\ncould have been determinative. Here, by contrast, the Arizona Supreme\\nCourt found that Spreitz’s evidence of intoxication did not rise to the\\nlevel of statutory mitigation. Accordingly, there is not the same\\ncontextual harmlessness problem as in Henry.\\n\\n SPREITZ V. RYAN 37\\n\\nfactor on which much of his other mitigation evidence\\nturned.\\n\\n In both his report and his testimony, Dr. Flynn discussed\\nat length the extent to which alcohol consumed Spreitz’s life\\nin the decade leading up to Reid’s murder. Spreitz began\\ndrinking when he was twelve or thirteen; by fifteen, he was\\ndrinking steadily on the weekends and drinking vodka before\\nschool. According to Dr. Flynn, “[d]rug and alcohol abuse\\ndominated his teenage years.” After he left home, Spreitz’s\\ndrinking intensified: Spreitz experienced blackouts multiple\\ntimes and, a week before the murder, Spreitz was\\n“intoxicated a majority of the time.” Dr. Flynn concluded\\nthat “[i]t appears completely clear from the available\\ninformation that . . . Spreitz had a long-standing problem\\nwith alcohol which probably reached the level of physical\\ndependence.”\\n\\n Critically, Spreitz’s decade-long struggle with alcohol\\nand other addictive substances reflects the true gravity of his\\n“subnormal upbringing,” a mitigating circumstance\\nrecognized by the Arizona Supreme Court. See Spreitz I,\\n945 P.2d at 1280. Both Spreitz’s sentencing memorandum\\nand Dr. Flynn detailed how his difficult childhood led to\\nSpreitz’s early dependence on alcohol. For instance, Dr.\\nFlynn explained that Spreitz turned to alcohol as a coping\\nmechanism to deal with the “pervasive subabusive\\nemotional battering and neglect that he received from his\\nmother,” whom he described as “punitive, controlling, [and]\\nemotionally cold.” Dr. Flynn found that Spreitz “slipped\\nquietly into alcoholic numbness” after his “failure to live up\\nto his mother’s standards or earn her love.” Reflecting\\nparental neglect, Spreitz’s parents “failed to arrange . . . for\\nan appropriate rehabilitation program” even though\\n“[v]irtually everyone . . . who spent much time with him\\n\\n38 SPREITZ V. RYAN\\n\\ndescribed him as a heavy drinker.” “Given the intensity and\\nduration” of Spreitz’s reliance on alcohol, “the mother (and\\nstep-father) were either so uninvolved with him to have\\nfailed to notice the problem, or they were aware of it and\\nfailed to arrange (or attempt to arrange) for an appropriate\\nrehabilitation program.”\\n\\n Because Spreitz did not receive treatment during his\\nteenage years, his dependence on alcohol only deepened. In\\nhis sentencing memorandum, Spreitz, too, argued that his\\nlong-term alcohol abuse reflects the impact that his\\nupbringing by an emotionally unavailable, manipulative,\\npunitive mother had on him: it was “[n]o wonder that by 12\\nor 13 [h]e began drinking alcohol and smoking marijuana.”\\nAlthough the Arizona Supreme Court recognized that\\nSpreitz’s “subnormal” upbringing was “a mitigating\\ncircumstance,” it “accord[ed] it little weight.” Spreitz I,\\n945 P.2d at 1280. Had the court properly considered\\nSpreitz’s longstanding history of alcohol abuse in\\nconjunction with and as a manifestation of his disturbing\\nchildhood, both mitigating circumstances—and most\\nimportantly, the two in combination—would have been\\nmore compelling.\\n\\n The dissent argues that Spreitz’s long-term history of\\nalcohol abuse cannot be central to his plea for leniency\\nbecause much of his mitigation evidence focused on his\\n“abusive and dysfunctional childhood, upbringing, and\\ninteractions with his mother,” Dissent at 66, not his history\\nof alcohol and substance abuse. This argument, however,\\nmisses the point, as it overlooks the interlinked relationship\\nbetween Spreitz’s alcoholism and his subnormal childhood:\\nSpreitz’s history of alcohol and substance abuse is central to\\nevaluating the actual significance of his childhood as a\\nmitigating factor, and vice versa.\\n\\n SPREITZ V. RYAN 39\\n\\n In addition to being a symptom of his dysfunctional\\nchildhood, Spreitz’s alcohol and substance abuse can be seen\\nas linked to his emotional immaturity, another nonstatutory\\nmitigating circumstance recognized by the Arizona Supreme\\nCourt but described as not “significant.” Spreitz I, 945 P.2d\\nat 1282. In his report, Dr. Flynn explained that Spreitz was\\n“very likely to have been socially and emotionally\\nimmature” due, at least in part, to “[y]ears of alcoholism\\nintoxication [that] wipe[d] out many of the . . . healthy\\ndevelopmental processes requisite to age-appropriate social\\nand emotional maturity.” Similarly, in his sentencing\\nmemorandum, Spreitz connected “[h]is developmental\\nfailure” with his “[n]umbing by use of alcohol.” Again, had\\nthe Arizona Supreme Court properly considered Spreitz’s\\nlongstanding alcohol and substance abuse, another\\nmitigating factor—Spreitz’s emotional immaturity—would\\nhave been brought into proper perspective.\\n\\n Thus, by failing to consider Spreitz’s longstanding\\nalcohol and substance abuse, the Arizona Supreme Court\\nwas left with a critical void in Spreitz’s narrative: Spreitz’s\\nsubnormal childhood was so emotionally disturbing that it\\nled him to drink by the age of twelve or thirteen, which, in\\nturn, disrupted his normal development and contributed to\\nhis emotional immaturity. As both a symptom and cause of\\nother mitigating factors, Spreitz’s longstanding alcohol\\nabuse was thus central to his plea for leniency. 16 Without\\n\\n 16\\n Contrary to the dissent’s suggestion, Spreitz did not focus on his\\nintoxication at the time of the crime to the exclusion of his history of\\nalcohol abuse. Dissent at 22–24. Rather, both he and Dr. Flynn\\ndistinguished between Spreitz’s history of alcohol abuse as nonstatutory\\nmitigation and his intoxication at the time of the crime as statutory\\nmitigation. It is Spreitz’s history of alcohol abuse that underlies and\\namplifies much of his other mitigation evidence, not his intoxication at\\nthe time of Reid’s murder.\\n\\n40 SPREITZ V. RYAN\\n\\nconsideration of the evidence of his long-term substance and\\nalcohol abuse, the Arizona Supreme Court’s conclusions\\nregarding all of Spretiz’s mitigation evidence were\\nfundamentally flawed.\\n\\n We come to this conclusion fully recognizing, as the\\ndissent emphasizes, that “the severity of an aggravating\\ncircumstance must be considered when determining whether\\nan Eddings error is harmless.” Dissent at 71. The Arizona\\nSupreme Court described what Spreitz’s “confession and\\nphysical evidence” revealed: that “Spreitz beat and raped\\n[Reid] in a brutal assault that lasted many minutes before he\\ncrushed her skull.” Spreitz I, 945 P.2d at 1279. For this\\nreason, the Arizona Supreme Court affirmed the sentencing\\ncourt’s finding that Reid’s murder was especially cruel. Id.\\nEven so, we found prejudicial nexus-error in McKinney\\nwhere the murder was done in a cruel manner for pecuniary\\ngain as part of a double murder. McKinney, 813 F.3d at 823.\\n“When the choice is between life and death,” “the risk that\\nthe death penalty will be imposed in spite of facts which may\\ncall for a less severe penalty. . . . is unacceptable.” Abdul-\\nKabir v. Quarterman, 550 U.S. 233, 264 (2007) (quoting\\nLockett v. Ohio, 438 U.S. 586, 605 (1978)). Accordingly,\\nthe sentencer “must be able to give meaningful consideration\\nand effect to all mitigating evidence that might provide a\\nbasis for refusing to impose the death penalty on a particular\\nindividual, notwithstanding the severity of his crime.” Id. at\\n247 (emphasis added).\\n\\n Here, the Arizona Supreme Court failed to give any\\nmeaning to Spreitz’s longstanding alcohol and substance\\nabuse, an oversight which, in turn, minimized the value of\\nother mitigating evidence as well. Because the court’s\\nanalysis was missing a core component, we “cannot say,\\nwith fair assurance” that the Arizona Supreme Court’s\\n\\n SPREITZ V. RYAN 41\\n\\nEddings error did not affect its decision. McKinney,\\n813 F.3d at 822. As a result, the Arizona Supreme Court’s\\nrefusal to consider the evidence was not harmless under\\nBrecht.\\n\\n IV. Conclusion\\n\\n We reverse in part the district court’s judgment denying\\na writ of habeas corpus. We remand with instructions to\\ngrant the writ with respect to Spreitz’s sentence unless the\\nstate, within a reasonable period, either corrects the\\nconstitutional error in his death sentence or vacates the\\nsentence and imposes a lesser sentence consistent with the\\nlaw.\\n\\n REVERSED in part and REMANDED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4373176/\", \"author_raw\": \"PAEZ, Circuit Judge:\"}, {\"author\": \"TALLMAN, Circuit Judge, dissenting\", \"type\": \"dissent\", \"text\": \"TALLMAN, Circuit Judge, dissenting:\\n\\n The court declares that Christopher Spreitz’s death\\nsentence should be vacated because, applying de novo\\nreview, the sentencing court and the Arizona Supreme Court\\nviolated Eddings v. Oklahoma, 455 U.S. 104 (1982), and that\\nthis error was not harmless. I respectfully dissent.\\n\\n First, the record does not establish that either court\\nunconstitutionally refused to consider relevant mitigating\\nevidence. They were simply not persuaded by it. What the\\nrecord clearly shows is the Arizona courts did not find that\\nSpreitz’s proffered evidence outweighed the aggravating\\nfactor of cruelty in the way by which he brutally murdered\\nRuby Reid. Second, even if we determined that the Arizona\\ncourts did violate Eddings, Spreitz cannot show that this\\nerror had a “substantial and injurious effect or influence” on\\n\\n42 SPREITZ V. RYAN\\n\\nhis ultimate sentence. Brecht v. Abrahamson, 507 U.S. 619,\\n623 (1993). Given the savage brutality of his crimes and the\\nweakness of all the mitigation evidence he presented, any\\nremand for a redetermination of the appropriate penalty\\nought to result in the same decision.\\n\\n Nonetheless, the majority attempts unsuccessfully to\\nnavigate the course that Spreitz’s Eddings claim requires\\nhim to sail to warrant federal habeas relief. In doing so, the\\nmajority has once again foundered on the shoals of the\\nAntiterrorism and Effective Death Penalty Act of 1996\\n(AEDPA), specifically 28 U.S.C. § 2254—where so many\\nof our prior habeas corpus decisions are wrecked—failing to\\nafford the respect due the decisions of our sister state courts\\nand taunting the Supreme Court for another capsizing in lieu\\nof safe passage.\\n\\n I\\n\\n We review the Arizona post-conviction relief (PCR)\\ncourt’s decision because it is the last reasoned state court\\ndecision on Spreitz’s Eddings claim. See Barker v. Fleming,\\n423 F.3d 1085, 1091 (9th Cir. 2005). Under AEDPA we first\\nmust decide the level of deference to give to that decision by\\ndetermining whether it “was contrary to, or involved an\\nunreasonable application of, clearly established Federal law,\\nas determined by the Supreme Court of the United States.”\\nId. § 2254(d)(1).\\n\\n I do not quarrel with the majority’s determination that\\nthe PCR court held a mistaken view of what Eddings\\nrequires, and that we therefore must review Spreitz’s\\nEddings claim de novo. See Castellanos v. Small, 766 F.3d\\n1137, 1146 (9th Cir. 2014). However, “we may not grant\\nhabeas relief simply because of [the PCR court’s]\\n§ 2254(d)(1) error . . . .” Frantz v. Hazey, 533 F.3d 724, 735\\n\\n SPREITZ V. RYAN 43\\n\\n(9th Cir. 2008) (en banc); see also id. at 728, 735–37\\n(explaining our “approach to reviewing state court decisions\\nthat rely on legal principles contradicting clearly established\\nSupreme Court law but do not necessarily reach the wrong\\nresult”).\\n\\n On de novo review, Spreitz can meet neither of the two\\nrequirements to obtain habeas relief on his Eddings claim.\\nFirst, he cannot demonstrate that the state sentencing courts\\nactually violated Eddings. And second, even if there was a\\nviolation, he cannot show that any Eddings error by those\\nstate courts was prejudicial under Brecht.\\n\\n II\\n\\n Spreitz’s Eddings claim rests in large part on the\\nerroneous assumption that the sentencing court and the\\nArizona Supreme Court failed to consider his history of\\nalcohol and substance abuse, unrelated to the offense, as\\nnon-statutory mitigation when determining his sentence. He\\ncontends that, by failing to consider all mitigating\\ncircumstances, the state courts violated Eddings, which\\nprovides that in capital cases “the sentencer may not refuse\\nto consider or be precluded from considering ‘any relevant\\nmitigating evidence.’” Skipper v. South Carolina, 476 U.S.\\n1, 4 (1986) (quoting Eddings, 455 U.S. at 114). Eddings also\\nprovides that the sentencer “may determine the weight to be\\ngiven relevant mitigating evidence.” 455 U.S. at 114–15.\\n\\n Recently, we adopted an erroneous presumption that the\\nArizona Supreme Court repeatedly violated the dictates of\\nEddings by “consistently” applying an unconstitutional\\ncausal nexus test to all capital cases between 1989 and 2005.\\nSee McKinney v. Ryan, 813 F.3d 798, 803 (9th Cir. 2015)\\n\\n44 SPREITZ V. RYAN\\n\\n(en banc), cert. denied, 137 S. Ct. 39 (2016). 1 “We did not\\nsay, however, that the Arizona [Supreme Court] always\\napplied it.” See Greenway v. Ryan, 866 F.3d 1095, 1095 (9th\\nCir. 2017) (per curiam). Notably, in listing the cases in\\nwhich the Arizona courts erroneously applied the causal\\nnexus test, McKinney did not include Spreitz’s case. See\\n813 F.3d at 815–16, 824–26. And in McKinney, our holding\\nresolved only the “precise question” whether the state court\\nin that specific case had applied the causal nexus test. Id. at\\n804; see also Hedlund v. Ryan, 854 F.3d 557, 586 n.22 (9th\\nCir. 2017) (“We express no opinion as to how to apply\\nMcKinney in future Arizona capital cases from the suspect\\ntime period.”). Each case must be reviewed on its unique\\nfacts. We therefore must examine the state court decisions\\nin Spreitz’s case to determine whether they properly took\\ninto account all mitigating factors. 2\\n\\n\\n\\n\\n 1\\n The real damage inflicted by the en banc decision in McKinney is\\nthe broad assumption it adopts that now infects every death sentence\\nimposed by Arizona courts for horrendous murders adjudicated in the\\nsixteen-year period. It is regrettable that the United States Supreme\\nCourt denied certiorari in McKinney. But this case presents yet another\\nopportunity to right the error in Ninth Circuit death penalty habeas\\njurisprudence.\\n 2\\n In reviewing the last reasoned state court decision, we may\\nconsider “the record that was before the state court that adjudicated the\\nclaim on the merits.” Cannedy v. Adams, 706 F.3d 1148, 1156 (9th Cir.\\n2013) (citation omitted). The record before the PCR court in this case\\nincluded the sentencing judge’s decision and the Arizona Supreme\\nCourt’s direct appeal ruling. Thus, despite the majority’s preference for\\ncabining its view of the record to only the Arizona Supreme Court’s\\ndecision, we must consider both decisions here.\\n\\n SPREITZ V. RYAN 45\\n\\n A\\n\\n At all relevant times in this case, the operative Arizona\\ndeath penalty statute required a sentencing court to “impose\\na sentence of death if the court finds one or more of the\\naggravating circumstances enumerated in [the statute and\\ndetermines] that there are no mitigating circumstances\\nsufficiently substantial to call for leniency.” Ariz. Rev. Stat.\\n§ 13-703(E) (1993) (current version at Ariz. Rev. Stat.\\n§ 13-751(E) (2012)). Arizona law provided five statutory\\nmitigating factors, see § 13-703(G), including a modified\\nform of diminished capacity: “The defendant’s capacity to\\nappreciate the wrongfulness of his conduct or to conform his\\nconduct to the requirements of the law was significantly\\nimpaired, but not so impaired as to constitute a defense to\\nprosecution.” Id. § 13-703(G)(1). Even our en banc court\\nacknowledged, “When applied solely in the context of\\nstatutory mitigation under § 13-703(G)(1), the causal nexus\\ntest does not violate Eddings.” McKinney, 813 F.3d at 810.\\n\\n But Arizona law also provided for consideration of\\ncatch-all, non-statutory mitigating evidence that\\nencompassed “any factors proffered by the defendant or the\\nstate that are relevant in determining whether to impose a\\nsentence less than death, including any aspect of the\\ndefendant’s character, propensities or record and any of the\\ncircumstances of the offense.” Ariz. Rev. Stat. § 13-703(G).\\nThe dictates of Eddings apply to such non-statutory\\nmitigating evidence. See McKinney, 813 F.3d at 810\\n(“When applied in the context of nonstatutory mitigation, the\\ncausal nexus test clearly violates Eddings.”). Lastly,\\nArizona law requires the Arizona Supreme Court to\\n“independently review the trial court’s findings of\\naggravation and mitigation and the propriety of the death\\n\\n46 SPREITZ V. RYAN\\n\\nsentence.” Ariz. Rev. Stat. § 13-703.01(A) (1994) (current\\nversion at Ariz. Rev. Stat. § 13-755(A) (2009)).\\n\\n B\\n\\n Here, Spreitz cannot show that the sentencing judge, the\\nHonorable William N. Sherrill, violated Eddings in\\ndetermining Spreitz’s sentence. At sentencing, Spreitz\\nsubmitted a memorandum that specifically discussed his\\nhistory of substance abuse and cited Eddings; defense\\npsychologist Dr. Todd C. Flynn provided the court with a\\npsychological evaluation report and testimony specifically\\ndiscussing Spreitz’s history of substance abuse as both\\nstatutory and non-statutory mitigating factors; and Judge\\nSherrill differentiated between the two types of mitigation\\nand ultimately decided that the mitigating evidence did not\\noutweigh the aggravating factor of cruelty based on how\\nSpreitz murdered Ms. Reid.\\n\\n In his written judgment, Judge Sherrill acknowledged\\nthat “many factors have been submitted in mitigation.”\\nAmong the mitigating evidence, he observed that Spreitz “in\\nhis life turned to substance abuse—alcohol and some\\nsuggestion he was using cocaine and other drugs”—and that\\nSpreitz’s “history of intoxication [was] longstanding.” He\\nalso found that Spreitz “had been abusing substances for\\nclose to ten years of his life.” Judge Sherrill’s discussion of\\nSpreitz’s history of substance abuse unrelated to the crime\\nestablishes that he considered that evidence in his sentencing\\ndecision. See Williams v. Stewart, 441 F.3d 1030, 1057 (9th\\nCir. 2006) (stating that a sentencing court does not violate\\nEddings where it is clear that “the sentencing court\\nconsidered all relevant mitigating evidence that was\\noffered”).\\n\\n SPREITZ V. RYAN 47\\n\\n There is an important distinction between Judge\\nSherrill’s consideration of Spreitz’s evidence of his long-\\nterm substance abuse, and Judge Sherrill’s subsequent\\nfindings regarding the weight he ultimately decided to give\\nthat evidence. After discussing Spreitz’s proffered\\nmitigating evidence, Judge Sherrill correctly observed that\\nthe issue was, in weighing the imposition of capital\\npunishment, “whether any of these mitigating\\ncircumstances, individually or cumulatively, can balance or\\noutweigh the aggravating circumstances.” And, he finally\\ndetermined that “the mitigating circumstances [were] not\\nsufficient to balance the aggravating circumstances, nor\\n[were] they sufficiently substantial to call for leniency.” In\\nreaching that conclusion, Judge Sherrill specified that he did\\n“not believe that intoxication [was] any sort of mitigating\\ncircumstance.” (Emphasis added.) One should fairly read\\nthat conclusion to mean that Spreitz’s long history of\\nsubstance abuse was neither a statutory nor a non-statutory\\nmitigating factor “sufficiently substantial to call for\\nleniency.” Ariz. Rev. Stat. § 13-703(E).\\n\\n Under Eddings, Judge Sherrill was free to conclude that\\nSpreitz’s mitigating evidence, whether statutory or non-\\nstatutory, simply did not warrant leniency given the\\naggravating circumstance of how Spreitz cruelly murdered\\nRuby Reid. See Mann v. Ryan, 828 F.3d 1143, 1159–60 (9th\\nCir. 2016) (holding that defendant “failed to show that the\\nmitigating circumstances he had presented,” including a\\nhistory of substance abuse, “outweighed the aggravating\\ncircumstances of his crimes”), cert. denied, 137 S. Ct. 1091\\n(2017), 137 S. Ct. 1205 (2017); Henry v. Ryan, 720 F.3d\\n1073, 1090 n.11 (9th Cir. 2013) (recognizing that “historical\\nalcoholism might [be] considered aggravating as well as\\nmitigating, depending on the perspective of the sentencing\\ncourt”). “It is sufficient that a sentencing court state that it\\n\\n48 SPREITZ V. RYAN\\n\\nfound no mitigating circumstances that outweigh the\\naggravating circumstances.” Williams, 441 F.3d at 1057\\n(quotation omitted).\\n\\n Eddings also did not prevent Judge Sherrill from giving\\nless weight to Spreitz’s substance abuse history based on a\\nlack of nexus to the murder. See Hedlund, 854 F.3d at 587\\nn.23 (stating that under Eddings, “a court is free to assign\\nless weight to mitigating factors that did not influence a\\ndefendant’s conduct at the time of the crime”); Styers v.\\nRyan, 811 F.3d 292, 298–99 (9th Cir. 2015) (holding that the\\nArizona Supreme Court did not violate Eddings in assigning\\nlittle weight to the petitioner’s PTSD in the absence of a\\ncausal connection to the crime), cert. denied, 137 S. Ct. 1332\\n(2017). Nor did Eddings prevent Judge Sherrill from\\nassigning no weight to Spreitz’s long-term substance abuse\\nunrelated to the offense, so long as Judge Sherrill considered\\nthat evidence, which the record shows he definitely did. See\\nWilliams, 441 F.3d at 1057 (“Once mitigating evidence is\\nallowed in, a finding that there are ‘no mitigating\\ncircumstances’ does not violate the Constitution.”); Ortiz v.\\nStewart, 149 F.3d 923, 943 (9th Cir. 1998) (“[T]he\\nsentencing court did not refuse to consider mitigating\\nevidence; it considered the evidence and found it inadequate\\nto justify leniency. That assessment did not violate the\\nConstitution.”), overruled in part on other grounds by\\nMartinez v. Ryan, 566 U.S. 1 (2012).\\n\\n In addition to the above findings, Judge Sherrill also\\nadded that he did not “believe that [Spreitz’s] substance\\nabuse or intoxication impaired [his] ability and capacity to\\nappreciate the wrongfulness of his conduct to any significant\\ndegree” under § 13-703(G)(1). At the time, Arizona courts\\nwere required by law to examine whether evidence of\\nalcohol and drug use bore a causal connection to the crime.\\n\\n SPREITZ V. RYAN 49\\n\\nAriz. Rev. Stat. § 13-703(G)(1) (requiring the sentencer to\\nconsider leniency if the defendant proved that his “capacity\\nto appreciate the wrongfulness of his conduct or to conform\\nhis conduct to the requirements of law was significantly\\nimpaired, but not so impaired as to constitute a defense to\\nprosecution”). And, there is no question that Spreitz pursued\\na mitigation strategy at sentencing that attempted to link his\\nlong-term substance abuse with the kidnap/murder to show\\nstatutory mitigation under § 13-703(G)(1).\\n\\n For example, Spreitz’s sentencing memorandum stated\\nthat Dr. Flynn’s insights about him were of “the nature, type\\nand character contemplated by A.R.S. Section\\n13-703(G)(1),” and that “Dr. Flynn’s report [was], indeed,\\nrelevant and should be considered by the Court as statutory\\nmitigating circumstances.” Dr. Flynn’s report and expert\\ntestimony went to great lengths to argue that Spreitz’s\\nalcohol abuse was “very strongly correlated with violent\\nbehavior . . . .” Judge Sherrill could not respond to Spreitz’s\\nargument that § 13-703(G)(1) called for leniency without\\nanalyzing whether Spreitz had shown a connection between\\nhis substance abuse and the murder. We cannot infer from\\nthat causal nexus analysis that Judge Sherrill flatly refused\\nto consider any evidence of substance abuse that was not\\ncausally related to the crime.\\n\\n Rather, this record shows that Judge Sherrill\\nconstitutionally applied a causal nexus test in the context of\\nan Arizona statutory mitigating factor, § 13-703(G)(1). See\\nMcKinney, 813 F.3d at 810 (“When applied solely in the\\ncontext of statutory mitigation . . . , the causal nexus test\\ndoes not violate Eddings.”). Judge Sherrill’s other\\nstatements repeatedly demonstrate that he understood\\nEddings’ mandate and considered all of Spreitz’s proffered\\nmitigating evidence, but ultimately found that evidence\\n\\n50 SPREITZ V. RYAN\\n\\ninsufficient to warrant leniency. We are in the same\\nsituation here that recently confronted our Court in Hedlund.\\nSee 854 F.3d at 591 (Bea, J., concurring) (“[Although] Judge\\nSheldon constitutionally applied a causal-nexus test in the\\ncontext of an Arizona statutory mitigating factor[,\\n§ 13-703(G)(1), that] statement does not show that Judge\\nSheldon excluded mitigating evidence from his\\nconsideration, and Judge Sheldon’s other statements\\nrepeatedly demonstrate otherwise.”).\\n\\n Nor does the record before us support the conclusion that\\nJudge Sherrill failed to consider evidence of Spreitz’s long-\\nterm substance abuse as a non-statutory mitigating factor.\\nWhat we can say for sure is that he was not persuaded by\\nthat evidence. Judge Sherrill listened to testimony and\\narguments at trial and in the penalty phase, read everything\\nsubmitted, and then deliberated for three days before\\nconcluding that Spreitz’s mitigation evidence was not\\nsufficient to outweigh the evidence in aggravation\\nwarranting death. There is no Eddings error here.\\n\\n C\\n\\n Similarly, in its “detailed independent review” of\\nSpreitz’s sentence, the Arizona Supreme Court “examined\\nthe entire record to weigh and consider the aggravating and\\nmitigating circumstances.” State v. Spreitz, 945 P.2d 1260,\\n1278 (Ariz. 1997). Consistent with Eddings, the Arizona\\nSupreme Court recognized “that the sentencing judge must\\nconsider any aspect of the defendant’s character or record\\nand any circumstance of the offense relevant to determining\\nwhether the death penalty should be imposed.” Id. at 1279\\n(quotation omitted). And, it recognized “that the weight\\naccorded such evidence is within the sentencing judge’s\\ndiscretion.” Id.; see Greenway, 866 F.3d at 1097 (citing\\nsimilar language in the Arizona Supreme Court’s review of\\n\\n SPREITZ V. RYAN 51\\n\\na capital sentence to conclude that the court “did not reject\\nany mitigating factor, as a matter of law, on the theory that\\nit was not related to the commission of the crime”).\\n\\n In that review, the Arizona Supreme Court specifically\\nobserved that the “record demonstrates defendant’s longtime\\nsubstance abuse problems.” Spreitz, 945 P.2d at 1280.\\nNonetheless, the court found that Spreitz’s “general\\nproblems with substance abuse [were] not essential to [its]\\ndecision” because, ultimately, “the aggravating\\ncircumstance of especial cruelty in [Spreitz]’s murder of\\nRuby Reid outweigh[ed] all factors mitigating in favor of\\nleniency.” Id. at 1280, 1282. The Arizona Supreme Court\\nthus rejected, on the merits, Spreitz’s argument that his long-\\nterm substance abuse constituted either a statutory or non-\\nstatutory mitigating factor warranting leniency in his case.\\nThere is no mitigation evidence it refused to consider.\\n\\n To the extent that the majority cites Poyson v. Ryan,\\n879 F.3d 875 (9th Cir. 2018), for the proposition that\\nMcKinney, 813 F.3d at 802–03, adopts a per se rule that the\\nArizona Supreme Court applied an unconstitutional causal\\nnexus test in all capital cases over a 15-year period, I fully\\nagree with Judge Ikuta’s reluctant concurrence in Poyson.\\n879 F.3d at 897–900. McKinney was wrongly decided and\\nflips the “presumption that state courts know and follow the\\nlaw.” Id. at 897–98 (citing Woodford v. Visciotti, 537 U.S.\\n19, 24 (2002)). The record here rebukes even the McKinney\\npresumption and in that respect is factually distinguishable\\nfrom both McKinney and Poyson.\\n\\n The majority nonetheless concludes that the Arizona\\nSupreme Court’s use of the term “not essential” proves that\\nthe court “require[d] that Spreitz establish a causal\\nconnection between his longstanding substance abuse and\\nthe murder before considering and weighing the evidence as\\n\\n52 SPREITZ V. RYAN\\n\\na nonstatutory mitigating factor.” Op. at 22–23, 29. Not so.\\nWe can reasonably interpret “not essential” to mean\\ninconsequential, which we have previously defined in a\\nsimilar context as having little effect on the outcome. See\\nLandrigan v. Stewart, 272 F.3d 1221, 1230 (9th Cir. 2001)\\n(“At any rate, any error in failing to consider Landrigan’s\\nuse of alcohol and drugs would have been inconsequential;\\nit would have had no effect whatsoever on the outcome.”),\\nadopted sub nom. Landrigan v. Schriro, 501 F.3d 1147 (9th\\nCir. 2007) (en banc).\\n\\n The majority reasons that, unlike the Arizona Supreme\\nCourt’s consideration of Spreitz’s substance abuse history\\nunrelated to the murder, “when the court considered other\\nmitigating factors, it specifically discussed the weight it\\nwould give to those factors . . . .” Op. at 31. But the law has\\nnever imposed an obligation on the sentencer to explicitly\\ndisclose the value it assigns to every piece of mitigating\\nevidence. “The U.S. Supreme Court has ‘never held that a\\nspecific method for balancing mitigating and aggravating\\nfactors in a capital sentencing proceeding is constitutionally\\nrequired.’” Styers, 811 F.3d at 298 (quoting Kansas v.\\nMarsh, 548 U.S. 163, 175 (2006)). Nor did McKinney\\nabrogate the principle that a sentencing court is not required\\nto “itemize and discuss every piece of evidence offered in\\nmitigation.” Jeffers v. Lewis, 38 F.3d 411, 418 (9th Cir.\\n1994) (en banc). “We have determined that a sentencing\\ncourt need not specifically discuss each individual item of\\nmitigating evidence so long as it appears to have considered\\nall relevant evidence.” Ortiz, 149 F.3d at 943. This is\\nparticularly important when reviewing Eddings claims\\nbecause otherwise we could infer that a sentencing court\\nrefused to consider any piece of evidence to which it did not\\nexpressly assign a value, even though it sufficiently\\ndiscussed that proffered evidence. That is not the law.\\n\\n SPREITZ V. RYAN 53\\n\\n For example, in Jeffers, our en banc court held that the\\nstate court’s “failure to list and discuss each item of evidence\\n[that the defendant] offered in mitigation” did not\\ndemonstrate that the “court neglected to consider all of his\\nmitigation evidence.” 38 F.3d at 417–18. Instead, we held\\nthat there was no clear requirement under either Arizona law\\nor the federal Constitution that the sentencer itemize and\\ndiscuss each piece of evidence offered in mitigation. Id. at\\n418. And, we held that the sentencing court there clearly\\nconsidered all such evidence. Id. In addition, a sentencing\\ncourt “need not exhaustively analyze each mitigating factor\\nas long as a reviewing federal court can discern from the\\nrecord that the state court did indeed consider all mitigating\\nevidence offered by the defendant.” Moormann v. Schriro,\\n426 F.3d 1044, 1055 (9th Cir. 2005) (quotation omitted); see\\nalso Lopez v. Schriro, 491 F.3d 1029, 1039 (9th Cir. 2007)\\n(observing that there is no “clearly established Supreme\\nCourt precedent setting forth the record the sentencing court\\nmust make in order to permit sufficient appellate review”\\n(quotation omitted)). We can easily make that determination\\nhere.\\n\\n The majority also focuses on the Arizona Supreme\\nCourt’s additional finding that Spreitz “was [not] impaired\\nby alcohol consumption to an extent that it interfered with\\nhis ‘capacity to appreciate the wrongfulness of his conduct\\nor to conform his conduct to the requirements of the law.’”\\nOp. at 28 (quoting Spreitz, 945 P.2d at 1280–81). But this\\nstatement does not mean that the Arizona Supreme Court\\nrefused to consider Spreitz’s long-term substance abuse\\nunrelated to the murder. On the contrary, as the majority\\nrecognizes, the Arizona Supreme Court’s “decision . . .\\nsuggest[ed] the court knew, understood, and applied the law\\nwith respect to Eddings.” Id. at 32. For example, the\\nArizona Supreme Court stated: “In our review, we have\\n\\n54 SPREITZ V. RYAN\\n\\nbeen mindful that the sentencing judge must consider any\\naspect of the defendant’s character or record and any\\ncircumstance of the offense relevant to determining whether\\nthe death penalty should be imposed.” Spreitz, 945 P.2d at\\n1279 (quotations omitted) (emphasis added). After\\nindependently “weighing the mitigating circumstances both\\nindividually and cumulatively against the aggravating\\ncircumstance,” it ultimately found that “the aggravator of\\nespecial cruelty outweighed all other circumstances.” Id. at\\n1280. Those findings do not violate Eddings. They reflect\\na careful consideration of all of the evidence, ultimately\\nconcluding that the death sentence was warranted here\\ndespite what Spreitz offered in mitigation. Thus, there was\\nno Eddings violation.\\n\\n III\\n\\n Even if we were to determine that the state courts\\ncommitted Eddings error by failing to consider Spreitz’s\\nlong-term substance abuse unrelated to the offense, Spreitz\\ncannot show that this error was prejudicial because the\\nviolent manner in which he abducted and murdered Ruby\\nReid trenchantly outweighed all evidence proffered in\\nmitigation. See Greenway, 866 F.3d at 1100 (denying\\nhabeas relief on petitioner’s Eddings claim because the\\nArizona Supreme Court’s Eddings error was harmless).\\nThus, even assuming that the state courts unconstitutionally\\nexcluded that evidence, their factual determinations make\\nclear that the evidence would have been entitled to little\\nmitigating weight in their ultimate sentencing\\ndeterminations.\\n\\n “The harmless-error standard on habeas review provides\\nthat ‘relief must be granted’ only if the error ‘had substantial\\nand injurious effect or influence in determining the [sentence\\nimposed].’” McKinney, 813 F.3d at 822 (quoting Brecht,\\n\\n SPREITZ V. RYAN 55\\n\\n507 U.S. at 623). The United States Supreme Court has\\ndistinguished this harmless error standard from the standard\\napplicable to cases on direct review, which require merely a\\n“reasonable possibility” that the error affected the outcome\\nof the proceedings. See Brecht, 507 U.S. at 634–38 (citing\\nChapman v. California, 386 U.S. 18, 24 (1967)). Unlike\\ncases on direct review, “granting habeas relief merely\\nbecause there is a ‘reasonable possibility’ that [the] error\\ncontributed to the [outcome of the proceedings] is at odds\\nwith the historic meaning of habeas corpus.” Id. at 637\\n(citation and quotations omitted). As such, “an error that\\nmay justify reversal on direct appeal will not necessarily\\nsupport a collateral attack on a final judgment.” Id. at 634\\n(citation and quotations omitted).\\n\\n In McKinney, we clarified that a state court’s Eddings\\nviolation is harmless if we can “say[] with fair assurance”\\nthat “the judgment was not substantially swayed by the\\nerror.” 813 F.3d at 822 (quoting Kotteakos v. United States,\\n328 U.S. 750, 765 (1946)) (emphasis added). On the other\\nhand, an Eddings error is not harmless if we are in “grave\\ndoubt” about whether it had a substantial and injurious effect\\nor influence in determining the sentence imposed. Id.\\n(quoting O’Neal v. McAninch, 513 U.S. 432, 436 (1995)).\\nMcKinney held that the excluded evidence there “was\\nimportant mitigating evidence” because it was “central to\\n[McKinney’s] plea for leniency.” 813 F.3d at 823 (emphasis\\nadded). It cited Coleman v. Calderon, 210 F.3d 1047 (9th\\nCir. 2000), where we held that an unconstitutional jury\\ninstruction at sentencing was not harmless because “it\\nundermined the very core of Coleman’s plea for life.” Id. at\\n1051 (emphasis added).\\n\\n Here, based on what the Arizona courts repeatedly said,\\nthere is no “grave doubt” as to whether any Eddings error\\n\\n56 SPREITZ V. RYAN\\n\\nsubstantially influenced or swayed Spreitz’s ultimate\\nsentence. The courts at every level consistently found that\\n“the aggravating circumstance of especial cruelty in\\n[Spreitz’s] murder of Ruby Reid outweigh[ed] all factors\\nmitigating in favor of leniency.” Spreitz, 945 P.2d at 1282.\\nIn considering the entire record, the Arizona Supreme Court\\nexplained:\\n\\n We agree with the sentencing judge that\\n defendant’s upbringing was subnormal. The\\n record supports the judge’s conclusion that\\n defendant’s home life was sadly lacking and\\n that his mother’s erratic behavior toward\\n defendant inhibited his emotional\\n development and social skills . . . . Although\\n we recognize defendant’s upbringing as a\\n mitigating circumstance, we accord it little\\n weight. While defendant’s inadequate\\n upbringing may have contributed to his\\n emotional immaturity and undeveloped\\n humanitarian skills, we concur with\\n defendant’s statement at his sentencing\\n hearing that “people that have had as bad a\\n background or worse haven’t killed. And I\\n don’t want what everyone has said about my\\n background to be an excuse for what’s\\n happened.”\\n\\n ....\\n\\n The record demonstrates defendant’s\\n longtime substance abuse problems. We\\n note, however, that defendant’s general\\n problems with substance abuse are not\\n essential to our decision here. We therefore\\n\\n SPREITZ V. RYAN 57\\n\\ndecline to conclude that defendant was\\nimpaired by alcohol consumption to an extent\\nthat it interfered with his “capacity to\\nappreciate the wrongfulness of his conduct or\\nto conform his conduct to the requirements of\\nthe law.” A.R.S. § 13-703(G)(1) . . . .\\n\\n[W]e find that defendant expressed remorse\\nfor the victim’s death on more than one\\noccasion . . . . We recognize remorse as a\\nnonstatutory mitigating factor ....\\nHowever, defendant’s remorse for his actions\\ndoes little to counterbalance especial cruelty\\nas a serious aggravating circumstance in Ms.\\nReid’s murder. According to defendant’s\\nconfession, when he left Ms. Reid in the\\ndesert early the morning of May 19, 1989, he\\ndid not know whether she was alive or dead.\\nHe confessed that he rode his bicycle out to\\nthe murder site several days later to see if her\\nbody was still there, hoping that it would not\\nbe, that she was still alive. We would find\\ndefendant’s remorse a more compelling\\nmitigating factor if, for example, it had\\nprompted him to report his actions toward\\nMs. Reid to the authorities.\\n\\nThe sentencing judge found that defendant’s\\nability to appreciate the wrongfulness of his\\nconduct was not impaired on the night of the\\nmurder to any significant extent by substance\\nabuse, emotional disorders, situational stress,\\nor by a combination of these. Our review of\\nthe record convinces us that the trial court’s\\nfinding was proper . . . .\\n\\n58 SPREITZ V. RYAN\\n\\n We agree that the record supports the\\n sentencing judge’s findings that defendant\\n had no previous adult felony convictions, no\\n prior record of acts of violence, and that\\n defendant is capable of rehabilitation. We\\n also find that the sentencing judge correctly\\n rejected defendant’s age of twenty-two as a\\n mitigating circumstance and properly found\\n that his emotional immaturity was not a\\n significant mitigating factor.\\n\\n After examining the entire record and\\n reweighing the applicable aggravating and\\n mitigating factors, we find that the\\n aggravating circumstance of especial cruelty\\n in defendant’s murder of Ruby Reid\\n outweighs all factors mitigating in favor of\\n leniency.\\n\\n ....\\n\\n We have conducted an independent review of\\n defendant’s aggravating and mitigating\\n circumstances as required by A.R.S. section\\n 13-703.01 and find that the mitigating\\n circumstances cumulatively are not\\n sufficiently substantial to warrant leniency in\\n relation to the aggravating circumstance of\\n cruelty.\\n\\nId. at 1280–83.\\n\\n SPREITZ V. RYAN 59\\n\\n Despite this explanatory language summarizing its\\nanalysis of all of the evidence from the sentencing hearing,\\nthe majority nonetheless insists that the Arizona Supreme\\nCourt refused to consider Spreitz’s long-term substance\\nabuse as a non-statutory mitigating factor, and concludes this\\nrefusal “had a substantial and injurious effect on [Spreitz’s\\nsentence].” Op. at 35–36. My colleagues reason that\\n“Spreitz’s long-term alcohol and substance abuse was\\nimportant mitigation evidence . . . central to his plea for\\nleniency.” Id. The record belies that characterization.\\n\\n A\\n\\n Unlike McKinney and Coleman, Spreitz’s history of\\nsubstance abuse unrelated to the murder was not “central” to\\nhis plea for leniency or at the very core of his plea for life.\\nInstead, Spreitz argued at sentencing that he deserved\\nleniency because of, among other things, (1) his intoxication,\\nage, and emotional immaturity at the time of the crime;\\n(2) his disruptive childhood; (3) his abusive home\\nenvironment; (4) his emotionally cold mother; (5) his poor\\nsocial adjustment; (6) a persistent pattern of rejection; (7) the\\nabsence of a healthy role model; (8) a lack of economic and\\nemotional support; (9) his lack of a high school education;\\n(10) the absence of adult convictions and violent behavior;\\n(11) his low risk of future danger; (12) his good character;\\n(13) his good behavior while in jail; and (14) his\\nrehabilitative nature. See, e.g., Spreitz, 945 P.2d at 1279\\n(“At sentencing, defendant argued as statutory and\\nnonstatutory mitigating factors: (1) his dysfunctional family\\nlife and lack of socialization; (2) a history of alcohol and\\ndrug abuse; (3) his expressions of remorse; (4) his impaired\\ncapacity to appreciate the wrongfulness of his conduct,\\nA.R.S. § 13-703(G)(1); (5) his good behavior while\\nincarcerated; (6) his lack of adult convictions; (7) no prior\\n\\n60 SPREITZ V. RYAN\\n\\nrecord of violent tendencies; and (8) his age at the time of\\nthe murder, A.R.S. § 13-703(G)(5).”). Specifically,\\nSpreitz’s sentencing memorandum posited that:\\n\\n The Defendant’s background and upbringing\\n are ripe with numerous mitigating factors:\\n\\n 1. He faced a disruptive middle childhood.\\n\\n 2. Punitive and abusive behavior toward\\n [Spreitz].\\n\\n 3. An emotionally cold mother.\\n\\n 4. Poor social adjustment with peers.\\n\\n 5. No healthy role model.\\n\\n 6. Drug and alcohol abuse.\\n\\n 7. [That] Spreitz was forced to live on his\\n own without economic or emotional support.\\n\\n 8. Persistent pattern of rejection.\\n\\n In addition, Spreitz devoted a large portion of his\\nsentencing memorandum to explaining how his childhood\\nabuse, dysfunctional upbringing, developmental immaturity,\\nand emotional problems contributed to his commission of\\nthe murder:\\n\\n SPREITZ V. RYAN 61\\n\\nFor Chris, as with other chronically rejected,\\nneglected and devalued children, there was\\nno escaping [his] deep seated anger and\\nresentment. The desperate, quiet, battle\\nwrestling inside of Chris as he sought to\\nplease [his mother] and was rejected could\\nnot be numbed forever by alcohol. His\\ndevelopmental failure included the non-\\ndevelopment of personal insight and conflict\\nresolution skills. He could not control,\\nunderstand or deal with the anger. Numbing\\nby use of alcohol and trysts with older women\\nwere but stop-gap measures. His fragile\\nfacade was ready to crack. Each failure\\nbrought Chris closer to the edge. He crossed\\nthat point with the victim. The result was\\ninevitable [as] Chris had no way to\\ncompensate for a lifetime of misdirected\\ndevelopment, through a dysfunctional life\\nand family, especially when combined with\\narrested and distorted development and\\nalcoholism . . . .\\n\\nChris’s background is a relevant and\\npowerful mitigating factor with a\\ndeterminative factor culminating in the\\ninstant offense. When a person’s background\\nhas an impact [on] his behavior beyond his\\ncontrol, as here, such a circumstance is\\nclearly mitigation entitled to great\\n[deference]. Add to that the family and\\nabusive environment and [this] mitigation\\nclearly outweighs any possible claim of\\naggravation . . . . The situational stress build\\nup, including rejection by his mother, his\\n\\n62 SPREITZ V. RYAN\\n\\n girlfriend, and again that night, coupled with\\n poorly developed coping skills and alcohol\\n certainly mitigates the act.\\n\\n(Emphasis added.) 3\\n\\n But the majority nonetheless wishes to recast the defense\\ntheories and reweigh the facts, as if we were conducting a\\ndirect review. The majority singles out Dr. Flynn’s\\nobservations that (1) Spreitz began drinking alcohol and\\nsmoking marijuana at age twelve or thirteen, (2) Spreitz\\ndrank steadily by age fifteen, (3) Spreitz’s “alcohol abuse\\ncontinued to intensify after he left home,” (4) Spreitz’s\\nalcoholism reached the level of physical dependence, and\\n(5) a “variety of persons” described Spreitz as a heavy\\ndrinker, as intoxicated most of the time, and as having\\nblackouts while drinking alcohol. Those observations,\\nhowever, do not make Spreitz’s evidence of his longstanding\\nalcoholism, with no connection to the crime, central to his\\nplea for leniency.\\n\\n Instead, Spreitz presented his long-term substance abuse\\nprimarily to establish the statutory mitigating factors under\\n§ 13-703(G)(1) and § 13-703(G)(5). As to § 13-703(G)(1),\\nDr. Flynn repeatedly emphasized that Spreitz’s lifetime of\\nsubstance abuse contributed to his murder of Ms. Reid.\\nNotably, Dr. Flynn’s report opined that:\\n\\n The research on violent behavior shows a\\n strong correlation with alcohol intoxication.\\n 3\\n Likewise, Spreitz claims in this appeal that his lawyer was\\nineffective at sentencing by failing to present witnesses who had\\nextensive evidence about his early development, family history, head\\ninjuries, substance abuse, and emotional problems. We have rejected\\nthat claim in a contemporaneously filed memorandum disposition.\\n\\n SPREITZ V. RYAN 63\\n\\nGiven [Spreitz’s] history, I can see a\\nperceived rejection by two women in the\\nsame evening as especially disturbing to him,\\nparticularly when intoxicated, to the point at\\nwhich he lost aggressive control, disinhibited\\nby the alcohol intoxication . . . .\\n\\nMy best guess is that he was drunk, hurt and\\nangry at the unwillingness of his girlfriend to\\nlet him into her apartment, startled into an\\nalcoholic rage at the angry confrontation by\\nthe victim, and ended up venting the years of\\nstored up rage at her.\\n\\nAnd although several factors that normally\\npredict violence were not present, there is\\nalcoholism—one of the strongest correlates\\nof violent behavior. And, there is the\\nemotional, sexual maladjustment that comes\\nfrom his lifetime of misdirected development\\nin a dysfunctional family, combined with\\narrested and/or distorted adult development\\nbecause of the alcoholism . . . .\\n\\n1. Given the history of alcoholism . . . a\\nsignificant but unknown degree of alcohol\\nintoxication is likely. Alcohol intoxication\\nhas a well-documented disinhibiting effect\\nwhich frequently includes losses of control of\\nangry emotions and aggressive behavior.\\n\\n2. [Spreitz’s] history strongly suggests years\\nof early experiences likely to have caused a\\nbuild-up of pent-up angry, aggressive\\nfeelings toward women generally (and older\\n\\n64 SPREITZ V. RYAN\\n\\n women especially) which may have burst\\n forth with uncontrollable intensity with or\\n without alcohol intoxication. Only trivial\\n provocation is required for this type of\\n aggression explosion, termed an Intermittent\\n Explosive Disorder by DSM-IV.\\n\\n 3. Still more likely is that a combination of\\n 1. and 2. above contributed to an\\n uncontrollable outburst of aggression.\\n\\n(Emphasis added.)\\n\\n Dr. Flynn also gave extensive testimony on how\\nSpreitz’s substance abuse contributed to the impairment of\\nhis mental state “in terms of conforming his behavior to his\\nunderstanding of the law and right and wrong generally.” At\\nthe aggravation-mitigation hearing, Dr. Flynn went to great\\nlengths to opine that Spreitz’s alcohol abuse was “very\\nstrongly correlated with violent behavior.” Dr. Flynn also\\ntestified that Spreitz’s parents’ failure to treat his alcoholism\\nduring his teenage years was significant because it increased\\nhis violence potential: “[A]lcohol is probably the most\\ncommon, strongest correlative . . . with violent behavior\\n. . . , doubling the risk of violent behavior and I think the\\nabsence of successful treatment for alcoholism is likely to\\nhave been a significant and maybe a sine qua non that went\\non in this offense.” (Emphasis added.)\\n\\n In light of the evidence presented by Dr. Flynn, it is hard\\nto say that the Arizona courts did not consider the effect of\\nboth his intoxication and his history of substance abuse as\\nrequired under § 13-703(G)(1). They just weren’t persuaded\\nby it. Rather, they reasonably and properly found that\\nSpreitz’s history of alcohol abuse was not sufficiently\\n\\n SPREITZ V. RYAN 65\\n\\ninfluential on his ability to conform his conduct to the law or\\nhis ability to appreciate the consequences at the time he\\nviciously murdered Ms. Reid. Contrary to the majority’s\\nsuggestion, Op. at 28–29, it cannot be said that the Arizona\\ncourts never considered his longstanding alcoholism under\\n§ 13-703(G)(1) in the first place.\\n\\n As to the statutory mitigating factor under\\n§ 13-703(G)(5), Dr. Flynn explained how Spreitz’s history\\nof substance abuse contributed to his emotional immaturity:\\n“When you spend most of your teenage years drunk or\\nstoned, it has not only been my observation in criminal\\npopulations but my observations clinically . . . that virtually\\nall heavily drugged teenagers [are] socially and emotionally\\nimmature as having substance abuse problems.” (Emphasis\\nadded.) Dr. Flynn also testified that the lack of a treatment\\neffort in Spreitz’s home to deal with his alcoholism was\\nsignificant because it showed a lack of proper supervision\\nand parenting during his developing years, which\\ncontributed to his emotional immaturity under\\n§ 13-703(G)(5):\\n\\n [The lack of any treatment effort for Spreitz’s\\n alcoholism] is consistent with an uncaring\\n parent to have a child which is stealing booze\\n out of the house from the early teenage years,\\n who shows up drunk in his early teenage life\\n and everybody in the person’s life recognizes\\n there’s a serious alcoholic problem and yet\\n the parent either does one of two things,\\n doesn’t care enough to notice or doesn’t care\\n enough to do anything about it. That’s\\n important in and of itself in my opinion from\\n the standpoint of an evaluation of parenting.\\n\\n66 SPREITZ V. RYAN\\n\\n Thus, where Spreitz did not present his long-term\\nsubstance abuse to explain his conduct at the time of the\\nmurder under § 13-703(G)(1), he did so to plead for leniency\\nbased on his emotional immaturity under § 13-703(G)(5).\\nAny evidence that Spreitz presented of his long-term\\nsubstance abuse as non-statutory mitigation was cursory and\\nonly ancillary to the considerable evidence he adduced to\\nexplain how his past addiction affected his conduct and\\nimmaturity at the time of the murder.\\n\\n Moreover, the significant evidence presented regarding\\nSpreitz’s abusive and dysfunctional childhood, upbringing,\\nand interactions with his mother stands as further support\\nthat Spreitz’s history of alcohol and substance abuse was not\\ncentral to his plea for leniency. Dr. Flynn testified to those\\nmatters at length:\\n\\n [Spreitz grew up in] a fully pathological\\n home environment, [which] include[d]\\n physical and emotional neglect, physical\\n abuse and emotional neglect and possibly\\n emotional abuse by the mother according to\\n which there was no hint of acceptance of him\\n in spite of ongoing efforts on his part to\\n please her. In fact of all the people that I\\n interviewed and all the interviews that I read,\\n the only favorable, the only person who had\\n a single favorable thing to say about the\\n mother was Mr. Spreitz himself. From all\\n other persons they described her in the most\\n negative terms. His sister described herself\\n as being saved from the pathological effects\\n of the home by the fact she ran away and was\\n placed in a group home. Mr. and Mrs. Spreitz\\n described most of the mother’s behavior in a\\n\\n SPREITZ V. RYAN 67\\n\\n manner that would be considered pathogenic\\n by anyone who knew anything about\\n psychology and parenting . . . .\\n\\n [There is] massive evidence from a\\n psychological point . . . that there was a cold\\n uncaring mother who was anything but\\n sensitive to his needs, whom he was never\\n able to please. By this hypothesis it is most\\n common for children so treated to harbor a\\n store of anger, and it is consistent with this\\n train of thought that the anger was breaking\\n down partially by chronicity and partially by\\n series of emotionally unsuccessful\\n relationships with women, a tendency to\\n orient toward older women in the months\\n leading up to the offense[.] [S]hortly before\\n the offense . . . there had been a rejection by\\n a current girlfriend . . . . [T]his by itself and\\n probably especially this under the\\n disinhibiting influence of alcohol possibly\\n result[ed] in an explosion of aggressive\\n behavior especially toward a woman and\\n especially toward an older woman. The\\n recently pent-up rage accumulated over the\\n year aggravated the more recent months\\n which may be considered by the Court to\\n constitute an outburst of behavior that means\\n an inability to conform one’s behavior [to the\\n requirements of the law].\\n\\n(Emphasis added.)\\n\\n And, when Judge Sherrill asked Dr. Flynn what he\\nconsidered were non-statutory mitigating factors in this\\n\\n68 SPREITZ V. RYAN\\n\\ncase, Dr. Flynn responded: “[T]he likely relationship\\nbetween the alcoholism and the violence generally and this\\noffense specifically and . . . the deprived pathogenic home\\nenvironment [are what] I would consider non-statutory\\nfactors.” (Emphasis added.) In the face of this record, I am\\nat a loss to understand how my colleagues can say that\\nSpreitz’s history of substance abuse as a non-statutory\\nmitigating factor was central to or at the very core of\\nSpreitz’s plea for leniency. 4\\n\\n B\\n\\n Two other cases where we have found Arizona courts’\\nEddings errors to be harmless, Henry v. Ryan, 720 F.3d 1073\\n(9th Cir. 2013), and Stokley v. Ryan, 705 F.3d 401 (9th Cir.\\n2012), are instructive here.\\n\\n 1\\n\\n In Henry, the Arizona courts found a statutory mitigating\\nfactor under § 13-703(G)(1)—namely, that “Henry’s\\ncapacity to appreciate the wrongfulness of his conduct or to\\nconform his conduct to the requirements of law was\\nsignificantly impaired . . . because of his consumption of\\nalcohol on the day of the murder.” 720 F.3d at 1089–90.\\nNonetheless, the courts imposed the death penalty because\\nthey “agreed that the mitigation was not sufficiently\\n\\n\\n\\n 4\\n The majority does not take issue with the fact that the Arizona\\ncourts considered Spreitz’s substance abuse history in the statutory\\nmitigating context or that they found it insufficient to outweigh the\\naggravating factor of cruelty. “When applied solely in the context of\\nstatutory mitigation,” the “causal nexus test does not violate Eddings.”\\nMcKinney, 813 F.3d at 810.\\n\\n SPREITZ V. RYAN 69\\n\\nsubstantial to call for leniency” in light of two aggravating\\nfactors. Id. at 1090.\\n\\n As with Spreitz here, on habeas review, we assumed that\\nthe state courts violated Eddings by failing to consider\\nHenry’s history of alcohol abuse as a non-statutory\\nmitigating factor. However, we denied habeas relief because\\nHenry failed to show that the Eddings error had a\\n“substantial and injurious effect or influence” on his\\nsentence. Id. at 1087–89 (quoting Brecht, 507 U.S. at 623).\\nWe reasoned that, “in imposing the death penalty, the\\nArizona courts already considered Henry’s intoxication at\\nthe time of the murder as a mitigating factor” and concluded\\nthat it was “not sufficient to call for leniency.” Id. at 1090.\\nAnd:\\n\\n Given the similar nature of the mitigation, the\\n additional evidence of Henry’s historical\\n alcoholism would have had minimal\\n mitigating value. If the state courts\\n concluded that intoxication with a causal\\n connection to the crime was not sufficient to\\n call for leniency, it is highly doubtful that\\n they would have considered alcoholism\\n without a causal connection to be sufficient.\\n\\nId. 5\\n\\n Similarly, here, the Arizona courts agreed that Spreitz’s\\nsubstance abuse at the time of the offense was not sufficient\\nto call for leniency. In addition, they found that Spreitz\\ncould not show statutory mitigation under § 13-703(G)(1)\\n\\n 5\\n Although Henry involved different facts, as the majority points out,\\nits reasoning applies fully to Spreitz’s case. Op. at 36, n.15.\\n\\n70 SPREITZ V. RYAN\\n\\nbecause his “ability to appreciate the wrongfulness of his\\nconduct was not impaired on the night of the murder to any\\nsignificant extent by substance abuse.” Spreitz, 945 P.2d at\\n1281. The majority does not take issue with these findings.\\nAs in Henry, the Arizona courts’ consideration of Spreitz’s\\nlong-term substance abuse unrelated to the offense would\\nhave had minimal mitigating value because of its similarity\\nto the evidence already rejected. Henry, 720 F.3d at 1090;\\nsee also Landrigan, 272 F.3d at 1230 & n.9 (holding that,\\nwhere the sentencing court rejected the petitioner’s “alleged\\nintoxication and past history of drug use as a statutory\\nmitigating factor” and “did not go on to consider them as a\\nnonstatutory mitigating factor,” any “error in failing to\\nconsider Landrigan’s use of alcohol and drugs would have\\nbeen inconsequential; it would have had no effect\\nwhatsoever on the outcome”); Wong v. Belmontes, 558 U.S.\\n15, 23 (2009) (“The sentencing jury was thus ‘well\\nacquainted’ with Belmontes’ background and potential\\nhumanizing features. Additional evidence on these points\\nwould have offered an insignificant benefit, if any at all.”\\n(citation omitted)).\\n\\n The majority thus errs by failing to apply Henry’s\\nreasoning to conclude that any Eddings error was harmless\\nhere. See Hart v. Massanari, 266 F.3d 1155, 1172 (9th Cir.\\n2001) (“[I]f a controlling precedent is determined to be on\\npoint, it must be followed.”). If direct evidence of\\nsubstance-induced intoxication on the night of the murder\\nwas insufficient to outweigh the aggravated circumstances\\nof the crime, a long history of drug and alcohol abuse would\\nfare no better.\\n\\n 2\\n\\n In Stokley, we assumed that the Arizona Supreme Court\\ncommitted Eddings error by failing to consider Stokley’s\\n\\n SPREITZ V. RYAN 71\\n\\nabusive childhood and good behavior in jail because of a\\nlack of causal nexus. 705 F.3d at 403–04. We held,\\nhowever, that this error was harmless because there were\\nthree aggravating circumstances, including that Stokley—\\nlike Spreitz—had “committed the offense in an especially\\nheinous, cruel, and depraved manner.” Id. at 404. We noted\\nthe gruesome details of Stokley’s offense, including (1) that\\nhe had sexual intercourse with his victims; (2) that the\\nvictims’ bodies (two young girls) were violently beaten; and\\n(3) that the victims’ bodies were dragged to and thrown\\ndown a mineshaft. Id. at 405 n.1. We also observed that\\nboth the sentencing court and the Arizona Supreme Court\\nconsistently determined that “leniency was inappropriate.”\\nId. at 405. We found no likelihood “that, but for a failure to\\nfully consider Stokley’s family history or his good behavior\\nin jail during pre-trial incarceration, the Arizona courts\\nwould have come to a different conclusion” as to Stokley’s\\nsentence. Id.\\n\\n Although Stokley’s facts were relatively more egregious\\nthan the facts here—Stokley involved three aggravating\\ncircumstances instead of one and Stokley’s victims were\\nunder the age of fifteen—Stokley nonetheless informs us that\\nthe severity of an aggravating circumstance must be\\nconsidered when determining whether an Eddings error is\\nharmless. Here, both the sentencing court and the Arizona\\nSupreme Court consistently ruled that leniency was\\ninappropriate because of the especially cruel manner in\\nwhich Spreitz kidnapped and murdered Ruby Reid. Spreitz,\\n945 P.2d at 1266, 1278–79 (citing Ariz. Rev. Stat. § 13-\\n703(F)(6)). The crime scene photographs alone give\\ncompelling support to that ruling, showing Ms. Reid’s body\\nbeaten beyond recognition by heavy rocks. The pathologist\\nunderscored the extensive damage to her body revealed on\\nautopsy.\\n\\n72 SPREITZ V. RYAN\\n\\n Arizona courts consider five factors in “determining\\nwhether a murder was especially heinous, cruel or depraved:\\n(1) relishing the murder, (2) infliction of gratuitous violence,\\n(3) needless mutilation of the victim, (4) senselessness of the\\ncrime, and (5) helplessness of the victim.” Gulbrandson v.\\nRyan, 738 F.3d 976, 984 (9th Cir. 2013) (citing State v.\\nGretzler, 659 P.2d 1, 11–12 (Ariz. 1983) (en banc)). “A\\nfinding of cruelty is warranted when the defendant inflicts\\non the victim mental anguish or physical abuse before the\\nvictim’s death.” Spreitz, 945 P.2d at 1278. “Cruelty is found\\nwhen the victim is conscious at the time of the offense in\\norder to suffer pain and distress.” Id. (citation and alterations\\nomitted).\\n\\n Here, the Arizona courts found that Ms. Reid suffered\\ntremendous mental and physical pain when she was forced\\ninto Spreitz’s car trunk and transported to the desert where\\nshe was beaten, sexually assaulted, and eventually murdered.\\nId. at 1265, 1278. Judge Sherrill observed: “At the scene of\\nthe murder the victim was thoroughly beaten. The autopsy\\nshowed, in addition to the death-causing blows to the head,\\nfive ribs broken, eighteen bruises and grab marks, internal\\nbruising and bleeding, lacerations[,] and a broken jaw.”\\nSpreitz did not confess to kidnapping Ms. Reid, but did\\nadmit to beating her as she fought back, removing her\\nclothes, having intercourse with her, and smashing her in the\\nhead with a rock when she would not stop yelling. Id.\\nPhysical evidence found at the scene of the murder\\ncorroborated Spreitz’s confession: Ms. Reid’s clothing,\\nincluding her torn brassiere, was strewn in one area of the\\nscene, and photographs of the area vividly depicted drag\\nmarks running from that area to the spot where Ms. Reid’s\\nbody and the bloody rocks were found. According to\\nSpreitz’s confession, when he left Ms. Reid lying in the\\ndesert on the morning of May 19, 1989, he did not know\\n\\n SPREITZ V. RYAN 73\\n\\nwhether she was still alive or dead. Also significant to Judge\\nSherrill’s finding of mental anguish was that Ms. Reid had\\ndefecated in and on her clothing. See id. at 1265, 1278–79.\\n\\n In its independent review of these facts, the Arizona\\nSupreme Court agreed that Spreitz had murdered Ms. Reid\\nin an especially cruel manner. Id. at 1278. The Arizona\\nSupreme Court highlighted Spreitz’s “own admission that he\\nbeat her as she fought back and hit her with the rock when\\nshe would not stop yelling,” which it found was “clear\\nevidence of [Ms. Reid’s] conscious suffering.” Id. at 1279.\\nThe court found that Spreitz’s confession and the physical\\nevidence all pointed to the conclusion that Spreitz “beat and\\nraped [Ms. Reid] in a brutal assault that lasted many minutes\\nbefore he crushed her skull.” Id. My colleagues in the\\nmajority do not quarrel with that finding.\\n\\n Thus, even assuming that the Arizona courts failed to\\nconsider Spreitz’s evidence of long-term substance abuse\\nunrelated to the crime, it is unlikely their consideration of\\nthat evidence would have substantially altered their ultimate\\ndecision to sentence Spreitz to death based on the aggravated\\nmanner in which Ms. Reid met her death. And we cannot\\nsay “with fair assurance” that the sentence they imposed was\\nsubstantially swayed by their failure to consider that\\nevidence. McKinney, 813 F.3d at 822 (quoting Kotteakos,\\n328 U.S. at 765).\\n\\n Further, Eddings allows the Arizona courts “to consider\\nthe absence of a causal connection when assessing the\\nquality and strength of [mitigating] evidence.” Henry,\\n720 F.3d at 1090. It does not prevent the Arizona courts\\nfrom “assign[ing] less weight to mitigating factors that did\\nnot influence a defendant’s conduct at the time of the crime.”\\nHedlund, 854 F.3d at 587 n.23. The Arizona Supreme Court\\nhas recognized this principle in its more recent cases. See,\\n\\n74 SPREITZ V. RYAN\\n\\ne.g., State v. Newell, 132 P.3d 833, 849 (Ariz. 2006) (en\\nbanc) (“We do not require that a nexus between the\\nmitigating factors and the crime be established before we\\nconsider the mitigation evidence. But the failure to establish\\nsuch a causal connection may be considered in assessing the\\nquality and strength of the mitigation evidence.” (citation\\nomitted)). Thus, we can confidently say that if Spreitz were\\nresentenced, the Arizona courts would give even less weight\\nto his history of substance abuse unrelated to the crimes in\\nbalancing that evidence against what he did to Ruby Reid.\\n\\n C\\n\\n The majority cites Henry for the broad proposition that\\n“a ‘history of substance abuse [is] substantially mitigating\\nwhen . . . combined with other mitigating evidence.’” Op.\\nat 36 (emphasis omitted) (quoting Henry, 720 F.3d at 1090).\\nBut the majority has not offered up any authority that reaches\\nthis conclusion under factually-similar circumstances.\\nIndeed, unlike the three cases Henry relies upon in making\\nthis statement, Spreitz’s defense counsel presented\\nconsiderable evidence at sentencing of his lifelong history of\\nsubstance abuse, his mental and emotional problems, and his\\nfamily dysfunction. Counsel retained Dr. Flynn, a\\npsychotherapist and forensic consultant, who personally\\nevaluated Spreitz, performed psychological assessments,\\nand interviewed people who were familiar with him. Dr.\\nFlynn submitted a report and testified extensively regarding\\nSpreitz’s alcohol and drug addictions, his family history of\\nsubstance abuse, his disruptive childhood, and his emotional\\nproblems. Thus, the majority’s suggestion that Spreitz’s\\ncase is comparable to those in which we have found evidence\\nof longstanding substance abuse to be substantially\\nmitigating is misplaced. This is particularly so in light of the\\nhorrific way in which Ruby Reid died, the impact of which\\n\\n SPREITZ V. RYAN 75\\n\\nwas not lost on all of the Arizona judges who reviewed this\\nrecord.\\n\\n D\\n\\n Lastly, Spreitz’s own arguments on appeal belie the\\nconclusion that his long-term substance abuse unrelated to\\nthe crime was central to his plea for leniency. Spreitz argues\\nthat whether he “was intoxicated at the time of the murder\\nwas critical” to his case, and that “the issue of [his]\\nintoxication [at the time of the crime] was critically\\nimportant.” He contends that he “was highly intoxicated at\\nthe time of the murder” and that “this was the primary cause\\nof him committing the murder.” And, he urges that there\\n“was ample evidence that [he] was highly intoxicated at the\\ntime of the murder and that his intoxication contributed\\ndirectly to his actions.” “This was highly relevant\\nmitigation,” he says: “[h]ad the judge heard the evidence\\ndescribed herein, it is highly probable that he would not have\\nsentenced [Spreitz] to death.” But the judge did hear all of\\nthat evidence. He just didn’t buy the argument based on the\\ntestimony of officers who interacted with Spreitz before and\\nafter he killed Ms. Reid. That credibility determination is\\nnot assailable on appeal.\\n\\n Spreitz’s defense theories have shifted on collateral\\nappeal. Contrary to his position in supplemental briefing\\nordered by this Court post-McKinney, Spreitz does not make\\nany similar arguments about his long-term substance abuse\\nas a non-statutory mitigating factor. Instead, he claims that\\nevidence of his childhood head injuries that his counsel\\nfailed to present were vital to his mitigation evidence.\\nSpreitz argues that the “importance of counsels’ failure to\\ninvestigate [his] history of head injuries cannot be\\noverstated;” that the “connection between childhood head\\ninjuries and impulsive murder” was “well established” at the\\n\\n76 SPREITZ V. RYAN\\n\\ntime of sentencing; that “there was a correlation between\\n[his] head injuries as a child and his impulsive behavior in\\nthe instant case;” and that “[expert] testimony could have\\nenlightened the court as to the connection between childhood\\nhead injuries and impulsive behavior in murderers in\\ngeneral, and could have developed the link between\\n[Spreitz]’s own head injuries and his impulsive behavior\\nduring this murder in particular.” He contends that:\\n\\n The murder in this case was a classic\\n impulsive act. It was not thought out or\\n premeditated beforehand. Appellant beat the\\n victim with a rock; he did not even bring a\\n weapon. The victim refused to have sex with\\n him, fought him, and he became enraged and\\n impulsively killed her. That is the classic\\n behavior of a murderer who suffered\\n childhood head injuries.\\n\\n(Emphasis added.) Spreitz’s shifting defense theories\\nundermine his latest argument that long-term substance\\nabuse was central to his plea for leniency all along.\\n\\n We will never know how Ruby Reid, forced into the\\ntrunk of Spreitz’s car and pondering her fate as he drove her\\ninto the desert, would have answered any of Spreitz’s\\ncontentions. But the forensic evidence of the way in which\\nshe died stands in silent, but compelling refutation of that\\ndefense. Simply put, Spreitz cannot show on this record that\\nhis long-term substance abuse, as a non-statutory mitigating\\nfactor and without any connection to the crime, was either\\n“central to his plea for leniency,” McKinney, 813 F.3d at\\n823, or at “the very core of [his] plea for life,” Coleman,\\n210 F.3d at 1051. Therefore, he cannot demonstrate that the\\nArizona courts’ alleged failure to consider that evidence\\n\\n SPREITZ V. RYAN 77\\n\\nsubstantially influenced or swayed his ultimate sentence.\\nNor can he show that, had the Arizona courts further\\nconsidered its marginal mitigating weight, the evidence\\nwould have substantially affected the outcome of his\\nsentence. See Brecht, 507 U.S. at 623. Because Spreitz\\ncannot show that the error he alleges was not harmless, the\\ndistrict court’s denial of habeas relief on Spreitz’s Eddings\\nclaim should be affirmed.\\n\\n From the majority’s conclusion otherwise, I respectfully\\ndissent.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4373176/\", \"author_raw\": \"TALLMAN, Circuit Judge, dissenting\"}]}"},"judge1":{"kind":"string","value":"RICHARD A PAEZ"},"judge2":{"kind":"string","value":"MARSHA S BERZON"},"judge3":{"kind":"string","value":"RICHARD C TALLMAN"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":2,"string":"2"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":1,"string":"1"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4595923/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Christopher Spreitz v. Charles Ryan\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-04\nOpinion type: majority\nHas dissenting opinion: Yes\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nCHRISTOPHER J. SPREITZ, No. 09-99006\n Petitioner-Appellant,\n D.C. No.\n v. 4:02-CV-00121-JMR\n\nCHARLES L. RYAN, Warden,\n Respondent-Appellee. OPINION\n\n\n\n Appeal from the United States District Court\n for the District of Arizona\n John M. Roll, District Judge, Presiding\n\n Argued July 11, 2013\n Submitted March 4, 2019\n San Francisco, California\n\n Filed March 4, 2019\n\n Before: Richard A. Paez, Marsha S. Berzon,\n and Richard C. Tallman, Circuit Judges.\n\n Opinion by Judge Paez;\n Dissent by Judge Tallman\n\n2 SPREITZ V. RYAN\n\n SUMMARY *\n\n\n Habeas Corpus / Death Penalty\n\n The panel reversed the district court’s denial of habeas\ncorpus relief with respect to Christopher J. Spreitz’s death\nsentence, and remanded, in a case in which Spreitz argued\nthat the Arizona Supreme Court violated Eddings v.\nOklahoma, 455 U.S. 104 (1982), by refusing to consider, as\na matter of law, mitigating evidence of Spreitz’s\nlongstanding alcohol and substance abuse on the ground that\nhe did not establish a causal connection between this\nmitigating evidence and the crime.\n\n The panel held that the district court erred in concluding\nthat Spreitz’s claim that the Arizona Supreme Court violated\nEddings is procedurally defaulted. The panel explained that\nthe first opportunity Spreitz had to raise that claim was\nbefore the post-conviction-relief (PCR) court, at which time\nhe did so.\n\n Because the decision of the PCR court – which first\ndeclared the claim waived, but proceeded to adjudicate the\nclaim on the merits – was contrary to clearly established\nSupreme Court precedent, the panel accorded that decision\nno deference and reviewed Spreitz’s Eddings claim de novo.\nThe panel concluded that the Arizona Supreme Court\nviolated Eddings by impermissibly requiring that Spreitz\nestablish a causal connection between his longstanding\nsubstance abuse and the murder before considering and\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\n SPREITZ V. RYAN 3\n\nweighing the evidence as a nonstatutory mitigating factor.\nThe panel concluded that the error was not harmless.\n\n The panel affirmed the district court’s judgment denying\nrelief with respect to Spreitz’s conviction in a concurrently\nfiled memorandum disposition.\n\n Dissenting, Judge Tallman wrote that the record does not\nestablish that either the sentencing court or the Arizona\nSupreme Court unconstitutionally refused to consider\nrelevant mitigating evidence; and that even if the Arizona\ncourts did violate Eddings, Spreitz cannot show that this\nerror had a “substantial and injurious effect or influence” on\nhis ultimate sentence.\n\n\n COUNSEL\n\nTimothy M. Gabrielsen (argued), Assistant Federal Public\nDefender; John M. Sands, Federal Public Defender; Office\nof the Federal Public Defender, Tucson, Arizona; Susan B.\nFox and Sean Bruner, Law Office of Sean Bruner Ltd.,\nTucson, Arizona; for Petitioner-Appellant.\n\nLacey Stover Gard (argued) and Jeffrey A. Zick, Section\nChief Counsel; Kent E. Cattani, Chief Counsel, Criminal\nAppeals/Capital Litigation Section; Mark Brnovich,\nAttorney General; Office of the Attorney General, Tucson,\nArizona; for Respondents-Appellees.\n\n4 SPREITZ V. RYAN\n\n OPINION\n\nPAEZ, Circuit Judge:\n\n In 1994, an Arizona jury convicted Christopher J. Spreitz\n(“Spreitz”) of first-degree murder. The victim was thirty-\nnine year old Ruby Reid (“Reid”). Finding that the cruelty\nof the murder outweighed any mitigating circumstances, the\ntrial judge sentenced Spreitz to death. Spreitz appeals the\ndistrict court’s denial of his petition for a writ of habeas\ncorpus challenging his conviction and sentence. We affirm\nthe district court’s judgment with respect to Spreitz’s\nconviction, 1 and reverse with respect to his sentence. 2\n\n In challenging his sentence, Spreitz argues that the\nArizona Supreme Court unconstitutionally affirmed his\ndeath sentence by failing to consider mitigating evidence of\nhis longstanding alcohol and substance abuse. He contends\nthat the state court refused to consider, as a matter of law,\nthis evidence in mitigation because he did not establish a\ncausal connection between the crime and his long-term\nalcohol and substance abuse. In Eddings v. Oklahoma,\n455 U.S. 104, 110 (1982), the Supreme Court held that under\nboth the Eighth and Fourteenth Amendments, a sentencer in\na capital case may not “refuse to consider, as a matter of law,\n\n 1\n We affirm the judgment with respect to Spreitz’s conviction in a\nconcurrently filed memorandum disposition.\n\n 2\n On February 13, 2015, we vacated submission of Spreitz’s case\npending final resolution of the en banc proceedings in McKinney v. Ryan,\n730 F.3d 903 (9th Cir. 2013). In December 2015, the en banc court\nissued an opinion in McKinney. 813 F.3d 798 (9th Circ. 2015) (en banc).\nAfter the Supreme Court denied the State’s petition for a writ of\ncertiorari, Ryan v. McKinney, 137 S. Ct. 39 (2016) (mem), we ordered\nsupplemental briefing on the applicability of McKinney to Spreitz’s case.\n\n SPREITZ V. RYAN 5\n\nany relevant mitigating evidence” offered by the defendant.\nId. at 114. Although a sentencer “may determine the weight\nto be given relevant mitigating evidence . . . they may not\ngive it no weight by excluding such evidence from their\nconsideration.” Id. at 114–15 (footnote omitted). In\ninterpreting and applying Eddings, the Supreme Court has\nexplained that “full consideration of evidence that mitigates\nagainst the death penalty is essential if the [sentencer] is to\ngive a reasoned moral response to the defendant’s\nbackground, character, and crime.” Penry v. Lynaugh\n(Penry I), 492 U.S. 302, 328 (1989), abrogated on other\ngrounds by Atkins v. Virginia, 536 U.S. 304 (2002) (internal\nquotation marks and citation omitted). Moreover, the\nSupreme Court has been clear: requiring a defendant to\nprove a causal nexus between his mitigating evidence and\nthe crime is “a test we never countenanced and now have\nunequivocally rejected.” Smith v. Texas, 543 U.S. 37, 45\n(2004) (per curiam).\n\n At the time of Spreitz’s sentencing, Arizona Revised\nStatute Annotated § 13-703(G)(1994) 3 listed five mitigating\nfactors, and Arizona case law additionally recognized\nnonstatutory mitigating factors, including, for example, a\ndefendant’s difficult family background or mental condition\nnot severe enough to qualify as a statutory mitigating factor.\nIn an en banc decision of our court, McKinney v. Ryan,\n813 F.3d 798 (9th Cir. 2015), cert denied, 137 S. Ct. 39\n(2016) (mem), we explained:\n\n For a period of a little over 15 years in capital\n cases, in clear violation of Eddings, the\n\n 3\n Arizona has since revised its death penalty sentencing scheme. All\nreferences to Arizona’s Revised Statute Annotated are to those\nprovisions in effect at the time of Spreitz’s sentencing.\n\n6 SPREITZ V. RYAN\n\n Supreme Court of Arizona articulated and\n applied a “causal nexus” test for nonstatutory\n mitigation that forbade as a matter of law\n giving weight to mitigating evidence, such as\n family background or mental condition,\n unless the background or mental condition\n was causally connected to the crime.\n\nId. at 802. As a result, we held in McKinney that\n“[a]pplication of the causal nexus test to nonstatutory\nmitigating factors violated Eddings, for it resulted in Arizona\ncourts being entirely forbidden, as a matter of state law, to\ntreat as a mitigating factor a family background or a mental\ncondition that was not causally connected to a defendant’s\ncrime.” Id. Spreitz argues that the Arizona Supreme Court\napplied its causal nexus test in his case, refusing to consider\nevidence of his long-term substance and alcohol abuse\nbecause he did not adequately establish a causal connection\nbetween that history of abuse and his crime.\n\n As in McKinney, “the precise question before us is\nwhether the Arizona Supreme Court applied its\nunconstitutional causal nexus test in affirming [Spreitz]’s\ndeath sentence on de novo review.” 4 Id. at 804 (emphasis\n\n\n 4\n The Arizona Supreme Court is required by statute to undertake an\nindependent review of a death sentence. Ariz. Rev. Stat. § 13-703.01(A).\nIn conducting that review for crimes that occurred before 2002, the court\n“independently review[s] the trial court’s findings of aggravation and\nmitigation and the propriety of the death sentence. In doing so, [the\ncourt] review[s] the record de novo, considering the quality and the\nstrength, not simply the number, of aggravating and mitigating factors.”\nState v. Lynch, 357 P.3d 119, 141 (Ariz. 2015) (internal quotation marks\nand citations omitted), rev’d on other grounds, Lynch v. Arizona, 136 S.\nCt. 1818 (2016).\n\n SPREITZ V. RYAN 7\n\nand internal quotation marks omitted). For the reasons that\nfollow, we conclude that it did.\n\n I.\n\n A. Spreitz’s Crimes, Conviction, and Sentence\n\n On May 25, 1989, the police arrested Spreitz after\ndiscovering Ruby Reid’s body in the desert. Upon\nquestioning, Spreitz confessed to murdering Reid. We\nbriefly provide the facts of the murder.\n\n On the evening of May 18, after drinking heavily and\nbeing rejected by the woman he was dating, Spreitz “picked\nup” Reid at a convenience store. State v. Spreitz (Spreitz I),\n945 P.2d 1260, 1264–65 (Ariz. 1997). 5 In his confession,\nSpreitz claimed that Reid voluntarily left with him and that\nhis understanding was that they would have sex later that\nevening. Id. at 1265. Spreitz further claimed that he drove\nher out to the desert, where Reid decided she no longer\nwanted to have sex. Id. The two fought as a result. Id.\nSpreitz explained that Reid slapped him and that he\nresponded by punching her in the mouth. Id. Spreitz then\nsexually assaulted Reid—“remov[ing] her clothing and\nha[ving] vaginal intercourse with her.” Id. Spreitz also\nrecounted that he hit Reid in the head multiple times with a\nrock “to make her stop yelling.” Id. He explained that he\nleft Reid without knowing whether she was alive or dead.\nId.\n\n Shortly after leaving Reid in the desert, Spreitz was\nstopped by a Tucson police department officer. Id. at 1264.\n\n 5\n Our recitation of the facts is adopted from the Arizona Supreme\nCourt’s opinion affirming Spreitz’s conviction and death sentence.\n\n8 SPREITZ V. RYAN\n\nThe officer observed that Spreitz had a ripped shirt, smelled\nof feces, and appeared to be covered in blood and fecal\nmatter. Id. In addition, when detectives later searched\nSpreitz’s car, they found blood spatter in the trunk, some of\nwhich was inconsistent with Spreitz’s blood characteristics.\nId. at 1265.\n\n On Monday morning, May 22, Reid’s naked and\ndecomposing body was discovered on the outskirts of\nTucson. Id. At trial, “the medical examiner testified that,\ndue to the advanced state of decomposition, he could not\ndetermine the full extent and nature of [Reid]’s injuries.” Id.\nEven so, he was able to observe “bruising on the legs, arms,\nand back; bruising and abrasions on the buttocks; several\nbroken ribs; internal bleeding; a broken jaw; several head\nlacerations; and a skull fracture where the skull had been\n‘shoved in.’” Id. The medical examiner concluded that Reid\nhad been killed by “blunt-force trauma to the head.” Id.\n\n In addition to finding Reid’s body “[a]t the scene, police\ndetectives observed tire tracks leading back to the pavement,\noil stains in the dirt, footprints, and drag marks in the dirt\nleading away from the body. They also found feces-stained\npants, tennis shoes, socks, a used tampon, and a torn\nbrassiere. Two blood-stained rocks lay next to the body.”\nId. A few days later, police arrested Spreitz. Id.\n\n On June 2, 1989, a grand jury indicted Spreitz for first-\ndegree murder, Ariz. Rev. Stat. Ann. §§ 13-1105, 13-703;\nsexual assault, Ariz. Rev. Stat. Ann. §§ 13-406(A) & (B);\nand kidnapping, Ariz. Rev. Stat. Ann. §§ 13-304(A)(3) &\n(B). Spreitz I, 945 P.2d at 1265. After five years of pre-trial\nproceedings mostly regarding the admissibility of DNA\nevidence, a seven-day jury trial began on August 9, 1994.\nId. at 1266. After the conclusion of the trial, the jury\nreturned guilty verdicts on all three counts: first-degree\n\n SPREITZ V. RYAN 9\n\nmurder (both premeditated and felony murder), sexual\nassault, and kidnapping. Id.\n\n Prior to both his aggravation-mitigation and sentencing\nhearings before the trial judge, Spreitz submitted evidence\nand a memorandum in support of certain mitigating\ncircumstances. As noted earlier, at the time of Spreitz’s\nsentencing, Arizona’s death penalty statutes provided a list\nof five specific mitigating factors; Arizona case law\nrecognized nonstatutory mitigating factors as well. See Ariz.\nRev. Stat. Ann. § 13-703(G); McKinney, 813 F.3d at 802.\nSpreitz argued as nonstatutory mitigating factors: “(1) his\ndysfunctional family life and lack of socialization; (2) a\nhistory of alcohol and drug abuse; (3) his expressions of\nremorse; [(4)] his good behavior while incarcerated; [(5)] his\nlack of adult convictions; [and (6)] no prior record of violent\ntendencies.” Spreitz I, 945 P.2d at 1279. Spreitz argued as\nstatutory mitigating factors: (1) his age at the time of the\nmurder, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and (2) that\nhis “capacity to appreciate the wrongfulness of his conduct\nor to conform his conduct to the requirements of law was\nsignificantly impaired [due to alcohol use], but not so\nimpaired as to constitute a defense to prosecution,” id.\n§ 13-703(G)(1); see Spreitz I, 945 P.2d at 1279.\n\n Spreitz provided evidence of and argued for all the\nforegoing mitigating circumstances but focused heavily on a\ncombination of his relationship with his mother and his long\nhistory of alcohol and substance abuse. To that end, Spreitz\nsubmitted a written report by and presented testimony from\nan examining psychologist, Dr. Todd Flynn, Ph.D. After\nconducting interviews and research, Dr. Flynn concluded\nthat Spreitz’s longstanding alcohol and substance abuse\nshould be considered as both a statutory and nonstatutory\nmitigating factor. In his report, which was admitted into\n\n10 SPREITZ V. RYAN\n\nevidence at the aggravation and mitigation hearing, Dr.\nFlynn repeatedly emphasized Spreitz’s longstanding\nsubstance abuse 6:\n\n By age twelve or thirteen, Chris Spreitz\n began drinking alcohol and smoking\n marijuana. By age 15, he drank steadily on\n weekends and would have a shot of vodka\n before school.\n\n The collateral information shows that the\n alcohol abuse continued to intensify after he\n left home. A variety of persons . . . described\n him as a heavy drinker. This includes a\n second cousin, Scott [Jouett], who saw him to\n be intoxicated, “a majority of the time,” when\n he was visiting Santa Barbara a week before\n the current offense. To the interviewing\n investigator, Mr. [Jouett] also described,\n “several different occasions when Chris has\n blackouts,” while drinking alcohol.\n\n It appears completely clear from the available\n information that Chris Spreitz had a long-\n standing problem with alcohol which\n probably reached the level of physical\n dependence. He described himself as\n drinking in the morning as early as age 15.\n Virtually everyone else who spent much time\n with him described him as a heavy drinker.\n\n\n\n 6\n Dr. Flynn testified consistently with his report. For ease of\nreference, we refer only to his report.\n\n SPREITZ V. RYAN 11\n\nUltimately, Dr. Flynn concluded that Spreitz’s alcohol\nabuse, childhood home life, and stunted development,\ncombined with rejection by his girlfriend on the day of the\ncrime, and intoxication at the time of the crime, led to the\nmurder.\n\n At the end of his report, Dr. Flynn summarized his\nfindings and opinions with respect to both statutory and\nnonstatutory mitigating factors. He opined that both\nSpreitz’s age, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and\nimpaired capacity to appreciate the wrongfulness of his\nconduct or conform his conduct to the law, Ariz. Rev. Stat.\nAnn. § 13-703(G)(1), were mitigating statutory factors. As\nto § 13-703(G)(5), specifically, Dr. Flynn opined that a\ncombination of a disturbing upbringing in a “pathogenic,\nemotionally neglectful home environment,” and “[y]ears of\nalcoholism intoxication” combined to cause major deficits in\nSpreitz’s social and emotional development. As to\n§ 13-703(G)(1), Dr. Flynn offered a similar conclusion. In\nlight of Spreitz’s “history of alcoholism . . . , a significant\nbut unknown degree of alcohol intoxication is likely” on the\nnight of the crime. In addition, Spreitz’s “history strongly\nsuggests years of early experiences likely to have caused a\nbuild-up of pent-up angry, aggressive feelings toward\nwomen generally (and older women especially) which may\nhave burst forth with uncontrollable intensity with or without\nalcohol intoxication.” Dr. Flynn concluded that Spreitz’s\nintoxication on the night of the crime coupled with his early\nchildhood experiences “likely . . . contributed to an\nuncontrollable outburst of aggression” and inability to\ncontrol his conduct.\n\n Dr. Flynn also concluded that certain nonstatutory\nmitigating factors were present. The nonstatutory mitigating\nfactors included, in his opinion: Spreitz’s low potential for\n\n12 SPREITZ V. RYAN\n\nfuture violence, the “failure of [Spreitz’s] parents to provide\ntreatment for alcohol abuse in [his] teenage years,” and the\n“emotionally deprived, physically punitive home\nenvironment” of Spreitz’s upbringing. Dr. Flynn also\nemphasized that even if both statutory mitigating factors\nfailed to satisfy the statutory threshold, they may still “be\nappropriately considered . . . as nonstatutory mitigation.”\n\n On November 28, 1994, the trial court conducted an\naggravation-mitigation hearing. At the hearing, Spreitz\ncalled three mitigation witnesses: Dr. Flynn and two\ncorrectional officers from the Pima County jail, where\nSpreitz was incarcerated. The State did not offer any\nwitnesses or evidence at the hearing. Dr. Flynn testified\nconsistently with his report, as detailed above.\n\n On the same date, a probation officer filed a Pre-\nSentence Report, which concluded:\n\n It appears the defendant became involved in\n the senseless commission of the instant\n offense due to his alcohol and drug abuse.\n After five years in custody, he now admits his\n substance abuse problem; however, this does\n not condone his involvement in the offense.\n It is unfortunate the victim died before the\n defendant had his revelation.\n\nPrior to the sentencing hearing, Spreitz also submitted\nseveral letters from friends, family, and jail personnel.\n\n On December 21, 1994, the trial court conducted a\nsentencing hearing. Spreitz, Spreitz’s counsel, the victim’s\n\n SPREITZ V. RYAN 13\n\nsister, and the prosecutor each addressed the court. 7 The\nprosecutor first disputed the mitigating evidence presented\nby Spreitz at the mitigation-aggravation hearing. With\nrespect to Spreitz’s alcohol abuse, the prosecutor argued that\nSpreitz’s intoxication at the time of the crime did not meet\nthe statutory definition for mitigation under § 13-703(G)(1)\nbecause his conduct both during the murder and afterwards\ndemonstrated that “he knew what he was doing.” The\nprosecutor emphasized that the evidence revealed that Reid\nhad been forcibly abducted and stuffed into the trunk of a\ncar, thus forcing her to spend time “contemplat[ing] the\nuncertainty of her fate.” The prosecutor argued that the\nevidence revealed signs of struggle and that the serious\ninjuries to Reid’s body belied any notion that Spreitz did not\nknow what he was doing. The prosecutor further argued that\nthe fact that Reid defecated on herself revealed that she was\nterrified prior to her murder. In sum, the prosecutor argued\nthat given both the heinous nature of the crime and the\nmanner in which Reid suffered, a finding in aggravation that\nReid was murdered in an “especially cruel manner” was\nwarranted.\n\n After a short recess, the sentencing judge rendered oral\nfindings addressing the aggravation and mitigation issues.\nHe then imposed a sentence of death. 8 Following the\n\n\n\n 7\n The trial court did not entertain argument at the conclusion of the\naggravation-mitigation hearing. As a result, counsel addressed the\naggravation-mitigation evidence and sentence options at the subsequent\nsentencing hearing.\n 8\n Ring v. Arizona, 536 U.S. 548 (2012), the Supreme Court decision\nholding judge-sentencing in capital cases unconstitutional, had yet to be\ndecided.\n\n14 SPREITZ V. RYAN\n\nhearing, the sentencing judge filed a judgment setting forth\nwritten findings.\n\n The judge first found one aggravating circumstance—\nthat the offense was committed in an “especially cruel\nmanner.” Ariz. Rev. Stat. Ann. § 13-703(F)(6). The judge\nproceeded to review the “many factors [that were] submitted\nin mitigation,” ultimately finding five mitigating\ncircumstances of various levels of significance:\n\n [1] A mitigating circumstance defense\n submitted was that the defendant had an\n extremely disruptive childhood . . . . The\n court finds the home was sub-normal, not\n even a minimally healthy one for developing\n children; and it is obvious the defendant\n suffered a disruptive middle childhood—had\n a punitive, controlling, cold mother, who he\n could not please, no matter what he did.\n\n The defendant in his life turned to substance\n abuse—alcohol and some suggestion he was\n using cocaine and other drugs. However, the\n court does not find such is a mitigating\n circumstance that impaired his ability to\n make a judgment on whether he was acting\n rightfully or wrongfully in the death of the\n victim.\n\n The defendant’s history of intoxication is\n longstanding. He had been abusing\n substances for close to ten years of his life at\n the time of this offense when he was twenty-\n two. Again, the court does not believe that\n the substance abuse or intoxication impaired\n\n SPREITZ V. RYAN 15\n\n the defendant’s ability and capacity to\n appreciate the wrongfulness of his conduct to\n any significant degree. . . . The court does not\n believe intoxication is any sort of mitigating\n circumstance.\n\n ...\n\n [2] The court acknowledges that the\n defendant has begun a process of\n improvement and emotional growth while\n confined at the Pima County Jail, where he\n has taken part in education programs.\n Correctional officers have testified he was a\n prisoner who caused no problems.\n\n [3] The age of the defendant at the time of the\n offense (twenty-two) is not a mitigating\n circumstance in and of itself. Immaturity\n probably is, but the court does not believe\n immaturity was a significant mitigating\n circumstance.\n\n [4] The court finds that Mr. Spreitz has no\n criminal history of a felony nature—there is\n no history or propensity for acts of violence.\n\n [5] The court believes the defendant is\n capable of being rehabilitated. The court\n does not know whether he has a good\n prognosis for the future, but the court\n believes he can be rehabilitated.\n\nIn conclusion, the judge found that “the mitigating\ncircumstances [were] not sufficient to balance the\n\n16 SPREITZ V. RYAN\n\naggravating circumstances, nor [were] they sufficiently\nsubstantial to call for leniency.” He thus imposed a death\nsentence as to the first-degree murder conviction.\n\n B. Spreitz’s Appeal and Post-Conviction Proceedings\n\n Spreitz appealed his conviction and sentence to the\nArizona Supreme Court with the assistance of new counsel.\nThe Arizona Supreme Court affirmed Spreitz’s conviction\nand sentence after conducting its own independent review of\nthe record. 9 In reviewing the mitigating evidence, the court\nfound four of the five mitigating circumstances that the\nsentencing judge had found, declining to find Spreitz’s good\nbehavior while in jail awaiting sentencing as a mitigating\ncircumstance. Spreitz I, 945 P.2d at 1280–81. The court\nexplained:\n\n [1] “We agree with the sentencing judge that\n defendant’s upbringing was subnormal.” Id.\n at 1280.\n\n [2] “We also find that the sentencing judge\n . . . properly found that his emotional\n immaturity was not a significant mitigating\n factor.” Id. at 1281.\n\n [3] “We agree that the record supports the\n sentencing judge’s findings that defendant\n had no previous adult felony convictions, no\n prior acts of violence, and\n\n\n 9\n See note 4, supra. Because the Arizona Supreme Court conducted\na de novo review, we focus on its analysis, rather than, as the dissent\ndoes in part, the analysis by the trial court.\n\n SPREITZ V. RYAN 17\n\n [4] that the defendant is capable of\n rehabilitation.” Id.\n\nThe Arizona Supreme Court also added that it considered\nremorse as an additional nonstatutory mitigating factor. Id.\n\n The court addressed Spreitz’s history of alcohol and\nsubstance abuse as follows:\n\n The record demonstrates defendant’s\n longtime substance abuse problems. We\n note, however, that defendant’s general\n problems with substance abuse are not\n essential to our decision here. We therefore\n decline to conclude that defendant was\n impaired by alcohol consumption to an extent\n that it interfered with his “capacity to\n appreciate the wrongfulness of his conduct or\n to conform his conduct to the requirements of\n the law.” A[riz]. R[ev]. S[tat]. [Ann.]\n § 13-703(G)(1); see also State v. Medrano,\n 185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)\n (citing [State v.] Stokley, 182 Ariz. [505,]\n 520, 898 P.2d [454,] 469 [1995]).\n\nId. at 1280–81. In the only other portion of its opinion\naddressing Spreitz’s history of alcohol and substance abuse,\nthe court said:\n\n The sentencing judge found that defendant’s\n ability to appreciate the wrongfulness of his\n conduct was not impaired on the night of the\n murder to any significant extent by substance\n abuse, emotional disorders, situational stress,\n or by a combination of these. Our review of\n\n18 SPREITZ V. RYAN\n\n the record convinces us that the trial court’s\n finding was proper.\n\nId. at 1281.\n\n The Arizona Supreme Court agreed with the trial court’s\naggravation finding that the murder was done in an\nespecially cruel manner, id. at 1279, and then proceeded to\nreweigh the applicable aggravating and mitigating factors,\nfinding that “the aggravating circumstance of especial\ncruelty in defendant’s murder of Ruby Reid outweigh[ed] all\nfactors mitigating in favor of leniency.” Id. at 1282. It\naffirmed his death sentence. Id. at 1283. Spreitz filed a\npetition for a writ of certiorari, which the United States\nSupreme Court denied. Spreitz v. Arizona, 523 U.S. 1027\n(1998) (mem).\n\n In March 2000, represented again by new counsel,\nSpreitz filed a petition for post-conviction relief in the\nArizona Superior Court (“PCR court”). See Ariz. R. Crim.\nP. 32.1. Spreitz alleged, inter alia, that the sentencing judge\nand the Arizona Supreme Court committed error by failing\nto consider his history of alcohol and substance abuse as a\nnonstatutory mitigating factor apart from its causal\nconnection to the murder—i.e., Eddings error. He also\nalleged that his appellate counsel was ineffective for failing\nto raise the sentencing judge’s Eddings error on appeal to the\nArizona Supreme Court. Spreitz additionally raised several\nnew claims of trial-counsel-ineffectiveness.\n\n The PCR court denied all of Spreitz’s claims and entered\nan order dismissing his petition for post-conviction relief. In\nPart 3 of its order, the PCR court discussed Spreitz’s claims\nof “nexus-error”—that both the trial court and the Arizona\nSupreme Court erred when they failed to consider Spreitz’s\nlongstanding alcohol and substance abuse on the basis that\n\n SPREITZ V. RYAN 19\n\nSpreitz had failed to establish a causal nexus between the\nlong-term substance abuse and the murder. 10 The PCR court\ndismissed the claim 11 as waived, reasoning that Spreitz had\nfailed to raise the issue on direct appeal. Nonetheless, the\nPCR court addressed the merits of the nexus-error claim in\nthe course of analyzing Spreitz’s argument that his appellate\ncounsel was ineffective for failing to raise it. In doing so,\nthe PCR court explained that the claim failed because:\n\n [I]t must be demonstrated, under A[riz].\n R[ev]. S[tat]. [Ann.] § 13-703(G)(l), that\n there is a causal link between the history of\n alcohol or substance abuse and the offense\n itself. E.g., State v. Stokley, [] 182 Ariz.\n [505,] 523 [898 P.2d 454, 472 (Ariz. 1995)].\n Without some basis for explaining or\n defining the individual’s behavior at the time\n of the offense, the Petitioner’s history of\n alcohol or substance abuse would be\n inconsequential (which is exactly what the\n trial court and Supreme Court concluded).\n State v. Kayer, 194 Ariz. 423, 984 P.2d 81\n ([Ariz.] 1999).\n\n At times, the court can and should consider\n an individual’s long-term alcoholism and\n\n 10\n For ease of reference, we refer to the Arizona courts’ alleged error\nas “nexus-error.”\n\n 11\n Although Spreitz raised, and the PCR court recognized, two\ndistinct claims—one with respect to the sentencing court and one with\nrespect to the Arizona Supreme Court—the PCR court’s analysis treats\nthe claims as one and the same and refers to them in the singular. To\nreflect accurately the PCR court’s discussion, our summary here likewise\nrefers to a singular “claim.”\n\n20 SPREITZ V. RYAN\n\n substance abuse, usually in conjunction with\n other factors or diagnosis, as non-statutory\n mitigation. However, the impact or effect of\n the alcoholism or substance abuse must be\n substantial and of such severity that it\n provides a sufficient basis for explaining the\n defendant’s conduct, character or ability to\n control his behavior at the time of the offense.\n [Citations omitted].\n\n As previously discussed, there is no evidence\n in Petitioner’s case to suggest that he suffered\n any long-term effects from his alcohol or\n drug abuse that precluded him from\n controlling his behavior. Petitioner did not\n suffer from any cognitive or emotional\n deficits that rendered him incapable of\n controlling his conduct. Therefore, the trial\n court did not err in failing to find Petitioner’s\n history of alcohol or substance abuse as a\n separate, non-statutory mitigating factor.\n [Citation omitted].\n\n Spreitz filed a petition for review in the Arizona\nSupreme Court challenging the PCR court’s judgment. The\nArizona Supreme Court summarily affirmed the PCR court’s\nmerits determination with respect to Spreitz’s Eddings\nclaims. State v. Spreitz (Spreitz II), 39 P.3d 525, 527 (Ariz.\n2002).\n\n Spreitz filed his federal habeas petition in February\n2003. He alleged in claim seven that the sentencing judge\nand the Arizona Supreme Court had both committed nexus-\nerror with respect to his long-time alcohol and substance\nabuse, and that appellate counsel was ineffective for failing\n\n SPREITZ V. RYAN 21\n\nto argue the sentencing judge’s nexus-error on direct appeal\nto the Arizona Supreme Court.\n\n The district court denied all of Spreitz’s claims. With\nrespect to the alleged nexus-error by both the sentencing\ncourt and the Arizona Supreme Court, the district court\nfound the claims procedurally barred, relying on the PCR\ncourt’s determination that Spreitz had waived the claims\nbecause he could have raised them on direct appeal but failed\nto do so. The district court concluded that Spreitz had\nproperly exhausted his ineffective assistance of appellate\ncounsel claim, and thus addressed the alleged nexus-error in\nthat context, ultimately concluding it was not meritorious.\nAccordingly, the district court concluded that “appellate\ncounsel’s failure to raise this issue on appeal does not\nconstitute ineffectiveness.”\n\n Spreitz timely appealed.\n\n II. Jurisdiction and Standard of Review\n\n We have jurisdiction pursuant to 28 U.S.C. §§ 1291\nand 2253(a), and review de novo the district court’s denial\nof a writ of habeas corpus. Poyson v. Ryan, 879 F.3d 875,\n887 (9th Cir. 2018). Because Spreitz filed his federal habeas\npetition after April 24, 1996, he must satisfy the\nAntiterrorism and Effective Death Penalty Act (“AEDPA”).\nUnder AEDPA, we may not grant habeas relief unless the\nstate’s adjudication of Spreitz’s claim (1) “was contrary to\n. . . clearly established federal law[] as determined by the\nSupreme Court,” (2) “involved an unreasonable application\nof” such law, or (3) “was based on an unreasonable\ndetermination of the facts in light of the evidence presented\nin the State court proceeding.” (28 U.S.C. § 2254(d)). “In\nmaking this determination, we look to the last state court\ndecision to address the claim,” White v. Ryan, 895 F.3d 641,\n\n22 SPREITZ V. RYAN\n\n665 (9th Cir. 2018) (citing Wilson v. Sellers, 138 S. Ct. 1188,\n1192 (2018)), which for Spreitz’s nexus-error claim is that\nof the PCR court.\n\n Spreitz argues that the PCR court’s decision was\n“contrary to” Eddings v. Oklahoma, 455 U.S. 104 (1982).\n“A state court’s decision is contrary to clearly established\nfederal law if it ‘applies a rule that contradicts the governing\nlaw set forth in [U.S. Supreme Court] cases’ or arrives at a\ndifferent result in a case that ‘confronts a set of facts that are\nmaterially indistinguishable from a decision of [the\nSupreme] Court.’” Castellanos v. Small, 766 F.3d 1137,\n1146 (9th Cir. 2014) (alteration in original) (quoting\nWilliams v. Taylor, 529 U.S. 362, 405–06 (2000)). “If the\nstate court applies a legal standard that contradicts clearly\nestablished [Supreme Court] law, we review de novo the\napplicant’s claims, applying the correct legal standard to\ndetermine whether the applicant is entitled to relief.” Id.\n(citation omitted).\n\n III. Discussion\n\n As discussed supra, the precise question we must decide\nis whether the Arizona Supreme Court applied its\nunconstitutional causal nexus test in violation of Eddings\nwhen it affirmed Spreitz’s death sentence. To answer that\nquestion, we must first determine whether the claim is\nproperly before us. After concluding that Spreitz’s claim\ncould not have been procedurally defaulted, we turn to the\nlevel of deference we must accord the PCR court’s ruling on\nSpreitz’s Eddings claim under AEDPA. Because the PCR\ncourt’s decision was contrary to clearly established Supreme\nCourt precedent, we accord that decision no deference and\nreview Spreitz’s Eddings claim de novo. We conclude that\nthe Arizona Supreme Court violated Eddings by\nimpermissibly requiring that Spreitz establish a causal\n\n SPREITZ V. RYAN 23\n\nconnection between his longstanding substance abuse and\nthe murder before considering and weighing the evidence as\na nonstatutory mitigating factor. Finally, having determined\nthat the Arizona Supreme Court applied an unconstitutional\ncausal nexus test in its sentencing procedure, we turn to\nwhether the error was harmless under Brecht v. Abrahamson,\n507 U.S. 619 (1993). We conclude that the error was not\nharmless and therefore reverse the district court’s judgment\nwith respect to Spreitz’s sentence.\n\n A.\n\n Under the doctrine of procedural default, a federal\nhabeas court “will not review the merits of claims, including\nconstitutional claims, that a state court declined to hear\nbecause the prisoner failed to abide by a state procedural\nrule.” Martinez v. Ryan, 566 U.S. 1, 9 (2012). “We review\nde novo a district court’s conclusion that a claim is\nprocedurally defaulted.” Cooper v. Neven, 641 F.3d 322,\n326 (9th Cir. 2011).\n\n Here, the district court concluded that both of Spreitz’s\nEddings claims—that the sentencing court and the Arizona\nSupreme Court applied an unconstitutional causal nexus\ntest—were procedurally defaulted, because the PCR court\nfound each claim waived for failure to raise them on direct\nappeal. This conclusion, however, is erroneous with respect\nto Spreitz’s claim that the Arizona Supreme Court violated\nEddings. After all, “[t]he Arizona Supreme Court reviews\ncapital sentences de novo, making its own determination of\nwhat constitute[s] legally relevant aggravating and\nmitigating factors, and then weighing those factors\nindependently. Ariz. Rev. Stat. Ann. § 13-755. The Arizona\nSupreme Court ‘conducts a thorough and independent\nreview of the record and of the aggravating and mitigating\nevidence to determine whether the sentence is justified.’”\n\n24 SPREITZ V. RYAN\n\nMcKinney, 813 F.3d at 819 (emphasis removed) (quoting\nState v. McKinney, 917 P.2d 1214, 1225 (Ariz. 1996)).\nSpreitz could not have raised on direct appeal his claim that\nthe Arizona Supreme Court violated Eddings in performing\nits de novo review of Spreitz’s death sentence. The first\nopportunity he had to raise that claim was before the PCR\ncourt, at which time he did so. Thus, Spreitz’s claim is not\nprocedurally defaulted and is properly before us. 12\n\n B.\n\n Because Spreitz’s claim that the Arizona Supreme Court\napplied an unconstitutional causal nexus test to his\nnonstatutory mitigating evidence of long-time alcohol and\n 12\n The district court declined to issue a certificate of appealability\n(“COA”) on Spreitz’s substantive Eddings claim, which he raised as\nclaim 1.7 on pages 108 through 111 in his federal habeas petition.\nAlthough Spreitz did not present his Eddings claim as a separate\nuncertified issue, we may grant a COA regarding any uncertified issue\ndiscussed in a petitioner’s opening brief. See 9th Cir. R. 22-1(e).\nSpreitz’s opening brief discusses the Arizona courts’ Eddings errors. We\ntherefore may exercise our discretion to expand the COA to encompass\nthe Eddings claim.\n\n To obtain a COA under 28 U.S.C. § 2253(c), Spreitz must\ndemonstrate that “reasonable jurists could debate whether . . . the\npetition should have been resolved in a different manner or that the issues\npresented were adequate to deserve encouragement to proceed further.”\nSlack v. McDaniel, 529 U.S. 473, 484 (2000). Reasonable jurists could\ndebate the merits of Spreitz’s Eddings claim. We thus exercise our\ndiscretion and expand the COA to encompass Spreitz’s substantive\nEddings claim with respect to the Arizona Supreme Court. We\nacknowledge that under Ninth Circuit Rule 22-1(f), the State would\nnormally be afforded an opportunity to submit supplemental briefing on\nthe uncertified Eddings issue. Here, however, the Eddings issue has been\nexhaustively briefed, in the context of Spreitz’s appellate counsel’s\nalleged ineffectiveness before the Arizona Supreme Court. Additional\nbriefing is unwarranted.\n\n SPREITZ V. RYAN 25\n\nsubstance abuse is properly before us, we next determine the\nlevel of deference to accord the PCR court’s adjudication of\nthat claim. As discussed above, although the PCR court first\ndeclared the claim waived, it proceeded to adjudicate the\nclaim on the merits. In denying the claim, the PCR court\nexplained:\n\n Without some basis for explaining or\n defining the individual’s behavior at the time\n of the offense, [Spreitz]’s history of alcohol\n or substance abuse would be inconsequential\n (which is exactly what the trial court and\n Supreme Court concluded).\n\n At times, the court can and should consider\n an individual’s long-term alcoholism and\n substance abuse, usually in conjunction with\n other factors or diagnosis, as non-statutory\n mitigation. However, the impact or effect of\n the alcoholism or substance abuse must be\n substantial and of such severity that it\n provides a sufficient basis for explaining the\n defendant’s conduct, character, or ability to\n control his behavior at the time of the offense.\n\n ....\n\n As previously discussed, there is no evidence\n in [Spreitz]’s case to suggest that he suffered\n any long-term effects from his alcohol or\n drug abuse that precluded him, from\n controlling his behavior. Petitioner did not\n suffer from any cognitive or emotional\n deficits that rendered him incapable of\n controlling his conduct. Therefore, the trial\n\n26 SPREITZ V. RYAN\n\n court did not err in failing to find [Spreitz]’s\n history of alcohol or substance abuse as a\n separate, non-statutory mitigating factor.\n\n In other words, the PCR court concluded that Spreitz’s\nlongstanding alcohol and substance abuse could only be\nconsidered a nonstatutory mitigating factor if it “provide[d]\na sufficient basis for explaining the defendant’s conduct,\ncharacter, or ability to control his behavior at the time of the\noffense.” This discussion both accurately described the\ncausal nexus test and approved of it.\n\n Such reasoning is contrary to clearly established federal\nlaw. As explained in McKinney, “the causal nexus test\nclearly violates Eddings.” 813 F.3d at 810. Therefore,\n“[b]ecause the state court used the wrong standard, we need\nnot defer to that decision.” Hardy v. Chappell, 849 F.3d 803,\n820 (9th Cir. 2016).\n\n C.\n\n Because the PCR court’s decision was “contrary to”\nclearly established law, we review de novo the merits of\nSpreitz’s Eddings claim. 28 U.S.C. § 2254(d)(1). We\ntherefore consider whether the Arizona Supreme Court\napplied its unconstitutional causal nexus test in reviewing\nand affirming Spreitz’s death sentence. In so doing, “we\nlook only to the decision of th[e Arizona Supreme Court],\n. . . only [considering] . . . the decision of the sentencing\njudge . . . to the degree it was adopted or substantially\nincorporated by the Arizona Supreme Court.” McKinney,\n813 F.3d at 819.\n\n SPREITZ V. RYAN 27\n\n 1.\n\n Our decision in McKinney frames our consideration of\nthe Arizona Supreme Court’s decision and Spreitz’s claim.\nMcKinney makes clear that the Arizona Supreme Court\nconsistently articulated and applied its unconstitutional\ncausal nexus test during the period in which Spreitz’s death\nsentence was litigated in the trial court and reviewed by the\nArizona Supreme Court. Id. at 824. Moreover, McKinney\nmakes plain that although on habeas review, we generally\napply a “presumption that state courts know and follow the\nlaw,” “the Arizona Supreme Court’s consistent articulation\nand application of its causal nexus test . . . make such a\ncourse impossible.” Id. at 803 (internal quotation marks and\ncitation omitted). In appeals heard from the late 1980s until\nat least 2002, see State v. Canez, 42 P.3d 564, 594 (Ariz.\n2002), it was unmistakably clear that “the Arizona Supreme\nCourt did not know and follow federal law.” Id. (internal\nquotation marks omitted). 13 Therefore, just as in McKinney\nand subsequent cases applying McKinney, see, e.g., Poyson,\n879 F.3d at 889, “the presumption” that the Arizona\nSupreme Court knew and followed the law “is rebutted” here\nas well. McKinney, 813 F.3d at 804.\n\n That said, McKinney does not dispose of Spreitz’s claim.\nWe must still review the Arizona Supreme Court’s decision\nto determine whether it did, in fact, apply its unconstitutional\ncausal nexus test in Spreitz’s case. In doing so, we are\nmindful of the fact that at the time the Arizona Supreme\nCourt decided Spreitz’s appeal, “if there is to be a\npresumption, it is that the Arizona Supreme Court violated\n\n\n 13\n In 2005, the Arizona Supreme Court repudiated its earlier nexus\ntest. See State v. Anderson, 111 P.3d 369, 391–92 (Ariz. 2005).\n\n28 SPREITZ V. RYAN\n\nthe dictates of Lockett and Eddings.” Greenway v. Ryan,\n866 F.3d 1094, 1095 (9th Cir. 2017).\n\n 2.\n\n Against this backdrop, we are convinced that the Arizona\nSupreme Court applied its causal nexus test with respect to\nSpreitz’s evidence of longstanding alcohol and substance\nabuse in violation of Eddings.\n\n We quote again the Arizona Supreme Court’s discussion\nof Spreitz’s history of alcohol and substance abuse:\n\n The record demonstrates defendant’s\n longtime substance abuse problems. We\n note, however, that defendant’s general\n problems with substance abuse are not\n essential to our decision here. We therefore\n decline to conclude that defendant was\n impaired by alcohol consumption to an extent\n that it interfered with his “capacity to\n appreciate the wrongfulness of his conduct or\n to conform his conduct to the requirements of\n the law.” A[riz]. R[ev]. S[tat]. [Ann.] § 13-\n 703(G)(1); see also State v. Medrano,\n 185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)\n (citing Stokley, 182 Ariz. at 520, 898 P.2d at\n 469).\n\n ....\n\n The sentencing judge found that defendant’s\n ability to appreciate the wrongfulness of his\n conduct was not impaired on the night of the\n murder to any significant extent by substance\n abuse, emotional disorders, situational stress,\n\n SPREITZ V. RYAN 29\n\n or by a combination of these. Our review of\n the record convinces us that the trial court’s\n finding was proper.\n\nSpreitz I, 945 P.2d at 1280–81. This discussion\ndemonstrates that the Arizona Supreme Court primarily\nconcluded that Spreitz failed to show statutory mitigation\nunder Ariz. Rev. Stat. Ann. § 13-703(G)(1) based on his\nintoxication on the night of the murder. “When applied\nsolely in the context of statutory mitigation under\n§ 13-703(G)(1), the causal nexus test does not violate\nEddings.” McKinney, 813 F.3d at 810.\n\n The plain text of the Arizona Supreme Court’s decision,\nhowever, also clearly demonstrates that the court did not\nsolely apply its causal nexus test to Spreitz’s evidence of\nstatutory mitigation. The court initially recognized Spreitz’s\n“general problems with substance abuse”—evidence that\nshould have been relevant as nonstatutory mitigation—but\nconcluded those problems were “not essential to [its]\ndecision” because they did not show Spreitz was “impaired\non the night of the murder.” Spreitz I, 945 P.2d at 1280–81.\nIn other words, the court did not acknowledge the relevance\nof longtime substance abuse as nonstatutory mitigation in the\nabsence of a causal relationship to the crime. Instead, the\ncourt held that because there was no causal relationship,\nSpreitz’s long-term alcohol abuse was not of significance—\ni.e. “not essential”—to the court’s overall determination of\neither statutory or nonstatutory mitigation.\n\n The Supreme Court, however, has been clear that “‘full\nconsideration of evidence that mitigates against the death\npenalty is essential if the sentencer is to give a reasoned\nmoral response to the defendant’s background, character,\nand crime.’” McKinney, 813 F.3d at 812 (first emphasis\n\n30 SPREITZ V. RYAN\n\nadded) (quoting Penry I, 492 U.S. at 319, 328).\nAccordingly, “Eddings requires that ‘[t]he sentencer must\nalso be able to consider and give effect’” to any relevant\nmitigation evidence proffered by the defendant, which\nincludes evidence going toward nonstatutory mitigation. Id.\n(quoting Penry I, 492 U.S. at 319, 328). The Arizona\nSupreme Court’s refusal to consider Spreitz’s alcohol and\nsubstance abuse beyond its connection, if any, to Reid’s\nmurder constitutes application of its unconstitutional causal\nnexus test to relevant nonstatutory mitigation evidence.\n\n Our understanding of the Arizona Supreme Court’s\nopinion is bolstered by the fact that in its discussion of\nSpreitz’s longstanding alcohol and substance abuse, the\ncourt cited to a prior decision that likewise applied the\nunconstitutional causal nexus test, State v. Medrano,\n914 P.2d 255 (Ariz. 1996). In McKinney, we explained that\nin Medrano, the Arizona Supreme Court upheld the\nsentencing judge’s determination that the defendant’s\ncocaine use failed as a nonstatutory mitigating circumstance\nbecause he did not show that it contributed to his conduct on\nthe night of the murder. McKinney, 813 F.3d at 825–26. In\nso doing, “[t]he Arizona Supreme Court applied the causal\nnexus test.” Id. at 825. Although the Arizona Supreme\nCourt’s citation to Medrano is not dispositive, it certainly\ncorroborates our understanding of the court’s opinion. See\nalso Poyson, 879 F.3d at 890 (emphasizing, among other\nconsiderations, that the Arizona Supreme Court cited a\npassage from a case specifically identified in McKinney as\napplying the unconstitutional causal nexus test in order to\nconclude that the state court also applied that test to Poyson’s\nevidence). Moreover, as we explained in McKinney, “the\nArizona Supreme Court applied its unconstitutional nexus\ntest consistently” for fifteen years because it had “a mistaken\nunderstanding of Eddings.” 813 F.3d at 826. It is thus\n\n SPREITZ V. RYAN 31\n\nentirely logical that in citing the exact portion of Medrano\nwhere it previously applied its causal nexus test, the Arizona\nSupreme Court was applying the same test in Spreitz’s case.\nSee id. at 821; see also Poyson, 879 F.3d at 890.\n\n We find further support for our understanding of the\nArizona Supreme Court’s unconstitutional treatment of\nSpreitz’s alcohol and substance abuse when we compare it\nwith the manner in which the court discussed other\nmitigating factors. As detailed earlier, in evaluating the\nevidence of Spreitz’s general history of alcohol and\nsubstance abuse, the court stated that this evidence was “not\nessential” to its decision. Spreitz I, 945 P.2d at 1280. By\ncontrast, when the court considered other mitigating factors,\nit specifically discussed the weight it would give those\nfactors, as opposed to simply dismissing them outright. For\nexample, when discussing Spreitz’s subnormal upbringing,\nthe court stated, “Although we recognize defendant’s\nupbringing as a mitigating circumstance, we accord it little\nweight.” Id. at 1280 (emphasis added). As another example,\nafter “finding that since his arrest, [Spreitz] has\ndemonstrated remorse,” the court stated that his “remorse for\nhis actions does little to counterbalance especial cruelty as a\nserious aggravating circumstance.” Id. at 1281 (emphasis\nadded). And finally, when discussing the sentencing judge’s\ndetermination regarding Spreitz’s emotional immaturity, the\ncourt stated, “We also find that the sentencing judge . . .\nproperly found that his emotional immaturity was not a\nsignificant mitigating factor.” Id. (emphasis added). In\nother words, when the Arizona Supreme Court wanted to\nassign weight to a given factor, it said so. This pattern stands\nin stark contrast to the court’s conclusion that anything\nrelated to Spreitz’s long-term alcohol or substance abuse that\nfell short of the statutory definition could not serve as\nmitigation. Id. at 1280–81.\n\n32 SPREITZ V. RYAN\n\n Finally, as also discussed earlier, the PCR court\ndistinctly interpreted the Arizona Supreme Court’s decision\nas having applied a causal nexus test. The PCR court\nexplained that “[w]ithout some basis for explaining or\ndefining the individual’s behavior at the time of the offense,\n[Spreitz]’s history of alcohol or substance abuse would be\ninconsequential (which is exactly what the trial court and\nSupreme Court concluded).” Moreover, in articulating how\nthe Arizona Supreme Court had applied its causal nexus test\nin Spreitz’s direct appeal, the PCR court, as had the Arizona\nSupreme Court, cited to a decision of the Arizona Supreme\nCourt indisputably applying an unconstitutional nexus test,\nState v. Kayer, 984 P.2d 31 (Ariz. 1999). See McKinney,\n813 F.3d at 816 (explaining that Kayer held “that the\ndefendant’s mental impairment ‘was not established as a\nnonstatutory mitigating factor’ in part because ‘defendant\noffered no evidence to show the requisite causal nexus that\nmental impairment affected his judgment or his actions at the\ntime of the murder’” (quoting Kayer, 984 P.2d at 46)). The\nfact that the PCR court interpreted the Arizona Supreme\nCourt’s opinion as we do—and then itself applied the\nconstitutionally erroneous nexus test—lends further support\nto our reading.\n\n We recognize that the Arizona Supreme Court’s decision\non the particular issue before us was framed by language that\nmight, in a decisional vacuum, suggest the court knew,\nunderstood, and applied the law with respect to Eddings.\nThe court acknowledged that Spreitz argued that his history\nof alcohol and drug abuse served as both a statutory and\nnonstatutory mitigating factor. Spreitz I, 945 P.2d at 1279.\nIt then stated that it “must consider any aspect of the\ndefendant’s character or record . . . relevant to determining\nwhether the death penalty should be imposed.” Id. And\nfinally, in reaching the conclusion to affirm Spreitz’s death\n\n SPREITZ V. RYAN 33\n\nsentence, the court explained that it had “examin[ed] the\nentire record and reweigh[ed] the applicable aggravating and\nmitigating factors.” Id. at 1281–82.\n\n The Arizona Supreme Court’s acknowledgment of what\nSpreitz argued does not indicate that it relied on or accepted\nthat argument as correct. Further, we cannot, given\nMcKinney, Poyson, and Greenway, proceed in a decisional\nvacuum.\n\n Considered in light of the baseline that the Arizona\nSupreme Court was consistently applying its\nunconstitutional causal nexus test in death penalty cases at\nthe time Spreitz’s appeal was decided, as well as its citations\nto cases held in McKinney to be applying that standard, we\nare confident that the manner in which the Arizona Supreme\nCourt disposed of Spreitz’s evidence of his longstanding\nalcohol and substance abuse confirms that it applied the\ncausal nexus test it was using during that period.\n\n Styers v. Schriro, 547 F.3d 1026 (9th Cir. 2008) supports\nthis conclusion. Styers stated that the mere fact that “the\nArizona Supreme Court stated that it had ‘considered all of\nthe proffered mitigation,’” is not sufficient to contradict the\nfact that “its analysis prior to this statement indicated\notherwise.” Id. at 1035 (internal citation omitted). In Styers,\nnotwithstanding assertions to the contrary, the Arizona\nSupreme Court refused to consider evidence of the\ndefendant’s Post Traumatic Stress Disorder (“PTSD”)\nbecause he had failed to connect it causally to his crimes. Id.\nHere, too, the Arizona Supreme Court’s actual analysis of\nSpreitz’s longstanding alcohol and substance abuse reveals\nthat it refused to consider this particular aspect of Spreitz’s\ncharacter free from any connection to the crime. Rather, the\ncourt’s discussion demonstrates that in evaluating Spreitz’s\nlong-term alcohol and substance abuse, it applied a causal\n\n34 SPREITZ V. RYAN\n\nnexus test in violation of Eddings and therefore refused, as a\nmatter of law, to consider it in assessing Spreitz’s evidence\nof nonstatutory mitigating circumstances.\n\n To the extent the state argues the Arizona Supreme\nCourt’s opinion is ambiguous as to whether it applied an\nunconstitutional causal nexus test or not—and we believe\nthat it is not—McKinney requires resolving that ambiguity\nin favor of Spreitz. McKinney held it clear that the Arizona\nSupreme Court “consistently articulated and applied its\ncausal nexus test” at the time the trial court imposed\nSpreitz’s death sentence and the Arizona Supreme Court\nreviewed and affirmed that sentence. McKinney, 813 F.3d\nat 803 (emphasis in original). Again, “if there is to be a\npresumption, it is that the Arizona Supreme Court violated\nthe dictates of Lockett and Eddings.” Greenway, 866 at\n1095.\n\n In sum, because (1) the Arizona Supreme Court clearly\nstated that Spreitz’s longstanding alcohol and substance\nabuse was “not essential” to its decision, which was in line\nwith the court’s consistent practice of applying a causal\nnexus test; (2) the Arizona Supreme Court cited to Medrano,\nin which it had previously articulated a causal nexus test;\n(3) the Arizona Supreme Court carefully laid out the weight\nit would accord other mitigating factors as opposed to simply\nignoring them; and (4) the PCR court likewise interpreted\nthe Arizona Supreme Court’s opinion as applying a causal\nnexus test, we conclude that the Arizona Supreme Court\nheld, as a matter of law, that Spreitz’s long-term alcohol and\nsubstance abuse was not a nonstatutory mitigating factor and\ntherefore refused to consider it. This holding was in clear\nviolation of Eddings.\n\n SPREITZ V. RYAN 35\n\n 3.\n\n In light of our conclusion that the Arizona Supreme\nCourt failed to comply with Eddings, we must next decide\nwhether Spreitz was prejudiced by this error. On federal\nhabeas review, relief is warranted if an error “had substantial\nand injurious effect or influence” on the challenged decision.\nSee Brecht v. Abrahamson, 507 U.S. 619, 623 (1993). The\npetitioner is “not entitled to habeas relief” unless he can\nestablish that the error “resulted in actual prejudice.” Davis\nv. Ayala, 135 S. Ct. 2187, 2197 (2015) (internal quotation\nmarks omitted). Here, because we are “in grave doubt about\nwhether [the Arizona Supreme Court’s] error of federal law\nhad substantial and injurious effect or influence in\ndetermining [whether to affirm Spreitz’s death sentence],\nthat error is not harmless. And, [Spreitz] must win.”\nMcKinney, 813 F.3d at 822 (quoting O’Neal v. McAninch,\n513 U.S. 432, 435 (1995)).\n\n We previously have required resentencing based on\nprejudicial Eddings error where the Arizona Supreme Court\nrefused to consider, as a matter of law, mitigation evidence\n“central to” the petitioner’s “plea for leniency.” See id. at\n823; see also Coleman v. Calderon, 210 F.3d 1047, 1051\n(9th Cir. 2000) (concluding that a penalty-phase jury\ninstruction error was not harmless under Brecht because “it\nundermined the very core of Coleman’s plea for life”). 14 For\nthe reasons that follow, it is clear that Spreitz’s long-term\n\n 14\n We do not suggest that Brecht prejudice may only be established\nif inappropriately considered mitigating evidence was at “the very core\nof [a defendant’s] plea for life.” Coleman, 210 F.3d at 1051. It is\nconceivable that other forms of Eddings error could have “a substantial\nand injurious effect” on a defendant’s sentence if they would be\nsignificant to the sentencer, despite the defendant’s failure to rely on\nthem as central to his or her argument. Brecht, 507 U.S. at 623.\n\n36 SPREITZ V. RYAN\n\nalcohol and substance abuse was important mitigation\nevidence, and was so closely related as to become central to\nhis plea for leniency. The Arizona Supreme Court’s refusal\nto consider that evidence had a substantial and injurious\neffect on its sentencing determination that requires\nresentencing.\n\n While a defendant’s evidence of longstanding alcohol\nand/or substance abuse can be probative mitigation evidence\non its own, a “history of substance abuse [is] substantially\nmitigating when . . . combined with other mitigating\nevidence.” Henry v. Ryan, 720 F.3d 1073, 1090 (9th Cir.\n2013) (emphasis added) (listing cases and holding that the\ndefendant’s alcoholism “would not have had a substantial\nand injurious effect” because it “st[ood] alone, was similar\nto evidence already considered by the sentencing courts, and\nwas of limited probative value”) (footnote omitted). 15 Here,\nSpreitz’s evidence regarding his history of alcohol and\nsubstance abuse—spanning nearly half his life by the time\nhe committed the crime at the age twenty-two—was the\n\n 15\n The dissent makes much of Henry, suggesting it dictates the\noutcome of our prejudice analysis. Dissent at 29. Not so. In that case,\nthe state court had found that Henry’s intoxication on the day of the\nmurder did constitute a statutory mitigating factor but was not\nsufficiently substantial to outweigh the two aggravating factors. Henry,\n720 F.3d at 1090. Henry argued that the state court erred by failing to\nconsider evidence of his “historical alcoholism” in addition to his\nintoxication on the day of the crime. Id. We concluded that any error\nwas harmless because the history of substance abuse “was similar to\nevidence already considered by the sentencing courts and was of limited\nprobative value.” Id. In other words, because the state court had found\ninsufficient evidence of alcohol use that was tied to the murder, we could\nnot conclude that general alcohol use that was not tied to the murder\ncould have been determinative. Here, by contrast, the Arizona Supreme\nCourt found that Spreitz’s evidence of intoxication did not rise to the\nlevel of statutory mitigation. Accordingly, there is not the same\ncontextual harmlessness problem as in Henry.\n\n SPREITZ V. RYAN 37\n\nfactor on which much of his other mitigation evidence\nturned.\n\n In both his report and his testimony, Dr. Flynn discussed\nat length the extent to which alcohol consumed Spreitz’s life\nin the decade leading up to Reid’s murder. Spreitz began\ndrinking when he was twelve or thirteen; by fifteen, he was\ndrinking steadily on the weekends and drinking vodka before\nschool. According to Dr. Flynn, “[d]rug and alcohol abuse\ndominated his teenage years.” After he left home, Spreitz’s\ndrinking intensified: Spreitz experienced blackouts multiple\ntimes and, a week before the murder, Spreitz was\n“intoxicated a majority of the time.” Dr. Flynn concluded\nthat “[i]t appears completely clear from the available\ninformation that . . . Spreitz had a long-standing problem\nwith alcohol which probably reached the level of physical\ndependence.”\n\n Critically, Spreitz’s decade-long struggle with alcohol\nand other addictive substances reflects the true gravity of his\n“subnormal upbringing,” a mitigating circumstance\nrecognized by the Arizona Supreme Court. See Spreitz I,\n945 P.2d at 1280. Both Spreitz’s sentencing memorandum\nand Dr. Flynn detailed how his difficult childhood led to\nSpreitz’s early dependence on alcohol. For instance, Dr.\nFlynn explained that Spreitz turned to alcohol as a coping\nmechanism to deal with the “pervasive subabusive\nemotional battering and neglect that he received from his\nmother,” whom he described as “punitive, controlling, [and]\nemotionally cold.” Dr. Flynn found that Spreitz “slipped\nquietly into alcoholic numbness” after his “failure to live up\nto his mother’s standards or earn her love.” Reflecting\nparental neglect, Spreitz’s parents “failed to arrange . . . for\nan appropriate rehabilitation program” even though\n“[v]irtually everyone . . . who spent much time with him\n\n38 SPREITZ V. RYAN\n\ndescribed him as a heavy drinker.” “Given the intensity and\nduration” of Spreitz’s reliance on alcohol, “the mother (and\nstep-father) were either so uninvolved with him to have\nfailed to notice the problem, or they were aware of it and\nfailed to arrange (or attempt to arrange) for an appropriate\nrehabilitation program.”\n\n Because Spreitz did not receive treatment during his\nteenage years, his dependence on alcohol only deepened. In\nhis sentencing memorandum, Spreitz, too, argued that his\nlong-term alcohol abuse reflects the impact that his\nupbringing by an emotionally unavailable, manipulative,\npunitive mother had on him: it was “[n]o wonder that by 12\nor 13 [h]e began drinking alcohol and smoking marijuana.”\nAlthough the Arizona Supreme Court recognized that\nSpreitz’s “subnormal” upbringing was “a mitigating\ncircumstance,” it “accord[ed] it little weight.” Spreitz I,\n945 P.2d at 1280. Had the court properly considered\nSpreitz’s longstanding history of alcohol abuse in\nconjunction with and as a manifestation of his disturbing\nchildhood, both mitigating circumstances—and most\nimportantly, the two in combination—would have been\nmore compelling.\n\n The dissent argues that Spreitz’s long-term history of\nalcohol abuse cannot be central to his plea for leniency\nbecause much of his mitigation evidence focused on his\n“abusive and dysfunctional childhood, upbringing, and\ninteractions with his mother,” Dissent at 66, not his history\nof alcohol and substance abuse. This argument, however,\nmisses the point, as it overlooks the interlinked relationship\nbetween Spreitz’s alcoholism and his subnormal childhood:\nSpreitz’s history of alcohol and substance abuse is central to\nevaluating the actual significance of his childhood as a\nmitigating factor, and vice versa.\n\n SPREITZ V. RYAN 39\n\n In addition to being a symptom of his dysfunctional\nchildhood, Spreitz’s alcohol and substance abuse can be seen\nas linked to his emotional immaturity, another nonstatutory\nmitigating circumstance recognized by the Arizona Supreme\nCourt but described as not “significant.” Spreitz I, 945 P.2d\nat 1282. In his report, Dr. Flynn explained that Spreitz was\n“very likely to have been socially and emotionally\nimmature” due, at least in part, to “[y]ears of alcoholism\nintoxication [that] wipe[d] out many of the . . . healthy\ndevelopmental processes requisite to age-appropriate social\nand emotional maturity.” Similarly, in his sentencing\nmemorandum, Spreitz connected “[h]is developmental\nfailure” with his “[n]umbing by use of alcohol.” Again, had\nthe Arizona Supreme Court properly considered Spreitz’s\nlongstanding alcohol and substance abuse, another\nmitigating factor—Spreitz’s emotional immaturity—would\nhave been brought into proper perspective.\n\n Thus, by failing to consider Spreitz’s longstanding\nalcohol and substance abuse, the Arizona Supreme Court\nwas left with a critical void in Spreitz’s narrative: Spreitz’s\nsubnormal childhood was so emotionally disturbing that it\nled him to drink by the age of twelve or thirteen, which, in\nturn, disrupted his normal development and contributed to\nhis emotional immaturity. As both a symptom and cause of\nother mitigating factors, Spreitz’s longstanding alcohol\nabuse was thus central to his plea for leniency. 16 Without\n\n 16\n Contrary to the dissent’s suggestion, Spreitz did not focus on his\nintoxication at the time of the crime to the exclusion of his history of\nalcohol abuse. Dissent at 22–24. Rather, both he and Dr. Flynn\ndistinguished between Spreitz’s history of alcohol abuse as nonstatutory\nmitigation and his intoxication at the time of the crime as statutory\nmitigation. It is Spreitz’s history of alcohol abuse that underlies and\namplifies much of his other mitigation evidence, not his intoxication at\nthe time of Reid’s murder.\n\n40 SPREITZ V. RYAN\n\nconsideration of the evidence of his long-term substance and\nalcohol abuse, the Arizona Supreme Court’s conclusions\nregarding all of Spretiz’s mitigation evidence were\nfundamentally flawed.\n\n We come to this conclusion fully recognizing, as the\ndissent emphasizes, that “the severity of an aggravating\ncircumstance must be considered when determining whether\nan Eddings error is harmless.” Dissent at 71. The Arizona\nSupreme Court described what Spreitz’s “confession and\nphysical evidence” revealed: that “Spreitz beat and raped\n[Reid] in a brutal assault that lasted many minutes before he\ncrushed her skull.” Spreitz I, 945 P.2d at 1279. For this\nreason, the Arizona Supreme Court affirmed the sentencing\ncourt’s finding that Reid’s murder was especially cruel. Id.\nEven so, we found prejudicial nexus-error in McKinney\nwhere the murder was done in a cruel manner for pecuniary\ngain as part of a double murder. McKinney, 813 F.3d at 823.\n“When the choice is between life and death,” “the risk that\nthe death penalty will be imposed in spite of facts which may\ncall for a less severe penalty. . . . is unacceptable.” Abdul-\nKabir v. Quarterman, 550 U.S. 233, 264 (2007) (quoting\nLockett v. Ohio, 438 U.S. 586, 605 (1978)). Accordingly,\nthe sentencer “must be able to give meaningful consideration\nand effect to all mitigating evidence that might provide a\nbasis for refusing to impose the death penalty on a particular\nindividual, notwithstanding the severity of his crime.” Id. at\n247 (emphasis added).\n\n Here, the Arizona Supreme Court failed to give any\nmeaning to Spreitz’s longstanding alcohol and substance\nabuse, an oversight which, in turn, minimized the value of\nother mitigating evidence as well. Because the court’s\nanalysis was missing a core component, we “cannot say,\nwith fair assurance” that the Arizona Supreme Court’s\n\n SPREITZ V. RYAN 41\n\nEddings error did not affect its decision. McKinney,\n813 F.3d at 822. As a result, the Arizona Supreme Court’s\nrefusal to consider the evidence was not harmless under\nBrecht.\n\n IV. Conclusion\n\n We reverse in part the district court’s judgment denying\na writ of habeas corpus. We remand with instructions to\ngrant the writ with respect to Spreitz’s sentence unless the\nstate, within a reasonable period, either corrects the\nconstitutional error in his death sentence or vacates the\nsentence and imposes a lesser sentence consistent with the\nlaw.\n\n REVERSED in part and REMANDED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Christopher Spreitz v. Charles Ryan\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-04\\nOpinion type: majority\\nHas dissenting opinion: Yes\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nCHRISTOPHER J. SPREITZ, No. 09-99006\\n Petitioner-Appellant,\\n D.C. No.\\n v. 4:02-CV-00121-JMR\\n\\nCHARLES L. RYAN, Warden,\\n Respondent-Appellee. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n John M. Roll, District Judge, Presiding\\n\\n Argued July 11, 2013\\n Submitted March 4, 2019\\n San Francisco, California\\n\\n Filed March 4, 2019\\n\\n Before: Richard A. Paez, Marsha S. Berzon,\\n and Richard C. Tallman, Circuit Judges.\\n\\n Opinion by Judge Paez;\\n Dissent by Judge Tallman\\n\\n2 SPREITZ V. RYAN\\n\\n SUMMARY *\\n\\n\\n Habeas Corpus / Death Penalty\\n\\n The panel reversed the district court’s denial of habeas\\ncorpus relief with respect to Christopher J. Spreitz’s death\\nsentence, and remanded, in a case in which Spreitz argued\\nthat the Arizona Supreme Court violated Eddings v.\\nOklahoma, 455 U.S. 104 (1982), by refusing to consider, as\\na matter of law, mitigating evidence of Spreitz’s\\nlongstanding alcohol and substance abuse on the ground that\\nhe did not establish a causal connection between this\\nmitigating evidence and the crime.\\n\\n The panel held that the district court erred in concluding\\nthat Spreitz’s claim that the Arizona Supreme Court violated\\nEddings is procedurally defaulted. The panel explained that\\nthe first opportunity Spreitz had to raise that claim was\\nbefore the post-conviction-relief (PCR) court, at which time\\nhe did so.\\n\\n Because the decision of the PCR court – which first\\ndeclared the claim waived, but proceeded to adjudicate the\\nclaim on the merits – was contrary to clearly established\\nSupreme Court precedent, the panel accorded that decision\\nno deference and reviewed Spreitz’s Eddings claim de novo.\\nThe panel concluded that the Arizona Supreme Court\\nviolated Eddings by impermissibly requiring that Spreitz\\nestablish a causal connection between his longstanding\\nsubstance abuse and the murder before considering and\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n SPREITZ V. RYAN 3\\n\\nweighing the evidence as a nonstatutory mitigating factor.\\nThe panel concluded that the error was not harmless.\\n\\n The panel affirmed the district court’s judgment denying\\nrelief with respect to Spreitz’s conviction in a concurrently\\nfiled memorandum disposition.\\n\\n Dissenting, Judge Tallman wrote that the record does not\\nestablish that either the sentencing court or the Arizona\\nSupreme Court unconstitutionally refused to consider\\nrelevant mitigating evidence; and that even if the Arizona\\ncourts did violate Eddings, Spreitz cannot show that this\\nerror had a “substantial and injurious effect or influence” on\\nhis ultimate sentence.\\n\\n\\n COUNSEL\\n\\nTimothy M. Gabrielsen (argued), Assistant Federal Public\\nDefender; John M. Sands, Federal Public Defender; Office\\nof the Federal Public Defender, Tucson, Arizona; Susan B.\\nFox and Sean Bruner, Law Office of Sean Bruner Ltd.,\\nTucson, Arizona; for Petitioner-Appellant.\\n\\nLacey Stover Gard (argued) and Jeffrey A. Zick, Section\\nChief Counsel; Kent E. Cattani, Chief Counsel, Criminal\\nAppeals/Capital Litigation Section; Mark Brnovich,\\nAttorney General; Office of the Attorney General, Tucson,\\nArizona; for Respondents-Appellees.\\n\\n4 SPREITZ V. RYAN\\n\\n OPINION\\n\\nPAEZ, Circuit Judge:\\n\\n In 1994, an Arizona jury convicted Christopher J. Spreitz\\n(“Spreitz”) of first-degree murder. The victim was thirty-\\nnine year old Ruby Reid (“Reid”). Finding that the cruelty\\nof the murder outweighed any mitigating circumstances, the\\ntrial judge sentenced Spreitz to death. Spreitz appeals the\\ndistrict court’s denial of his petition for a writ of habeas\\ncorpus challenging his conviction and sentence. We affirm\\nthe district court’s judgment with respect to Spreitz’s\\nconviction, 1 and reverse with respect to his sentence. 2\\n\\n In challenging his sentence, Spreitz argues that the\\nArizona Supreme Court unconstitutionally affirmed his\\ndeath sentence by failing to consider mitigating evidence of\\nhis longstanding alcohol and substance abuse. He contends\\nthat the state court refused to consider, as a matter of law,\\nthis evidence in mitigation because he did not establish a\\ncausal connection between the crime and his long-term\\nalcohol and substance abuse. In Eddings v. Oklahoma,\\n455 U.S. 104, 110 (1982), the Supreme Court held that under\\nboth the Eighth and Fourteenth Amendments, a sentencer in\\na capital case may not “refuse to consider, as a matter of law,\\n\\n 1\\n We affirm the judgment with respect to Spreitz’s conviction in a\\nconcurrently filed memorandum disposition.\\n\\n 2\\n On February 13, 2015, we vacated submission of Spreitz’s case\\npending final resolution of the en banc proceedings in McKinney v. Ryan,\\n730 F.3d 903 (9th Cir. 2013). In December 2015, the en banc court\\nissued an opinion in McKinney. 813 F.3d 798 (9th Circ. 2015) (en banc).\\nAfter the Supreme Court denied the State’s petition for a writ of\\ncertiorari, Ryan v. McKinney, 137 S. Ct. 39 (2016) (mem), we ordered\\nsupplemental briefing on the applicability of McKinney to Spreitz’s case.\\n\\n SPREITZ V. RYAN 5\\n\\nany relevant mitigating evidence” offered by the defendant.\\nId. at 114. Although a sentencer “may determine the weight\\nto be given relevant mitigating evidence . . . they may not\\ngive it no weight by excluding such evidence from their\\nconsideration.” Id. at 114–15 (footnote omitted). In\\ninterpreting and applying Eddings, the Supreme Court has\\nexplained that “full consideration of evidence that mitigates\\nagainst the death penalty is essential if the [sentencer] is to\\ngive a reasoned moral response to the defendant’s\\nbackground, character, and crime.” Penry v. Lynaugh\\n(Penry I), 492 U.S. 302, 328 (1989), abrogated on other\\ngrounds by Atkins v. Virginia, 536 U.S. 304 (2002) (internal\\nquotation marks and citation omitted). Moreover, the\\nSupreme Court has been clear: requiring a defendant to\\nprove a causal nexus between his mitigating evidence and\\nthe crime is “a test we never countenanced and now have\\nunequivocally rejected.” Smith v. Texas, 543 U.S. 37, 45\\n(2004) (per curiam).\\n\\n At the time of Spreitz’s sentencing, Arizona Revised\\nStatute Annotated § 13-703(G)(1994) 3 listed five mitigating\\nfactors, and Arizona case law additionally recognized\\nnonstatutory mitigating factors, including, for example, a\\ndefendant’s difficult family background or mental condition\\nnot severe enough to qualify as a statutory mitigating factor.\\nIn an en banc decision of our court, McKinney v. Ryan,\\n813 F.3d 798 (9th Cir. 2015), cert denied, 137 S. Ct. 39\\n(2016) (mem), we explained:\\n\\n For a period of a little over 15 years in capital\\n cases, in clear violation of Eddings, the\\n\\n 3\\n Arizona has since revised its death penalty sentencing scheme. All\\nreferences to Arizona’s Revised Statute Annotated are to those\\nprovisions in effect at the time of Spreitz’s sentencing.\\n\\n6 SPREITZ V. RYAN\\n\\n Supreme Court of Arizona articulated and\\n applied a “causal nexus” test for nonstatutory\\n mitigation that forbade as a matter of law\\n giving weight to mitigating evidence, such as\\n family background or mental condition,\\n unless the background or mental condition\\n was causally connected to the crime.\\n\\nId. at 802. As a result, we held in McKinney that\\n“[a]pplication of the causal nexus test to nonstatutory\\nmitigating factors violated Eddings, for it resulted in Arizona\\ncourts being entirely forbidden, as a matter of state law, to\\ntreat as a mitigating factor a family background or a mental\\ncondition that was not causally connected to a defendant’s\\ncrime.” Id. Spreitz argues that the Arizona Supreme Court\\napplied its causal nexus test in his case, refusing to consider\\nevidence of his long-term substance and alcohol abuse\\nbecause he did not adequately establish a causal connection\\nbetween that history of abuse and his crime.\\n\\n As in McKinney, “the precise question before us is\\nwhether the Arizona Supreme Court applied its\\nunconstitutional causal nexus test in affirming [Spreitz]’s\\ndeath sentence on de novo review.” 4 Id. at 804 (emphasis\\n\\n\\n 4\\n The Arizona Supreme Court is required by statute to undertake an\\nindependent review of a death sentence. Ariz. Rev. Stat. § 13-703.01(A).\\nIn conducting that review for crimes that occurred before 2002, the court\\n“independently review[s] the trial court’s findings of aggravation and\\nmitigation and the propriety of the death sentence. In doing so, [the\\ncourt] review[s] the record de novo, considering the quality and the\\nstrength, not simply the number, of aggravating and mitigating factors.”\\nState v. Lynch, 357 P.3d 119, 141 (Ariz. 2015) (internal quotation marks\\nand citations omitted), rev’d on other grounds, Lynch v. Arizona, 136 S.\\nCt. 1818 (2016).\\n\\n SPREITZ V. RYAN 7\\n\\nand internal quotation marks omitted). For the reasons that\\nfollow, we conclude that it did.\\n\\n I.\\n\\n A. Spreitz’s Crimes, Conviction, and Sentence\\n\\n On May 25, 1989, the police arrested Spreitz after\\ndiscovering Ruby Reid’s body in the desert. Upon\\nquestioning, Spreitz confessed to murdering Reid. We\\nbriefly provide the facts of the murder.\\n\\n On the evening of May 18, after drinking heavily and\\nbeing rejected by the woman he was dating, Spreitz “picked\\nup” Reid at a convenience store. State v. Spreitz (Spreitz I),\\n945 P.2d 1260, 1264–65 (Ariz. 1997). 5 In his confession,\\nSpreitz claimed that Reid voluntarily left with him and that\\nhis understanding was that they would have sex later that\\nevening. Id. at 1265. Spreitz further claimed that he drove\\nher out to the desert, where Reid decided she no longer\\nwanted to have sex. Id. The two fought as a result. Id.\\nSpreitz explained that Reid slapped him and that he\\nresponded by punching her in the mouth. Id. Spreitz then\\nsexually assaulted Reid—“remov[ing] her clothing and\\nha[ving] vaginal intercourse with her.” Id. Spreitz also\\nrecounted that he hit Reid in the head multiple times with a\\nrock “to make her stop yelling.” Id. He explained that he\\nleft Reid without knowing whether she was alive or dead.\\nId.\\n\\n Shortly after leaving Reid in the desert, Spreitz was\\nstopped by a Tucson police department officer. Id. at 1264.\\n\\n 5\\n Our recitation of the facts is adopted from the Arizona Supreme\\nCourt’s opinion affirming Spreitz’s conviction and death sentence.\\n\\n8 SPREITZ V. RYAN\\n\\nThe officer observed that Spreitz had a ripped shirt, smelled\\nof feces, and appeared to be covered in blood and fecal\\nmatter. Id. In addition, when detectives later searched\\nSpreitz’s car, they found blood spatter in the trunk, some of\\nwhich was inconsistent with Spreitz’s blood characteristics.\\nId. at 1265.\\n\\n On Monday morning, May 22, Reid’s naked and\\ndecomposing body was discovered on the outskirts of\\nTucson. Id. At trial, “the medical examiner testified that,\\ndue to the advanced state of decomposition, he could not\\ndetermine the full extent and nature of [Reid]’s injuries.” Id.\\nEven so, he was able to observe “bruising on the legs, arms,\\nand back; bruising and abrasions on the buttocks; several\\nbroken ribs; internal bleeding; a broken jaw; several head\\nlacerations; and a skull fracture where the skull had been\\n‘shoved in.’” Id. The medical examiner concluded that Reid\\nhad been killed by “blunt-force trauma to the head.” Id.\\n\\n In addition to finding Reid’s body “[a]t the scene, police\\ndetectives observed tire tracks leading back to the pavement,\\noil stains in the dirt, footprints, and drag marks in the dirt\\nleading away from the body. They also found feces-stained\\npants, tennis shoes, socks, a used tampon, and a torn\\nbrassiere. Two blood-stained rocks lay next to the body.”\\nId. A few days later, police arrested Spreitz. Id.\\n\\n On June 2, 1989, a grand jury indicted Spreitz for first-\\ndegree murder, Ariz. Rev. Stat. Ann. §§ 13-1105, 13-703;\\nsexual assault, Ariz. Rev. Stat. Ann. §§ 13-406(A) & (B);\\nand kidnapping, Ariz. Rev. Stat. Ann. §§ 13-304(A)(3) &\\n(B). Spreitz I, 945 P.2d at 1265. After five years of pre-trial\\nproceedings mostly regarding the admissibility of DNA\\nevidence, a seven-day jury trial began on August 9, 1994.\\nId. at 1266. After the conclusion of the trial, the jury\\nreturned guilty verdicts on all three counts: first-degree\\n\\n SPREITZ V. RYAN 9\\n\\nmurder (both premeditated and felony murder), sexual\\nassault, and kidnapping. Id.\\n\\n Prior to both his aggravation-mitigation and sentencing\\nhearings before the trial judge, Spreitz submitted evidence\\nand a memorandum in support of certain mitigating\\ncircumstances. As noted earlier, at the time of Spreitz’s\\nsentencing, Arizona’s death penalty statutes provided a list\\nof five specific mitigating factors; Arizona case law\\nrecognized nonstatutory mitigating factors as well. See Ariz.\\nRev. Stat. Ann. § 13-703(G); McKinney, 813 F.3d at 802.\\nSpreitz argued as nonstatutory mitigating factors: “(1) his\\ndysfunctional family life and lack of socialization; (2) a\\nhistory of alcohol and drug abuse; (3) his expressions of\\nremorse; [(4)] his good behavior while incarcerated; [(5)] his\\nlack of adult convictions; [and (6)] no prior record of violent\\ntendencies.” Spreitz I, 945 P.2d at 1279. Spreitz argued as\\nstatutory mitigating factors: (1) his age at the time of the\\nmurder, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and (2) that\\nhis “capacity to appreciate the wrongfulness of his conduct\\nor to conform his conduct to the requirements of law was\\nsignificantly impaired [due to alcohol use], but not so\\nimpaired as to constitute a defense to prosecution,” id.\\n§ 13-703(G)(1); see Spreitz I, 945 P.2d at 1279.\\n\\n Spreitz provided evidence of and argued for all the\\nforegoing mitigating circumstances but focused heavily on a\\ncombination of his relationship with his mother and his long\\nhistory of alcohol and substance abuse. To that end, Spreitz\\nsubmitted a written report by and presented testimony from\\nan examining psychologist, Dr. Todd Flynn, Ph.D. After\\nconducting interviews and research, Dr. Flynn concluded\\nthat Spreitz’s longstanding alcohol and substance abuse\\nshould be considered as both a statutory and nonstatutory\\nmitigating factor. In his report, which was admitted into\\n\\n10 SPREITZ V. RYAN\\n\\nevidence at the aggravation and mitigation hearing, Dr.\\nFlynn repeatedly emphasized Spreitz’s longstanding\\nsubstance abuse 6:\\n\\n By age twelve or thirteen, Chris Spreitz\\n began drinking alcohol and smoking\\n marijuana. By age 15, he drank steadily on\\n weekends and would have a shot of vodka\\n before school.\\n\\n The collateral information shows that the\\n alcohol abuse continued to intensify after he\\n left home. A variety of persons . . . described\\n him as a heavy drinker. This includes a\\n second cousin, Scott [Jouett], who saw him to\\n be intoxicated, “a majority of the time,” when\\n he was visiting Santa Barbara a week before\\n the current offense. To the interviewing\\n investigator, Mr. [Jouett] also described,\\n “several different occasions when Chris has\\n blackouts,” while drinking alcohol.\\n\\n It appears completely clear from the available\\n information that Chris Spreitz had a long-\\n standing problem with alcohol which\\n probably reached the level of physical\\n dependence. He described himself as\\n drinking in the morning as early as age 15.\\n Virtually everyone else who spent much time\\n with him described him as a heavy drinker.\\n\\n\\n\\n 6\\n Dr. Flynn testified consistently with his report. For ease of\\nreference, we refer only to his report.\\n\\n SPREITZ V. RYAN 11\\n\\nUltimately, Dr. Flynn concluded that Spreitz’s alcohol\\nabuse, childhood home life, and stunted development,\\ncombined with rejection by his girlfriend on the day of the\\ncrime, and intoxication at the time of the crime, led to the\\nmurder.\\n\\n At the end of his report, Dr. Flynn summarized his\\nfindings and opinions with respect to both statutory and\\nnonstatutory mitigating factors. He opined that both\\nSpreitz’s age, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and\\nimpaired capacity to appreciate the wrongfulness of his\\nconduct or conform his conduct to the law, Ariz. Rev. Stat.\\nAnn. § 13-703(G)(1), were mitigating statutory factors. As\\nto § 13-703(G)(5), specifically, Dr. Flynn opined that a\\ncombination of a disturbing upbringing in a “pathogenic,\\nemotionally neglectful home environment,” and “[y]ears of\\nalcoholism intoxication” combined to cause major deficits in\\nSpreitz’s social and emotional development. As to\\n§ 13-703(G)(1), Dr. Flynn offered a similar conclusion. In\\nlight of Spreitz’s “history of alcoholism . . . , a significant\\nbut unknown degree of alcohol intoxication is likely” on the\\nnight of the crime. In addition, Spreitz’s “history strongly\\nsuggests years of early experiences likely to have caused a\\nbuild-up of pent-up angry, aggressive feelings toward\\nwomen generally (and older women especially) which may\\nhave burst forth with uncontrollable intensity with or without\\nalcohol intoxication.” Dr. Flynn concluded that Spreitz’s\\nintoxication on the night of the crime coupled with his early\\nchildhood experiences “likely . . . contributed to an\\nuncontrollable outburst of aggression” and inability to\\ncontrol his conduct.\\n\\n Dr. Flynn also concluded that certain nonstatutory\\nmitigating factors were present. The nonstatutory mitigating\\nfactors included, in his opinion: Spreitz’s low potential for\\n\\n12 SPREITZ V. RYAN\\n\\nfuture violence, the “failure of [Spreitz’s] parents to provide\\ntreatment for alcohol abuse in [his] teenage years,” and the\\n“emotionally deprived, physically punitive home\\nenvironment” of Spreitz’s upbringing. Dr. Flynn also\\nemphasized that even if both statutory mitigating factors\\nfailed to satisfy the statutory threshold, they may still “be\\nappropriately considered . . . as nonstatutory mitigation.”\\n\\n On November 28, 1994, the trial court conducted an\\naggravation-mitigation hearing. At the hearing, Spreitz\\ncalled three mitigation witnesses: Dr. Flynn and two\\ncorrectional officers from the Pima County jail, where\\nSpreitz was incarcerated. The State did not offer any\\nwitnesses or evidence at the hearing. Dr. Flynn testified\\nconsistently with his report, as detailed above.\\n\\n On the same date, a probation officer filed a Pre-\\nSentence Report, which concluded:\\n\\n It appears the defendant became involved in\\n the senseless commission of the instant\\n offense due to his alcohol and drug abuse.\\n After five years in custody, he now admits his\\n substance abuse problem; however, this does\\n not condone his involvement in the offense.\\n It is unfortunate the victim died before the\\n defendant had his revelation.\\n\\nPrior to the sentencing hearing, Spreitz also submitted\\nseveral letters from friends, family, and jail personnel.\\n\\n On December 21, 1994, the trial court conducted a\\nsentencing hearing. Spreitz, Spreitz’s counsel, the victim’s\\n\\n SPREITZ V. RYAN 13\\n\\nsister, and the prosecutor each addressed the court. 7 The\\nprosecutor first disputed the mitigating evidence presented\\nby Spreitz at the mitigation-aggravation hearing. With\\nrespect to Spreitz’s alcohol abuse, the prosecutor argued that\\nSpreitz’s intoxication at the time of the crime did not meet\\nthe statutory definition for mitigation under § 13-703(G)(1)\\nbecause his conduct both during the murder and afterwards\\ndemonstrated that “he knew what he was doing.” The\\nprosecutor emphasized that the evidence revealed that Reid\\nhad been forcibly abducted and stuffed into the trunk of a\\ncar, thus forcing her to spend time “contemplat[ing] the\\nuncertainty of her fate.” The prosecutor argued that the\\nevidence revealed signs of struggle and that the serious\\ninjuries to Reid’s body belied any notion that Spreitz did not\\nknow what he was doing. The prosecutor further argued that\\nthe fact that Reid defecated on herself revealed that she was\\nterrified prior to her murder. In sum, the prosecutor argued\\nthat given both the heinous nature of the crime and the\\nmanner in which Reid suffered, a finding in aggravation that\\nReid was murdered in an “especially cruel manner” was\\nwarranted.\\n\\n After a short recess, the sentencing judge rendered oral\\nfindings addressing the aggravation and mitigation issues.\\nHe then imposed a sentence of death. 8 Following the\\n\\n\\n\\n 7\\n The trial court did not entertain argument at the conclusion of the\\naggravation-mitigation hearing. As a result, counsel addressed the\\naggravation-mitigation evidence and sentence options at the subsequent\\nsentencing hearing.\\n 8\\n Ring v. Arizona, 536 U.S. 548 (2012), the Supreme Court decision\\nholding judge-sentencing in capital cases unconstitutional, had yet to be\\ndecided.\\n\\n14 SPREITZ V. RYAN\\n\\nhearing, the sentencing judge filed a judgment setting forth\\nwritten findings.\\n\\n The judge first found one aggravating circumstance—\\nthat the offense was committed in an “especially cruel\\nmanner.” Ariz. Rev. Stat. Ann. § 13-703(F)(6). The judge\\nproceeded to review the “many factors [that were] submitted\\nin mitigation,” ultimately finding five mitigating\\ncircumstances of various levels of significance:\\n\\n [1] A mitigating circumstance defense\\n submitted was that the defendant had an\\n extremely disruptive childhood . . . . The\\n court finds the home was sub-normal, not\\n even a minimally healthy one for developing\\n children; and it is obvious the defendant\\n suffered a disruptive middle childhood—had\\n a punitive, controlling, cold mother, who he\\n could not please, no matter what he did.\\n\\n The defendant in his life turned to substance\\n abuse—alcohol and some suggestion he was\\n using cocaine and other drugs. However, the\\n court does not find such is a mitigating\\n circumstance that impaired his ability to\\n make a judgment on whether he was acting\\n rightfully or wrongfully in the death of the\\n victim.\\n\\n The defendant’s history of intoxication is\\n longstanding. He had been abusing\\n substances for close to ten years of his life at\\n the time of this offense when he was twenty-\\n two. Again, the court does not believe that\\n the substance abuse or intoxication impaired\\n\\n SPREITZ V. RYAN 15\\n\\n the defendant’s ability and capacity to\\n appreciate the wrongfulness of his conduct to\\n any significant degree. . . . The court does not\\n believe intoxication is any sort of mitigating\\n circumstance.\\n\\n ...\\n\\n [2] The court acknowledges that the\\n defendant has begun a process of\\n improvement and emotional growth while\\n confined at the Pima County Jail, where he\\n has taken part in education programs.\\n Correctional officers have testified he was a\\n prisoner who caused no problems.\\n\\n [3] The age of the defendant at the time of the\\n offense (twenty-two) is not a mitigating\\n circumstance in and of itself. Immaturity\\n probably is, but the court does not believe\\n immaturity was a significant mitigating\\n circumstance.\\n\\n [4] The court finds that Mr. Spreitz has no\\n criminal history of a felony nature—there is\\n no history or propensity for acts of violence.\\n\\n [5] The court believes the defendant is\\n capable of being rehabilitated. The court\\n does not know whether he has a good\\n prognosis for the future, but the court\\n believes he can be rehabilitated.\\n\\nIn conclusion, the judge found that “the mitigating\\ncircumstances [were] not sufficient to balance the\\n\\n16 SPREITZ V. RYAN\\n\\naggravating circumstances, nor [were] they sufficiently\\nsubstantial to call for leniency.” He thus imposed a death\\nsentence as to the first-degree murder conviction.\\n\\n B. Spreitz’s Appeal and Post-Conviction Proceedings\\n\\n Spreitz appealed his conviction and sentence to the\\nArizona Supreme Court with the assistance of new counsel.\\nThe Arizona Supreme Court affirmed Spreitz’s conviction\\nand sentence after conducting its own independent review of\\nthe record. 9 In reviewing the mitigating evidence, the court\\nfound four of the five mitigating circumstances that the\\nsentencing judge had found, declining to find Spreitz’s good\\nbehavior while in jail awaiting sentencing as a mitigating\\ncircumstance. Spreitz I, 945 P.2d at 1280–81. The court\\nexplained:\\n\\n [1] “We agree with the sentencing judge that\\n defendant’s upbringing was subnormal.” Id.\\n at 1280.\\n\\n [2] “We also find that the sentencing judge\\n . . . properly found that his emotional\\n immaturity was not a significant mitigating\\n factor.” Id. at 1281.\\n\\n [3] “We agree that the record supports the\\n sentencing judge’s findings that defendant\\n had no previous adult felony convictions, no\\n prior acts of violence, and\\n\\n\\n 9\\n See note 4, supra. Because the Arizona Supreme Court conducted\\na de novo review, we focus on its analysis, rather than, as the dissent\\ndoes in part, the analysis by the trial court.\\n\\n SPREITZ V. RYAN 17\\n\\n [4] that the defendant is capable of\\n rehabilitation.” Id.\\n\\nThe Arizona Supreme Court also added that it considered\\nremorse as an additional nonstatutory mitigating factor. Id.\\n\\n The court addressed Spreitz’s history of alcohol and\\nsubstance abuse as follows:\\n\\n The record demonstrates defendant’s\\n longtime substance abuse problems. We\\n note, however, that defendant’s general\\n problems with substance abuse are not\\n essential to our decision here. We therefore\\n decline to conclude that defendant was\\n impaired by alcohol consumption to an extent\\n that it interfered with his “capacity to\\n appreciate the wrongfulness of his conduct or\\n to conform his conduct to the requirements of\\n the law.” A[riz]. R[ev]. S[tat]. [Ann.]\\n § 13-703(G)(1); see also State v. Medrano,\\n 185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)\\n (citing [State v.] Stokley, 182 Ariz. [505,]\\n 520, 898 P.2d [454,] 469 [1995]).\\n\\nId. at 1280–81. In the only other portion of its opinion\\naddressing Spreitz’s history of alcohol and substance abuse,\\nthe court said:\\n\\n The sentencing judge found that defendant’s\\n ability to appreciate the wrongfulness of his\\n conduct was not impaired on the night of the\\n murder to any significant extent by substance\\n abuse, emotional disorders, situational stress,\\n or by a combination of these. Our review of\\n\\n18 SPREITZ V. RYAN\\n\\n the record convinces us that the trial court’s\\n finding was proper.\\n\\nId. at 1281.\\n\\n The Arizona Supreme Court agreed with the trial court’s\\naggravation finding that the murder was done in an\\nespecially cruel manner, id. at 1279, and then proceeded to\\nreweigh the applicable aggravating and mitigating factors,\\nfinding that “the aggravating circumstance of especial\\ncruelty in defendant’s murder of Ruby Reid outweigh[ed] all\\nfactors mitigating in favor of leniency.” Id. at 1282. It\\naffirmed his death sentence. Id. at 1283. Spreitz filed a\\npetition for a writ of certiorari, which the United States\\nSupreme Court denied. Spreitz v. Arizona, 523 U.S. 1027\\n(1998) (mem).\\n\\n In March 2000, represented again by new counsel,\\nSpreitz filed a petition for post-conviction relief in the\\nArizona Superior Court (“PCR court”). See Ariz. R. Crim.\\nP. 32.1. Spreitz alleged, inter alia, that the sentencing judge\\nand the Arizona Supreme Court committed error by failing\\nto consider his history of alcohol and substance abuse as a\\nnonstatutory mitigating factor apart from its causal\\nconnection to the murder—i.e., Eddings error. He also\\nalleged that his appellate counsel was ineffective for failing\\nto raise the sentencing judge’s Eddings error on appeal to the\\nArizona Supreme Court. Spreitz additionally raised several\\nnew claims of trial-counsel-ineffectiveness.\\n\\n The PCR court denied all of Spreitz’s claims and entered\\nan order dismissing his petition for post-conviction relief. In\\nPart 3 of its order, the PCR court discussed Spreitz’s claims\\nof “nexus-error”—that both the trial court and the Arizona\\nSupreme Court erred when they failed to consider Spreitz’s\\nlongstanding alcohol and substance abuse on the basis that\\n\\n SPREITZ V. RYAN 19\\n\\nSpreitz had failed to establish a causal nexus between the\\nlong-term substance abuse and the murder. 10 The PCR court\\ndismissed the claim 11 as waived, reasoning that Spreitz had\\nfailed to raise the issue on direct appeal. Nonetheless, the\\nPCR court addressed the merits of the nexus-error claim in\\nthe course of analyzing Spreitz’s argument that his appellate\\ncounsel was ineffective for failing to raise it. In doing so,\\nthe PCR court explained that the claim failed because:\\n\\n [I]t must be demonstrated, under A[riz].\\n R[ev]. S[tat]. [Ann.] § 13-703(G)(l), that\\n there is a causal link between the history of\\n alcohol or substance abuse and the offense\\n itself. E.g., State v. Stokley, [] 182 Ariz.\\n [505,] 523 [898 P.2d 454, 472 (Ariz. 1995)].\\n Without some basis for explaining or\\n defining the individual’s behavior at the time\\n of the offense, the Petitioner’s history of\\n alcohol or substance abuse would be\\n inconsequential (which is exactly what the\\n trial court and Supreme Court concluded).\\n State v. Kayer, 194 Ariz. 423, 984 P.2d 81\\n ([Ariz.] 1999).\\n\\n At times, the court can and should consider\\n an individual’s long-term alcoholism and\\n\\n 10\\n For ease of reference, we refer to the Arizona courts’ alleged error\\nas “nexus-error.”\\n\\n 11\\n Although Spreitz raised, and the PCR court recognized, two\\ndistinct claims—one with respect to the sentencing court and one with\\nrespect to the Arizona Supreme Court—the PCR court’s analysis treats\\nthe claims as one and the same and refers to them in the singular. To\\nreflect accurately the PCR court’s discussion, our summary here likewise\\nrefers to a singular “claim.”\\n\\n20 SPREITZ V. RYAN\\n\\n substance abuse, usually in conjunction with\\n other factors or diagnosis, as non-statutory\\n mitigation. However, the impact or effect of\\n the alcoholism or substance abuse must be\\n substantial and of such severity that it\\n provides a sufficient basis for explaining the\\n defendant’s conduct, character or ability to\\n control his behavior at the time of the offense.\\n [Citations omitted].\\n\\n As previously discussed, there is no evidence\\n in Petitioner’s case to suggest that he suffered\\n any long-term effects from his alcohol or\\n drug abuse that precluded him from\\n controlling his behavior. Petitioner did not\\n suffer from any cognitive or emotional\\n deficits that rendered him incapable of\\n controlling his conduct. Therefore, the trial\\n court did not err in failing to find Petitioner’s\\n history of alcohol or substance abuse as a\\n separate, non-statutory mitigating factor.\\n [Citation omitted].\\n\\n Spreitz filed a petition for review in the Arizona\\nSupreme Court challenging the PCR court’s judgment. The\\nArizona Supreme Court summarily affirmed the PCR court’s\\nmerits determination with respect to Spreitz’s Eddings\\nclaims. State v. Spreitz (Spreitz II), 39 P.3d 525, 527 (Ariz.\\n2002).\\n\\n Spreitz filed his federal habeas petition in February\\n2003. He alleged in claim seven that the sentencing judge\\nand the Arizona Supreme Court had both committed nexus-\\nerror with respect to his long-time alcohol and substance\\nabuse, and that appellate counsel was ineffective for failing\\n\\n SPREITZ V. RYAN 21\\n\\nto argue the sentencing judge’s nexus-error on direct appeal\\nto the Arizona Supreme Court.\\n\\n The district court denied all of Spreitz’s claims. With\\nrespect to the alleged nexus-error by both the sentencing\\ncourt and the Arizona Supreme Court, the district court\\nfound the claims procedurally barred, relying on the PCR\\ncourt’s determination that Spreitz had waived the claims\\nbecause he could have raised them on direct appeal but failed\\nto do so. The district court concluded that Spreitz had\\nproperly exhausted his ineffective assistance of appellate\\ncounsel claim, and thus addressed the alleged nexus-error in\\nthat context, ultimately concluding it was not meritorious.\\nAccordingly, the district court concluded that “appellate\\ncounsel’s failure to raise this issue on appeal does not\\nconstitute ineffectiveness.”\\n\\n Spreitz timely appealed.\\n\\n II. Jurisdiction and Standard of Review\\n\\n We have jurisdiction pursuant to 28 U.S.C. §§ 1291\\nand 2253(a), and review de novo the district court’s denial\\nof a writ of habeas corpus. Poyson v. Ryan, 879 F.3d 875,\\n887 (9th Cir. 2018). Because Spreitz filed his federal habeas\\npetition after April 24, 1996, he must satisfy the\\nAntiterrorism and Effective Death Penalty Act (“AEDPA”).\\nUnder AEDPA, we may not grant habeas relief unless the\\nstate’s adjudication of Spreitz’s claim (1) “was contrary to\\n. . . clearly established federal law[] as determined by the\\nSupreme Court,” (2) “involved an unreasonable application\\nof” such law, or (3) “was based on an unreasonable\\ndetermination of the facts in light of the evidence presented\\nin the State court proceeding.” (28 U.S.C. § 2254(d)). “In\\nmaking this determination, we look to the last state court\\ndecision to address the claim,” White v. Ryan, 895 F.3d 641,\\n\\n22 SPREITZ V. RYAN\\n\\n665 (9th Cir. 2018) (citing Wilson v. Sellers, 138 S. Ct. 1188,\\n1192 (2018)), which for Spreitz’s nexus-error claim is that\\nof the PCR court.\\n\\n Spreitz argues that the PCR court’s decision was\\n“contrary to” Eddings v. Oklahoma, 455 U.S. 104 (1982).\\n“A state court’s decision is contrary to clearly established\\nfederal law if it ‘applies a rule that contradicts the governing\\nlaw set forth in [U.S. Supreme Court] cases’ or arrives at a\\ndifferent result in a case that ‘confronts a set of facts that are\\nmaterially indistinguishable from a decision of [the\\nSupreme] Court.’” Castellanos v. Small, 766 F.3d 1137,\\n1146 (9th Cir. 2014) (alteration in original) (quoting\\nWilliams v. Taylor, 529 U.S. 362, 405–06 (2000)). “If the\\nstate court applies a legal standard that contradicts clearly\\nestablished [Supreme Court] law, we review de novo the\\napplicant’s claims, applying the correct legal standard to\\ndetermine whether the applicant is entitled to relief.” Id.\\n(citation omitted).\\n\\n III. Discussion\\n\\n As discussed supra, the precise question we must decide\\nis whether the Arizona Supreme Court applied its\\nunconstitutional causal nexus test in violation of Eddings\\nwhen it affirmed Spreitz’s death sentence. To answer that\\nquestion, we must first determine whether the claim is\\nproperly before us. After concluding that Spreitz’s claim\\ncould not have been procedurally defaulted, we turn to the\\nlevel of deference we must accord the PCR court’s ruling on\\nSpreitz’s Eddings claim under AEDPA. Because the PCR\\ncourt’s decision was contrary to clearly established Supreme\\nCourt precedent, we accord that decision no deference and\\nreview Spreitz’s Eddings claim de novo. We conclude that\\nthe Arizona Supreme Court violated Eddings by\\nimpermissibly requiring that Spreitz establish a causal\\n\\n SPREITZ V. RYAN 23\\n\\nconnection between his longstanding substance abuse and\\nthe murder before considering and weighing the evidence as\\na nonstatutory mitigating factor. Finally, having determined\\nthat the Arizona Supreme Court applied an unconstitutional\\ncausal nexus test in its sentencing procedure, we turn to\\nwhether the error was harmless under Brecht v. Abrahamson,\\n507 U.S. 619 (1993). We conclude that the error was not\\nharmless and therefore reverse the district court’s judgment\\nwith respect to Spreitz’s sentence.\\n\\n A.\\n\\n Under the doctrine of procedural default, a federal\\nhabeas court “will not review the merits of claims, including\\nconstitutional claims, that a state court declined to hear\\nbecause the prisoner failed to abide by a state procedural\\nrule.” Martinez v. Ryan, 566 U.S. 1, 9 (2012). “We review\\nde novo a district court’s conclusion that a claim is\\nprocedurally defaulted.” Cooper v. Neven, 641 F.3d 322,\\n326 (9th Cir. 2011).\\n\\n Here, the district court concluded that both of Spreitz’s\\nEddings claims—that the sentencing court and the Arizona\\nSupreme Court applied an unconstitutional causal nexus\\ntest—were procedurally defaulted, because the PCR court\\nfound each claim waived for failure to raise them on direct\\nappeal. This conclusion, however, is erroneous with respect\\nto Spreitz’s claim that the Arizona Supreme Court violated\\nEddings. After all, “[t]he Arizona Supreme Court reviews\\ncapital sentences de novo, making its own determination of\\nwhat constitute[s] legally relevant aggravating and\\nmitigating factors, and then weighing those factors\\nindependently. Ariz. Rev. Stat. Ann. § 13-755. The Arizona\\nSupreme Court ‘conducts a thorough and independent\\nreview of the record and of the aggravating and mitigating\\nevidence to determine whether the sentence is justified.’”\\n\\n24 SPREITZ V. RYAN\\n\\nMcKinney, 813 F.3d at 819 (emphasis removed) (quoting\\nState v. McKinney, 917 P.2d 1214, 1225 (Ariz. 1996)).\\nSpreitz could not have raised on direct appeal his claim that\\nthe Arizona Supreme Court violated Eddings in performing\\nits de novo review of Spreitz’s death sentence. The first\\nopportunity he had to raise that claim was before the PCR\\ncourt, at which time he did so. Thus, Spreitz’s claim is not\\nprocedurally defaulted and is properly before us. 12\\n\\n B.\\n\\n Because Spreitz’s claim that the Arizona Supreme Court\\napplied an unconstitutional causal nexus test to his\\nnonstatutory mitigating evidence of long-time alcohol and\\n 12\\n The district court declined to issue a certificate of appealability\\n(“COA”) on Spreitz’s substantive Eddings claim, which he raised as\\nclaim 1.7 on pages 108 through 111 in his federal habeas petition.\\nAlthough Spreitz did not present his Eddings claim as a separate\\nuncertified issue, we may grant a COA regarding any uncertified issue\\ndiscussed in a petitioner’s opening brief. See 9th Cir. R. 22-1(e).\\nSpreitz’s opening brief discusses the Arizona courts’ Eddings errors. We\\ntherefore may exercise our discretion to expand the COA to encompass\\nthe Eddings claim.\\n\\n To obtain a COA under 28 U.S.C. § 2253(c), Spreitz must\\ndemonstrate that “reasonable jurists could debate whether . . . the\\npetition should have been resolved in a different manner or that the issues\\npresented were adequate to deserve encouragement to proceed further.”\\nSlack v. McDaniel, 529 U.S. 473, 484 (2000). Reasonable jurists could\\ndebate the merits of Spreitz’s Eddings claim. We thus exercise our\\ndiscretion and expand the COA to encompass Spreitz’s substantive\\nEddings claim with respect to the Arizona Supreme Court. We\\nacknowledge that under Ninth Circuit Rule 22-1(f), the State would\\nnormally be afforded an opportunity to submit supplemental briefing on\\nthe uncertified Eddings issue. Here, however, the Eddings issue has been\\nexhaustively briefed, in the context of Spreitz’s appellate counsel’s\\nalleged ineffectiveness before the Arizona Supreme Court. Additional\\nbriefing is unwarranted.\\n\\n SPREITZ V. RYAN 25\\n\\nsubstance abuse is properly before us, we next determine the\\nlevel of deference to accord the PCR court’s adjudication of\\nthat claim. As discussed above, although the PCR court first\\ndeclared the claim waived, it proceeded to adjudicate the\\nclaim on the merits. In denying the claim, the PCR court\\nexplained:\\n\\n Without some basis for explaining or\\n defining the individual’s behavior at the time\\n of the offense, [Spreitz]’s history of alcohol\\n or substance abuse would be inconsequential\\n (which is exactly what the trial court and\\n Supreme Court concluded).\\n\\n At times, the court can and should consider\\n an individual’s long-term alcoholism and\\n substance abuse, usually in conjunction with\\n other factors or diagnosis, as non-statutory\\n mitigation. However, the impact or effect of\\n the alcoholism or substance abuse must be\\n substantial and of such severity that it\\n provides a sufficient basis for explaining the\\n defendant’s conduct, character, or ability to\\n control his behavior at the time of the offense.\\n\\n ....\\n\\n As previously discussed, there is no evidence\\n in [Spreitz]’s case to suggest that he suffered\\n any long-term effects from his alcohol or\\n drug abuse that precluded him, from\\n controlling his behavior. Petitioner did not\\n suffer from any cognitive or emotional\\n deficits that rendered him incapable of\\n controlling his conduct. Therefore, the trial\\n\\n26 SPREITZ V. RYAN\\n\\n court did not err in failing to find [Spreitz]’s\\n history of alcohol or substance abuse as a\\n separate, non-statutory mitigating factor.\\n\\n In other words, the PCR court concluded that Spreitz’s\\nlongstanding alcohol and substance abuse could only be\\nconsidered a nonstatutory mitigating factor if it “provide[d]\\na sufficient basis for explaining the defendant’s conduct,\\ncharacter, or ability to control his behavior at the time of the\\noffense.” This discussion both accurately described the\\ncausal nexus test and approved of it.\\n\\n Such reasoning is contrary to clearly established federal\\nlaw. As explained in McKinney, “the causal nexus test\\nclearly violates Eddings.” 813 F.3d at 810. Therefore,\\n“[b]ecause the state court used the wrong standard, we need\\nnot defer to that decision.” Hardy v. Chappell, 849 F.3d 803,\\n820 (9th Cir. 2016).\\n\\n C.\\n\\n Because the PCR court’s decision was “contrary to”\\nclearly established law, we review de novo the merits of\\nSpreitz’s Eddings claim. 28 U.S.C. § 2254(d)(1). We\\ntherefore consider whether the Arizona Supreme Court\\napplied its unconstitutional causal nexus test in reviewing\\nand affirming Spreitz’s death sentence. In so doing, “we\\nlook only to the decision of th[e Arizona Supreme Court],\\n. . . only [considering] . . . the decision of the sentencing\\njudge . . . to the degree it was adopted or substantially\\nincorporated by the Arizona Supreme Court.” McKinney,\\n813 F.3d at 819.\\n\\n SPREITZ V. RYAN 27\\n\\n 1.\\n\\n Our decision in McKinney frames our consideration of\\nthe Arizona Supreme Court’s decision and Spreitz’s claim.\\nMcKinney makes clear that the Arizona Supreme Court\\nconsistently articulated and applied its unconstitutional\\ncausal nexus test during the period in which Spreitz’s death\\nsentence was litigated in the trial court and reviewed by the\\nArizona Supreme Court. Id. at 824. Moreover, McKinney\\nmakes plain that although on habeas review, we generally\\napply a “presumption that state courts know and follow the\\nlaw,” “the Arizona Supreme Court’s consistent articulation\\nand application of its causal nexus test . . . make such a\\ncourse impossible.” Id. at 803 (internal quotation marks and\\ncitation omitted). In appeals heard from the late 1980s until\\nat least 2002, see State v. Canez, 42 P.3d 564, 594 (Ariz.\\n2002), it was unmistakably clear that “the Arizona Supreme\\nCourt did not know and follow federal law.” Id. (internal\\nquotation marks omitted). 13 Therefore, just as in McKinney\\nand subsequent cases applying McKinney, see, e.g., Poyson,\\n879 F.3d at 889, “the presumption” that the Arizona\\nSupreme Court knew and followed the law “is rebutted” here\\nas well. McKinney, 813 F.3d at 804.\\n\\n That said, McKinney does not dispose of Spreitz’s claim.\\nWe must still review the Arizona Supreme Court’s decision\\nto determine whether it did, in fact, apply its unconstitutional\\ncausal nexus test in Spreitz’s case. In doing so, we are\\nmindful of the fact that at the time the Arizona Supreme\\nCourt decided Spreitz’s appeal, “if there is to be a\\npresumption, it is that the Arizona Supreme Court violated\\n\\n\\n 13\\n In 2005, the Arizona Supreme Court repudiated its earlier nexus\\ntest. See State v. Anderson, 111 P.3d 369, 391–92 (Ariz. 2005).\\n\\n28 SPREITZ V. RYAN\\n\\nthe dictates of Lockett and Eddings.” Greenway v. Ryan,\\n866 F.3d 1094, 1095 (9th Cir. 2017).\\n\\n 2.\\n\\n Against this backdrop, we are convinced that the Arizona\\nSupreme Court applied its causal nexus test with respect to\\nSpreitz’s evidence of longstanding alcohol and substance\\nabuse in violation of Eddings.\\n\\n We quote again the Arizona Supreme Court’s discussion\\nof Spreitz’s history of alcohol and substance abuse:\\n\\n The record demonstrates defendant’s\\n longtime substance abuse problems. We\\n note, however, that defendant’s general\\n problems with substance abuse are not\\n essential to our decision here. We therefore\\n decline to conclude that defendant was\\n impaired by alcohol consumption to an extent\\n that it interfered with his “capacity to\\n appreciate the wrongfulness of his conduct or\\n to conform his conduct to the requirements of\\n the law.” A[riz]. R[ev]. S[tat]. [Ann.] § 13-\\n 703(G)(1); see also State v. Medrano,\\n 185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)\\n (citing Stokley, 182 Ariz. at 520, 898 P.2d at\\n 469).\\n\\n ....\\n\\n The sentencing judge found that defendant’s\\n ability to appreciate the wrongfulness of his\\n conduct was not impaired on the night of the\\n murder to any significant extent by substance\\n abuse, emotional disorders, situational stress,\\n\\n SPREITZ V. RYAN 29\\n\\n or by a combination of these. Our review of\\n the record convinces us that the trial court’s\\n finding was proper.\\n\\nSpreitz I, 945 P.2d at 1280–81. This discussion\\ndemonstrates that the Arizona Supreme Court primarily\\nconcluded that Spreitz failed to show statutory mitigation\\nunder Ariz. Rev. Stat. Ann. § 13-703(G)(1) based on his\\nintoxication on the night of the murder. “When applied\\nsolely in the context of statutory mitigation under\\n§ 13-703(G)(1), the causal nexus test does not violate\\nEddings.” McKinney, 813 F.3d at 810.\\n\\n The plain text of the Arizona Supreme Court’s decision,\\nhowever, also clearly demonstrates that the court did not\\nsolely apply its causal nexus test to Spreitz’s evidence of\\nstatutory mitigation. The court initially recognized Spreitz’s\\n“general problems with substance abuse”—evidence that\\nshould have been relevant as nonstatutory mitigation—but\\nconcluded those problems were “not essential to [its]\\ndecision” because they did not show Spreitz was “impaired\\non the night of the murder.” Spreitz I, 945 P.2d at 1280–81.\\nIn other words, the court did not acknowledge the relevance\\nof longtime substance abuse as nonstatutory mitigation in the\\nabsence of a causal relationship to the crime. Instead, the\\ncourt held that because there was no causal relationship,\\nSpreitz’s long-term alcohol abuse was not of significance—\\ni.e. “not essential”—to the court’s overall determination of\\neither statutory or nonstatutory mitigation.\\n\\n The Supreme Court, however, has been clear that “‘full\\nconsideration of evidence that mitigates against the death\\npenalty is essential if the sentencer is to give a reasoned\\nmoral response to the defendant’s background, character,\\nand crime.’” McKinney, 813 F.3d at 812 (first emphasis\\n\\n30 SPREITZ V. RYAN\\n\\nadded) (quoting Penry I, 492 U.S. at 319, 328).\\nAccordingly, “Eddings requires that ‘[t]he sentencer must\\nalso be able to consider and give effect’” to any relevant\\nmitigation evidence proffered by the defendant, which\\nincludes evidence going toward nonstatutory mitigation. Id.\\n(quoting Penry I, 492 U.S. at 319, 328). The Arizona\\nSupreme Court’s refusal to consider Spreitz’s alcohol and\\nsubstance abuse beyond its connection, if any, to Reid’s\\nmurder constitutes application of its unconstitutional causal\\nnexus test to relevant nonstatutory mitigation evidence.\\n\\n Our understanding of the Arizona Supreme Court’s\\nopinion is bolstered by the fact that in its discussion of\\nSpreitz’s longstanding alcohol and substance abuse, the\\ncourt cited to a prior decision that likewise applied the\\nunconstitutional causal nexus test, State v. Medrano,\\n914 P.2d 255 (Ariz. 1996). In McKinney, we explained that\\nin Medrano, the Arizona Supreme Court upheld the\\nsentencing judge’s determination that the defendant’s\\ncocaine use failed as a nonstatutory mitigating circumstance\\nbecause he did not show that it contributed to his conduct on\\nthe night of the murder. McKinney, 813 F.3d at 825–26. In\\nso doing, “[t]he Arizona Supreme Court applied the causal\\nnexus test.” Id. at 825. Although the Arizona Supreme\\nCourt’s citation to Medrano is not dispositive, it certainly\\ncorroborates our understanding of the court’s opinion. See\\nalso Poyson, 879 F.3d at 890 (emphasizing, among other\\nconsiderations, that the Arizona Supreme Court cited a\\npassage from a case specifically identified in McKinney as\\napplying the unconstitutional causal nexus test in order to\\nconclude that the state court also applied that test to Poyson’s\\nevidence). Moreover, as we explained in McKinney, “the\\nArizona Supreme Court applied its unconstitutional nexus\\ntest consistently” for fifteen years because it had “a mistaken\\nunderstanding of Eddings.” 813 F.3d at 826. It is thus\\n\\n SPREITZ V. RYAN 31\\n\\nentirely logical that in citing the exact portion of Medrano\\nwhere it previously applied its causal nexus test, the Arizona\\nSupreme Court was applying the same test in Spreitz’s case.\\nSee id. at 821; see also Poyson, 879 F.3d at 890.\\n\\n We find further support for our understanding of the\\nArizona Supreme Court’s unconstitutional treatment of\\nSpreitz’s alcohol and substance abuse when we compare it\\nwith the manner in which the court discussed other\\nmitigating factors. As detailed earlier, in evaluating the\\nevidence of Spreitz’s general history of alcohol and\\nsubstance abuse, the court stated that this evidence was “not\\nessential” to its decision. Spreitz I, 945 P.2d at 1280. By\\ncontrast, when the court considered other mitigating factors,\\nit specifically discussed the weight it would give those\\nfactors, as opposed to simply dismissing them outright. For\\nexample, when discussing Spreitz’s subnormal upbringing,\\nthe court stated, “Although we recognize defendant’s\\nupbringing as a mitigating circumstance, we accord it little\\nweight.” Id. at 1280 (emphasis added). As another example,\\nafter “finding that since his arrest, [Spreitz] has\\ndemonstrated remorse,” the court stated that his “remorse for\\nhis actions does little to counterbalance especial cruelty as a\\nserious aggravating circumstance.” Id. at 1281 (emphasis\\nadded). And finally, when discussing the sentencing judge’s\\ndetermination regarding Spreitz’s emotional immaturity, the\\ncourt stated, “We also find that the sentencing judge . . .\\nproperly found that his emotional immaturity was not a\\nsignificant mitigating factor.” Id. (emphasis added). In\\nother words, when the Arizona Supreme Court wanted to\\nassign weight to a given factor, it said so. This pattern stands\\nin stark contrast to the court’s conclusion that anything\\nrelated to Spreitz’s long-term alcohol or substance abuse that\\nfell short of the statutory definition could not serve as\\nmitigation. Id. at 1280–81.\\n\\n32 SPREITZ V. RYAN\\n\\n Finally, as also discussed earlier, the PCR court\\ndistinctly interpreted the Arizona Supreme Court’s decision\\nas having applied a causal nexus test. The PCR court\\nexplained that “[w]ithout some basis for explaining or\\ndefining the individual’s behavior at the time of the offense,\\n[Spreitz]’s history of alcohol or substance abuse would be\\ninconsequential (which is exactly what the trial court and\\nSupreme Court concluded).” Moreover, in articulating how\\nthe Arizona Supreme Court had applied its causal nexus test\\nin Spreitz’s direct appeal, the PCR court, as had the Arizona\\nSupreme Court, cited to a decision of the Arizona Supreme\\nCourt indisputably applying an unconstitutional nexus test,\\nState v. Kayer, 984 P.2d 31 (Ariz. 1999). See McKinney,\\n813 F.3d at 816 (explaining that Kayer held “that the\\ndefendant’s mental impairment ‘was not established as a\\nnonstatutory mitigating factor’ in part because ‘defendant\\noffered no evidence to show the requisite causal nexus that\\nmental impairment affected his judgment or his actions at the\\ntime of the murder’” (quoting Kayer, 984 P.2d at 46)). The\\nfact that the PCR court interpreted the Arizona Supreme\\nCourt’s opinion as we do—and then itself applied the\\nconstitutionally erroneous nexus test—lends further support\\nto our reading.\\n\\n We recognize that the Arizona Supreme Court’s decision\\non the particular issue before us was framed by language that\\nmight, in a decisional vacuum, suggest the court knew,\\nunderstood, and applied the law with respect to Eddings.\\nThe court acknowledged that Spreitz argued that his history\\nof alcohol and drug abuse served as both a statutory and\\nnonstatutory mitigating factor. Spreitz I, 945 P.2d at 1279.\\nIt then stated that it “must consider any aspect of the\\ndefendant’s character or record . . . relevant to determining\\nwhether the death penalty should be imposed.” Id. And\\nfinally, in reaching the conclusion to affirm Spreitz’s death\\n\\n SPREITZ V. RYAN 33\\n\\nsentence, the court explained that it had “examin[ed] the\\nentire record and reweigh[ed] the applicable aggravating and\\nmitigating factors.” Id. at 1281–82.\\n\\n The Arizona Supreme Court’s acknowledgment of what\\nSpreitz argued does not indicate that it relied on or accepted\\nthat argument as correct. Further, we cannot, given\\nMcKinney, Poyson, and Greenway, proceed in a decisional\\nvacuum.\\n\\n Considered in light of the baseline that the Arizona\\nSupreme Court was consistently applying its\\nunconstitutional causal nexus test in death penalty cases at\\nthe time Spreitz’s appeal was decided, as well as its citations\\nto cases held in McKinney to be applying that standard, we\\nare confident that the manner in which the Arizona Supreme\\nCourt disposed of Spreitz’s evidence of his longstanding\\nalcohol and substance abuse confirms that it applied the\\ncausal nexus test it was using during that period.\\n\\n Styers v. Schriro, 547 F.3d 1026 (9th Cir. 2008) supports\\nthis conclusion. Styers stated that the mere fact that “the\\nArizona Supreme Court stated that it had ‘considered all of\\nthe proffered mitigation,’” is not sufficient to contradict the\\nfact that “its analysis prior to this statement indicated\\notherwise.” Id. at 1035 (internal citation omitted). In Styers,\\nnotwithstanding assertions to the contrary, the Arizona\\nSupreme Court refused to consider evidence of the\\ndefendant’s Post Traumatic Stress Disorder (“PTSD”)\\nbecause he had failed to connect it causally to his crimes. Id.\\nHere, too, the Arizona Supreme Court’s actual analysis of\\nSpreitz’s longstanding alcohol and substance abuse reveals\\nthat it refused to consider this particular aspect of Spreitz’s\\ncharacter free from any connection to the crime. Rather, the\\ncourt’s discussion demonstrates that in evaluating Spreitz’s\\nlong-term alcohol and substance abuse, it applied a causal\\n\\n34 SPREITZ V. RYAN\\n\\nnexus test in violation of Eddings and therefore refused, as a\\nmatter of law, to consider it in assessing Spreitz’s evidence\\nof nonstatutory mitigating circumstances.\\n\\n To the extent the state argues the Arizona Supreme\\nCourt’s opinion is ambiguous as to whether it applied an\\nunconstitutional causal nexus test or not—and we believe\\nthat it is not—McKinney requires resolving that ambiguity\\nin favor of Spreitz. McKinney held it clear that the Arizona\\nSupreme Court “consistently articulated and applied its\\ncausal nexus test” at the time the trial court imposed\\nSpreitz’s death sentence and the Arizona Supreme Court\\nreviewed and affirmed that sentence. McKinney, 813 F.3d\\nat 803 (emphasis in original). Again, “if there is to be a\\npresumption, it is that the Arizona Supreme Court violated\\nthe dictates of Lockett and Eddings.” Greenway, 866 at\\n1095.\\n\\n In sum, because (1) the Arizona Supreme Court clearly\\nstated that Spreitz’s longstanding alcohol and substance\\nabuse was “not essential” to its decision, which was in line\\nwith the court’s consistent practice of applying a causal\\nnexus test; (2) the Arizona Supreme Court cited to Medrano,\\nin which it had previously articulated a causal nexus test;\\n(3) the Arizona Supreme Court carefully laid out the weight\\nit would accord other mitigating factors as opposed to simply\\nignoring them; and (4) the PCR court likewise interpreted\\nthe Arizona Supreme Court’s opinion as applying a causal\\nnexus test, we conclude that the Arizona Supreme Court\\nheld, as a matter of law, that Spreitz’s long-term alcohol and\\nsubstance abuse was not a nonstatutory mitigating factor and\\ntherefore refused to consider it. This holding was in clear\\nviolation of Eddings.\\n\\n SPREITZ V. RYAN 35\\n\\n 3.\\n\\n In light of our conclusion that the Arizona Supreme\\nCourt failed to comply with Eddings, we must next decide\\nwhether Spreitz was prejudiced by this error. On federal\\nhabeas review, relief is warranted if an error “had substantial\\nand injurious effect or influence” on the challenged decision.\\nSee Brecht v. Abrahamson, 507 U.S. 619, 623 (1993). The\\npetitioner is “not entitled to habeas relief” unless he can\\nestablish that the error “resulted in actual prejudice.” Davis\\nv. Ayala, 135 S. Ct. 2187, 2197 (2015) (internal quotation\\nmarks omitted). Here, because we are “in grave doubt about\\nwhether [the Arizona Supreme Court’s] error of federal law\\nhad substantial and injurious effect or influence in\\ndetermining [whether to affirm Spreitz’s death sentence],\\nthat error is not harmless. And, [Spreitz] must win.”\\nMcKinney, 813 F.3d at 822 (quoting O’Neal v. McAninch,\\n513 U.S. 432, 435 (1995)).\\n\\n We previously have required resentencing based on\\nprejudicial Eddings error where the Arizona Supreme Court\\nrefused to consider, as a matter of law, mitigation evidence\\n“central to” the petitioner’s “plea for leniency.” See id. at\\n823; see also Coleman v. Calderon, 210 F.3d 1047, 1051\\n(9th Cir. 2000) (concluding that a penalty-phase jury\\ninstruction error was not harmless under Brecht because “it\\nundermined the very core of Coleman’s plea for life”). 14 For\\nthe reasons that follow, it is clear that Spreitz’s long-term\\n\\n 14\\n We do not suggest that Brecht prejudice may only be established\\nif inappropriately considered mitigating evidence was at “the very core\\nof [a defendant’s] plea for life.” Coleman, 210 F.3d at 1051. It is\\nconceivable that other forms of Eddings error could have “a substantial\\nand injurious effect” on a defendant’s sentence if they would be\\nsignificant to the sentencer, despite the defendant’s failure to rely on\\nthem as central to his or her argument. Brecht, 507 U.S. at 623.\\n\\n36 SPREITZ V. RYAN\\n\\nalcohol and substance abuse was important mitigation\\nevidence, and was so closely related as to become central to\\nhis plea for leniency. The Arizona Supreme Court’s refusal\\nto consider that evidence had a substantial and injurious\\neffect on its sentencing determination that requires\\nresentencing.\\n\\n While a defendant’s evidence of longstanding alcohol\\nand/or substance abuse can be probative mitigation evidence\\non its own, a “history of substance abuse [is] substantially\\nmitigating when . . . combined with other mitigating\\nevidence.” Henry v. Ryan, 720 F.3d 1073, 1090 (9th Cir.\\n2013) (emphasis added) (listing cases and holding that the\\ndefendant’s alcoholism “would not have had a substantial\\nand injurious effect” because it “st[ood] alone, was similar\\nto evidence already considered by the sentencing courts, and\\nwas of limited probative value”) (footnote omitted). 15 Here,\\nSpreitz’s evidence regarding his history of alcohol and\\nsubstance abuse—spanning nearly half his life by the time\\nhe committed the crime at the age twenty-two—was the\\n\\n 15\\n The dissent makes much of Henry, suggesting it dictates the\\noutcome of our prejudice analysis. Dissent at 29. Not so. In that case,\\nthe state court had found that Henry’s intoxication on the day of the\\nmurder did constitute a statutory mitigating factor but was not\\nsufficiently substantial to outweigh the two aggravating factors. Henry,\\n720 F.3d at 1090. Henry argued that the state court erred by failing to\\nconsider evidence of his “historical alcoholism” in addition to his\\nintoxication on the day of the crime. Id. We concluded that any error\\nwas harmless because the history of substance abuse “was similar to\\nevidence already considered by the sentencing courts and was of limited\\nprobative value.” Id. In other words, because the state court had found\\ninsufficient evidence of alcohol use that was tied to the murder, we could\\nnot conclude that general alcohol use that was not tied to the murder\\ncould have been determinative. Here, by contrast, the Arizona Supreme\\nCourt found that Spreitz’s evidence of intoxication did not rise to the\\nlevel of statutory mitigation. Accordingly, there is not the same\\ncontextual harmlessness problem as in Henry.\\n\\n SPREITZ V. RYAN 37\\n\\nfactor on which much of his other mitigation evidence\\nturned.\\n\\n In both his report and his testimony, Dr. Flynn discussed\\nat length the extent to which alcohol consumed Spreitz’s life\\nin the decade leading up to Reid’s murder. Spreitz began\\ndrinking when he was twelve or thirteen; by fifteen, he was\\ndrinking steadily on the weekends and drinking vodka before\\nschool. According to Dr. Flynn, “[d]rug and alcohol abuse\\ndominated his teenage years.” After he left home, Spreitz’s\\ndrinking intensified: Spreitz experienced blackouts multiple\\ntimes and, a week before the murder, Spreitz was\\n“intoxicated a majority of the time.” Dr. Flynn concluded\\nthat “[i]t appears completely clear from the available\\ninformation that . . . Spreitz had a long-standing problem\\nwith alcohol which probably reached the level of physical\\ndependence.”\\n\\n Critically, Spreitz’s decade-long struggle with alcohol\\nand other addictive substances reflects the true gravity of his\\n“subnormal upbringing,” a mitigating circumstance\\nrecognized by the Arizona Supreme Court. See Spreitz I,\\n945 P.2d at 1280. Both Spreitz’s sentencing memorandum\\nand Dr. Flynn detailed how his difficult childhood led to\\nSpreitz’s early dependence on alcohol. For instance, Dr.\\nFlynn explained that Spreitz turned to alcohol as a coping\\nmechanism to deal with the “pervasive subabusive\\nemotional battering and neglect that he received from his\\nmother,” whom he described as “punitive, controlling, [and]\\nemotionally cold.” Dr. Flynn found that Spreitz “slipped\\nquietly into alcoholic numbness” after his “failure to live up\\nto his mother’s standards or earn her love.” Reflecting\\nparental neglect, Spreitz’s parents “failed to arrange . . . for\\nan appropriate rehabilitation program” even though\\n“[v]irtually everyone . . . who spent much time with him\\n\\n38 SPREITZ V. RYAN\\n\\ndescribed him as a heavy drinker.” “Given the intensity and\\nduration” of Spreitz’s reliance on alcohol, “the mother (and\\nstep-father) were either so uninvolved with him to have\\nfailed to notice the problem, or they were aware of it and\\nfailed to arrange (or attempt to arrange) for an appropriate\\nrehabilitation program.”\\n\\n Because Spreitz did not receive treatment during his\\nteenage years, his dependence on alcohol only deepened. In\\nhis sentencing memorandum, Spreitz, too, argued that his\\nlong-term alcohol abuse reflects the impact that his\\nupbringing by an emotionally unavailable, manipulative,\\npunitive mother had on him: it was “[n]o wonder that by 12\\nor 13 [h]e began drinking alcohol and smoking marijuana.”\\nAlthough the Arizona Supreme Court recognized that\\nSpreitz’s “subnormal” upbringing was “a mitigating\\ncircumstance,” it “accord[ed] it little weight.” Spreitz I,\\n945 P.2d at 1280. Had the court properly considered\\nSpreitz’s longstanding history of alcohol abuse in\\nconjunction with and as a manifestation of his disturbing\\nchildhood, both mitigating circumstances—and most\\nimportantly, the two in combination—would have been\\nmore compelling.\\n\\n The dissent argues that Spreitz’s long-term history of\\nalcohol abuse cannot be central to his plea for leniency\\nbecause much of his mitigation evidence focused on his\\n“abusive and dysfunctional childhood, upbringing, and\\ninteractions with his mother,” Dissent at 66, not his history\\nof alcohol and substance abuse. This argument, however,\\nmisses the point, as it overlooks the interlinked relationship\\nbetween Spreitz’s alcoholism and his subnormal childhood:\\nSpreitz’s history of alcohol and substance abuse is central to\\nevaluating the actual significance of his childhood as a\\nmitigating factor, and vice versa.\\n\\n SPREITZ V. RYAN 39\\n\\n In addition to being a symptom of his dysfunctional\\nchildhood, Spreitz’s alcohol and substance abuse can be seen\\nas linked to his emotional immaturity, another nonstatutory\\nmitigating circumstance recognized by the Arizona Supreme\\nCourt but described as not “significant.” Spreitz I, 945 P.2d\\nat 1282. In his report, Dr. Flynn explained that Spreitz was\\n“very likely to have been socially and emotionally\\nimmature” due, at least in part, to “[y]ears of alcoholism\\nintoxication [that] wipe[d] out many of the . . . healthy\\ndevelopmental processes requisite to age-appropriate social\\nand emotional maturity.” Similarly, in his sentencing\\nmemorandum, Spreitz connected “[h]is developmental\\nfailure” with his “[n]umbing by use of alcohol.” Again, had\\nthe Arizona Supreme Court properly considered Spreitz’s\\nlongstanding alcohol and substance abuse, another\\nmitigating factor—Spreitz’s emotional immaturity—would\\nhave been brought into proper perspective.\\n\\n Thus, by failing to consider Spreitz’s longstanding\\nalcohol and substance abuse, the Arizona Supreme Court\\nwas left with a critical void in Spreitz’s narrative: Spreitz’s\\nsubnormal childhood was so emotionally disturbing that it\\nled him to drink by the age of twelve or thirteen, which, in\\nturn, disrupted his normal development and contributed to\\nhis emotional immaturity. As both a symptom and cause of\\nother mitigating factors, Spreitz’s longstanding alcohol\\nabuse was thus central to his plea for leniency. 16 Without\\n\\n 16\\n Contrary to the dissent’s suggestion, Spreitz did not focus on his\\nintoxication at the time of the crime to the exclusion of his history of\\nalcohol abuse. Dissent at 22–24. Rather, both he and Dr. Flynn\\ndistinguished between Spreitz’s history of alcohol abuse as nonstatutory\\nmitigation and his intoxication at the time of the crime as statutory\\nmitigation. It is Spreitz’s history of alcohol abuse that underlies and\\namplifies much of his other mitigation evidence, not his intoxication at\\nthe time of Reid’s murder.\\n\\n40 SPREITZ V. RYAN\\n\\nconsideration of the evidence of his long-term substance and\\nalcohol abuse, the Arizona Supreme Court’s conclusions\\nregarding all of Spretiz’s mitigation evidence were\\nfundamentally flawed.\\n\\n We come to this conclusion fully recognizing, as the\\ndissent emphasizes, that “the severity of an aggravating\\ncircumstance must be considered when determining whether\\nan Eddings error is harmless.” Dissent at 71. The Arizona\\nSupreme Court described what Spreitz’s “confession and\\nphysical evidence” revealed: that “Spreitz beat and raped\\n[Reid] in a brutal assault that lasted many minutes before he\\ncrushed her skull.” Spreitz I, 945 P.2d at 1279. For this\\nreason, the Arizona Supreme Court affirmed the sentencing\\ncourt’s finding that Reid’s murder was especially cruel. Id.\\nEven so, we found prejudicial nexus-error in McKinney\\nwhere the murder was done in a cruel manner for pecuniary\\ngain as part of a double murder. McKinney, 813 F.3d at 823.\\n“When the choice is between life and death,” “the risk that\\nthe death penalty will be imposed in spite of facts which may\\ncall for a less severe penalty. . . . is unacceptable.” Abdul-\\nKabir v. Quarterman, 550 U.S. 233, 264 (2007) (quoting\\nLockett v. Ohio, 438 U.S. 586, 605 (1978)). Accordingly,\\nthe sentencer “must be able to give meaningful consideration\\nand effect to all mitigating evidence that might provide a\\nbasis for refusing to impose the death penalty on a particular\\nindividual, notwithstanding the severity of his crime.” Id. at\\n247 (emphasis added).\\n\\n Here, the Arizona Supreme Court failed to give any\\nmeaning to Spreitz’s longstanding alcohol and substance\\nabuse, an oversight which, in turn, minimized the value of\\nother mitigating evidence as well. Because the court’s\\nanalysis was missing a core component, we “cannot say,\\nwith fair assurance” that the Arizona Supreme Court’s\\n\\n SPREITZ V. RYAN 41\\n\\nEddings error did not affect its decision. McKinney,\\n813 F.3d at 822. As a result, the Arizona Supreme Court’s\\nrefusal to consider the evidence was not harmless under\\nBrecht.\\n\\n IV. Conclusion\\n\\n We reverse in part the district court’s judgment denying\\na writ of habeas corpus. We remand with instructions to\\ngrant the writ with respect to Spreitz’s sentence unless the\\nstate, within a reasonable period, either corrects the\\nconstitutional error in his death sentence or vacates the\\nsentence and imposes a lesser sentence consistent with the\\nlaw.\\n\\n REVERSED in part and REMANDED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1272,"cells":{"id":{"kind":"number","value":4596569,"string":"4,596,569"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Christopher Spreitz v. Charles Ryan"},"decision_date":{"kind":"string","value":"2019-03-04"},"docket_number":{"kind":"string","value":"09-99006"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Richard A. Paez, Marsha S. Berzon, and Richard C. Tallman, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PAEZ, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nCHRISTOPHER J. SPREITZ, No. 09-99006\\n Petitioner-Appellant,\\n D.C. No.\\n v. 4:02-CV-00121-JMR\\n\\nCHARLES L. RYAN, Warden,\\n Respondent-Appellee. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n John M. Roll, District Judge, Presiding\\n\\n Argued July 11, 2013\\n Submitted March 4, 2019\\n San Francisco, California\\n\\n Filed March 4, 2019\\n\\n Before: Richard A. Paez, Marsha S. Berzon,\\n and Richard C. Tallman, Circuit Judges.\\n\\n Opinion by Judge Paez;\\n Dissent by Judge Tallman\\n\\n2 SPREITZ V. RYAN\\n\\n SUMMARY *\\n\\n\\n Habeas Corpus / Death Penalty\\n\\n The panel reversed the district court’s denial of habeas\\ncorpus relief with respect to Christopher J. Spreitz’s death\\nsentence, and remanded, in a case in which Spreitz argued\\nthat the Arizona Supreme Court violated Eddings v.\\nOklahoma, 455 U.S. 104 (1982), by refusing to consider, as\\na matter of law, mitigating evidence of Spreitz’s\\nlongstanding alcohol and substance abuse on the ground that\\nhe did not establish a causal connection between this\\nmitigating evidence and the crime.\\n\\n The panel held that the district court erred in concluding\\nthat Spreitz’s claim that the Arizona Supreme Court violated\\nEddings is procedurally defaulted. The panel explained that\\nthe first opportunity Spreitz had to raise that claim was\\nbefore the post-conviction-relief (PCR) court, at which time\\nhe did so.\\n\\n Because the decision of the PCR court – which first\\ndeclared the claim waived, but proceeded to adjudicate the\\nclaim on the merits – was contrary to clearly established\\nSupreme Court precedent, the panel accorded that decision\\nno deference and reviewed Spreitz’s Eddings claim de novo.\\nThe panel concluded that the Arizona Supreme Court\\nviolated Eddings by impermissibly requiring that Spreitz\\nestablish a causal connection between his longstanding\\nsubstance abuse and the murder before considering and\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n SPREITZ V. RYAN 3\\n\\nweighing the evidence as a nonstatutory mitigating factor.\\nThe panel concluded that the error was not harmless.\\n\\n The panel affirmed the district court’s judgment denying\\nrelief with respect to Spreitz’s conviction in a concurrently\\nfiled memorandum disposition.\\n\\n Dissenting, Judge Tallman wrote that the record does not\\nestablish that either the sentencing court or the Arizona\\nSupreme Court unconstitutionally refused to consider\\nrelevant mitigating evidence; and that even if the Arizona\\ncourts did violate Eddings, Spreitz cannot show that this\\nerror had a “substantial and injurious effect or influence” on\\nhis ultimate sentence.\\n\\n\\n COUNSEL\\n\\nTimothy M. Gabrielsen (argued), Assistant Federal Public\\nDefender; John M. Sands, Federal Public Defender; Office\\nof the Federal Public Defender, Tucson, Arizona; Susan B.\\nFox and Sean Bruner, Law Office of Sean Bruner Ltd.,\\nTucson, Arizona; for Petitioner-Appellant.\\n\\nLacey Stover Gard (argued) and Jeffrey A. Zick, Section\\nChief Counsel; Kent E. Cattani, Chief Counsel, Criminal\\nAppeals/Capital Litigation Section; Mark Brnovich,\\nAttorney General; Office of the Attorney General, Tucson,\\nArizona; for Respondents-Appellees.\\n\\n4 SPREITZ V. RYAN\\n\\n OPINION\\n\\nPAEZ, Circuit Judge:\\n\\n In 1994, an Arizona jury convicted Christopher J. Spreitz\\n(“Spreitz”) of first-degree murder. The victim was thirty-\\nnine year old Ruby Reid (“Reid”). Finding that the cruelty\\nof the murder outweighed any mitigating circumstances, the\\ntrial judge sentenced Spreitz to death. Spreitz appeals the\\ndistrict court’s denial of his petition for a writ of habeas\\ncorpus challenging his conviction and sentence. We affirm\\nthe district court’s judgment with respect to Spreitz’s\\nconviction, 1 and reverse with respect to his sentence. 2\\n\\n In challenging his sentence, Spreitz argues that the\\nArizona Supreme Court unconstitutionally affirmed his\\ndeath sentence by failing to consider mitigating evidence of\\nhis longstanding alcohol and substance abuse. He contends\\nthat the state court refused to consider, as a matter of law,\\nthis evidence in mitigation because he did not establish a\\ncausal connection between the crime and his long-term\\nalcohol and substance abuse. In Eddings v. Oklahoma,\\n455 U.S. 104, 110 (1982), the Supreme Court held that under\\nboth the Eighth and Fourteenth Amendments, a sentencer in\\na capital case may not “refuse to consider, as a matter of law,\\n\\n 1\\n We affirm the judgment with respect to Spreitz’s conviction in a\\nconcurrently filed memorandum disposition.\\n\\n 2\\n On February 13, 2015, we vacated submission of Spreitz’s case\\npending final resolution of the en banc proceedings in McKinney v. Ryan,\\n730 F.3d 903 (9th Cir. 2013). In December 2015, the en banc court\\nissued an opinion in McKinney. 813 F.3d 798 (9th Circ. 2015) (en banc).\\nAfter the Supreme Court denied the State’s petition for a writ of\\ncertiorari, Ryan v. McKinney, 137 S. Ct. 39 (2016) (mem), we ordered\\nsupplemental briefing on the applicability of McKinney to Spreitz’s case.\\n\\n SPREITZ V. RYAN 5\\n\\nany relevant mitigating evidence” offered by the defendant.\\nId. at 114. Although a sentencer “may determine the weight\\nto be given relevant mitigating evidence . . . they may not\\ngive it no weight by excluding such evidence from their\\nconsideration.” Id. at 114–15 (footnote omitted). In\\ninterpreting and applying Eddings, the Supreme Court has\\nexplained that “full consideration of evidence that mitigates\\nagainst the death penalty is essential if the [sentencer] is to\\ngive a reasoned moral response to the defendant’s\\nbackground, character, and crime.” Penry v. Lynaugh\\n(Penry I), 492 U.S. 302, 328 (1989), abrogated on other\\ngrounds by Atkins v. Virginia, 536 U.S. 304 (2002) (internal\\nquotation marks and citation omitted). Moreover, the\\nSupreme Court has been clear: requiring a defendant to\\nprove a causal nexus between his mitigating evidence and\\nthe crime is “a test we never countenanced and now have\\nunequivocally rejected.” Smith v. Texas, 543 U.S. 37, 45\\n(2004) (per curiam).\\n\\n At the time of Spreitz’s sentencing, Arizona Revised\\nStatute Annotated § 13-703(G)(1994) 3 listed five mitigating\\nfactors, and Arizona case law additionally recognized\\nnonstatutory mitigating factors, including, for example, a\\ndefendant’s difficult family background or mental condition\\nnot severe enough to qualify as a statutory mitigating factor.\\nIn an en banc decision of our court, McKinney v. Ryan,\\n813 F.3d 798 (9th Cir. 2015), cert denied, 137 S. Ct. 39\\n(2016) (mem), we explained:\\n\\n For a period of a little over 15 years in capital\\n cases, in clear violation of Eddings, the\\n\\n 3\\n Arizona has since revised its death penalty sentencing scheme. All\\nreferences to Arizona’s Revised Statute Annotated are to those\\nprovisions in effect at the time of Spreitz’s sentencing.\\n\\n6 SPREITZ V. RYAN\\n\\n Supreme Court of Arizona articulated and\\n applied a “causal nexus” test for nonstatutory\\n mitigation that forbade as a matter of law\\n giving weight to mitigating evidence, such as\\n family background or mental condition,\\n unless the background or mental condition\\n was causally connected to the crime.\\n\\nId. at 802. As a result, we held in McKinney that\\n“[a]pplication of the causal nexus test to nonstatutory\\nmitigating factors violated Eddings, for it resulted in Arizona\\ncourts being entirely forbidden, as a matter of state law, to\\ntreat as a mitigating factor a family background or a mental\\ncondition that was not causally connected to a defendant’s\\ncrime.” Id. Spreitz argues that the Arizona Supreme Court\\napplied its causal nexus test in his case, refusing to consider\\nevidence of his long-term substance and alcohol abuse\\nbecause he did not adequately establish a causal connection\\nbetween that history of abuse and his crime.\\n\\n As in McKinney, “the precise question before us is\\nwhether the Arizona Supreme Court applied its\\nunconstitutional causal nexus test in affirming [Spreitz]’s\\ndeath sentence on de novo review.” 4 Id. at 804 (emphasis\\n\\n\\n 4\\n The Arizona Supreme Court is required by statute to undertake an\\nindependent review of a death sentence. Ariz. Rev. Stat. § 13-703.01(A).\\nIn conducting that review for crimes that occurred before 2002, the court\\n“independently review[s] the trial court’s findings of aggravation and\\nmitigation and the propriety of the death sentence. In doing so, [the\\ncourt] review[s] the record de novo, considering the quality and the\\nstrength, not simply the number, of aggravating and mitigating factors.”\\nState v. Lynch, 357 P.3d 119, 141 (Ariz. 2015) (internal quotation marks\\nand citations omitted), rev’d on other grounds, Lynch v. Arizona, 136 S.\\nCt. 1818 (2016).\\n\\n SPREITZ V. RYAN 7\\n\\nand internal quotation marks omitted). For the reasons that\\nfollow, we conclude that it did.\\n\\n I.\\n\\n A. Spreitz’s Crimes, Conviction, and Sentence\\n\\n On May 25, 1989, the police arrested Spreitz after\\ndiscovering Ruby Reid’s body in the desert. Upon\\nquestioning, Spreitz confessed to murdering Reid. We\\nbriefly provide the facts of the murder.\\n\\n On the evening of May 18, after drinking heavily and\\nbeing rejected by the woman he was dating, Spreitz “picked\\nup” Reid at a convenience store. State v. Spreitz (Spreitz I),\\n945 P.2d 1260, 1264–65 (Ariz. 1997). 5 In his confession,\\nSpreitz claimed that Reid voluntarily left with him and that\\nhis understanding was that they would have sex later that\\nevening. Id. at 1265. Spreitz further claimed that he drove\\nher out to the desert, where Reid decided she no longer\\nwanted to have sex. Id. The two fought as a result. Id.\\nSpreitz explained that Reid slapped him and that he\\nresponded by punching her in the mouth. Id. Spreitz then\\nsexually assaulted Reid—“remov[ing] her clothing and\\nha[ving] vaginal intercourse with her.” Id. Spreitz also\\nrecounted that he hit Reid in the head multiple times with a\\nrock “to make her stop yelling.” Id. He explained that he\\nleft Reid without knowing whether she was alive or dead.\\nId.\\n\\n Shortly after leaving Reid in the desert, Spreitz was\\nstopped by a Tucson police department officer. Id. at 1264.\\n\\n 5\\n Our recitation of the facts is adopted from the Arizona Supreme\\nCourt’s opinion affirming Spreitz’s conviction and death sentence.\\n\\n8 SPREITZ V. RYAN\\n\\nThe officer observed that Spreitz had a ripped shirt, smelled\\nof feces, and appeared to be covered in blood and fecal\\nmatter. Id. In addition, when detectives later searched\\nSpreitz’s car, they found blood spatter in the trunk, some of\\nwhich was inconsistent with Spreitz’s blood characteristics.\\nId. at 1265.\\n\\n On Monday morning, May 22, Reid’s naked and\\ndecomposing body was discovered on the outskirts of\\nTucson. Id. At trial, “the medical examiner testified that,\\ndue to the advanced state of decomposition, he could not\\ndetermine the full extent and nature of [Reid]’s injuries.” Id.\\nEven so, he was able to observe “bruising on the legs, arms,\\nand back; bruising and abrasions on the buttocks; several\\nbroken ribs; internal bleeding; a broken jaw; several head\\nlacerations; and a skull fracture where the skull had been\\n‘shoved in.’” Id. The medical examiner concluded that Reid\\nhad been killed by “blunt-force trauma to the head.” Id.\\n\\n In addition to finding Reid’s body “[a]t the scene, police\\ndetectives observed tire tracks leading back to the pavement,\\noil stains in the dirt, footprints, and drag marks in the dirt\\nleading away from the body. They also found feces-stained\\npants, tennis shoes, socks, a used tampon, and a torn\\nbrassiere. Two blood-stained rocks lay next to the body.”\\nId. A few days later, police arrested Spreitz. Id.\\n\\n On June 2, 1989, a grand jury indicted Spreitz for first-\\ndegree murder, Ariz. Rev. Stat. Ann. §§ 13-1105, 13-703;\\nsexual assault, Ariz. Rev. Stat. Ann. §§ 13-406(A) & (B);\\nand kidnapping, Ariz. Rev. Stat. Ann. §§ 13-304(A)(3) &\\n(B). Spreitz I, 945 P.2d at 1265. After five years of pre-trial\\nproceedings mostly regarding the admissibility of DNA\\nevidence, a seven-day jury trial began on August 9, 1994.\\nId. at 1266. After the conclusion of the trial, the jury\\nreturned guilty verdicts on all three counts: first-degree\\n\\n SPREITZ V. RYAN 9\\n\\nmurder (both premeditated and felony murder), sexual\\nassault, and kidnapping. Id.\\n\\n Prior to both his aggravation-mitigation and sentencing\\nhearings before the trial judge, Spreitz submitted evidence\\nand a memorandum in support of certain mitigating\\ncircumstances. As noted earlier, at the time of Spreitz’s\\nsentencing, Arizona’s death penalty statutes provided a list\\nof five specific mitigating factors; Arizona case law\\nrecognized nonstatutory mitigating factors as well. See Ariz.\\nRev. Stat. Ann. § 13-703(G); McKinney, 813 F.3d at 802.\\nSpreitz argued as nonstatutory mitigating factors: “(1) his\\ndysfunctional family life and lack of socialization; (2) a\\nhistory of alcohol and drug abuse; (3) his expressions of\\nremorse; [(4)] his good behavior while incarcerated; [(5)] his\\nlack of adult convictions; [and (6)] no prior record of violent\\ntendencies.” Spreitz I, 945 P.2d at 1279. Spreitz argued as\\nstatutory mitigating factors: (1) his age at the time of the\\nmurder, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and (2) that\\nhis “capacity to appreciate the wrongfulness of his conduct\\nor to conform his conduct to the requirements of law was\\nsignificantly impaired [due to alcohol use], but not so\\nimpaired as to constitute a defense to prosecution,” id.\\n§ 13-703(G)(1); see Spreitz I, 945 P.2d at 1279.\\n\\n Spreitz provided evidence of and argued for all the\\nforegoing mitigating circumstances but focused heavily on a\\ncombination of his relationship with his mother and his long\\nhistory of alcohol and substance abuse. To that end, Spreitz\\nsubmitted a written report by and presented testimony from\\nan examining psychologist, Dr. Todd Flynn, Ph.D. After\\nconducting interviews and research, Dr. Flynn concluded\\nthat Spreitz’s longstanding alcohol and substance abuse\\nshould be considered as both a statutory and nonstatutory\\nmitigating factor. In his report, which was admitted into\\n\\n10 SPREITZ V. RYAN\\n\\nevidence at the aggravation and mitigation hearing, Dr.\\nFlynn repeatedly emphasized Spreitz’s longstanding\\nsubstance abuse 6:\\n\\n By age twelve or thirteen, Chris Spreitz\\n began drinking alcohol and smoking\\n marijuana. By age 15, he drank steadily on\\n weekends and would have a shot of vodka\\n before school.\\n\\n The collateral information shows that the\\n alcohol abuse continued to intensify after he\\n left home. A variety of persons . . . described\\n him as a heavy drinker. This includes a\\n second cousin, Scott [Jouett], who saw him to\\n be intoxicated, “a majority of the time,” when\\n he was visiting Santa Barbara a week before\\n the current offense. To the interviewing\\n investigator, Mr. [Jouett] also described,\\n “several different occasions when Chris has\\n blackouts,” while drinking alcohol.\\n\\n It appears completely clear from the available\\n information that Chris Spreitz had a long-\\n standing problem with alcohol which\\n probably reached the level of physical\\n dependence. He described himself as\\n drinking in the morning as early as age 15.\\n Virtually everyone else who spent much time\\n with him described him as a heavy drinker.\\n\\n\\n\\n 6\\n Dr. Flynn testified consistently with his report. For ease of\\nreference, we refer only to his report.\\n\\n SPREITZ V. RYAN 11\\n\\nUltimately, Dr. Flynn concluded that Spreitz’s alcohol\\nabuse, childhood home life, and stunted development,\\ncombined with rejection by his girlfriend on the day of the\\ncrime, and intoxication at the time of the crime, led to the\\nmurder.\\n\\n At the end of his report, Dr. Flynn summarized his\\nfindings and opinions with respect to both statutory and\\nnonstatutory mitigating factors. He opined that both\\nSpreitz’s age, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and\\nimpaired capacity to appreciate the wrongfulness of his\\nconduct or conform his conduct to the law, Ariz. Rev. Stat.\\nAnn. § 13-703(G)(1), were mitigating statutory factors. As\\nto § 13-703(G)(5), specifically, Dr. Flynn opined that a\\ncombination of a disturbing upbringing in a “pathogenic,\\nemotionally neglectful home environment,” and “[y]ears of\\nalcoholism intoxication” combined to cause major deficits in\\nSpreitz’s social and emotional development. As to\\n§ 13-703(G)(1), Dr. Flynn offered a similar conclusion. In\\nlight of Spreitz’s “history of alcoholism . . . , a significant\\nbut unknown degree of alcohol intoxication is likely” on the\\nnight of the crime. In addition, Spreitz’s “history strongly\\nsuggests years of early experiences likely to have caused a\\nbuild-up of pent-up angry, aggressive feelings toward\\nwomen generally (and older women especially) which may\\nhave burst forth with uncontrollable intensity with or without\\nalcohol intoxication.” Dr. Flynn concluded that Spreitz’s\\nintoxication on the night of the crime coupled with his early\\nchildhood experiences “likely . . . contributed to an\\nuncontrollable outburst of aggression” and inability to\\ncontrol his conduct.\\n\\n Dr. Flynn also concluded that certain nonstatutory\\nmitigating factors were present. The nonstatutory mitigating\\nfactors included, in his opinion: Spreitz’s low potential for\\n\\n12 SPREITZ V. RYAN\\n\\nfuture violence, the “failure of [Spreitz’s] parents to provide\\ntreatment for alcohol abuse in [his] teenage years,” and the\\n“emotionally deprived, physically punitive home\\nenvironment” of Spreitz’s upbringing. Dr. Flynn also\\nemphasized that even if both statutory mitigating factors\\nfailed to satisfy the statutory threshold, they may still “be\\nappropriately considered . . . as nonstatutory mitigation.”\\n\\n On November 28, 1994, the trial court conducted an\\naggravation-mitigation hearing. At the hearing, Spreitz\\ncalled three mitigation witnesses: Dr. Flynn and two\\ncorrectional officers from the Pima County jail, where\\nSpreitz was incarcerated. The State did not offer any\\nwitnesses or evidence at the hearing. Dr. Flynn testified\\nconsistently with his report, as detailed above.\\n\\n On the same date, a probation officer filed a Pre-\\nSentence Report, which concluded:\\n\\n It appears the defendant became involved in\\n the senseless commission of the instant\\n offense due to his alcohol and drug abuse.\\n After five years in custody, he now admits his\\n substance abuse problem; however, this does\\n not condone his involvement in the offense.\\n It is unfortunate the victim died before the\\n defendant had his revelation.\\n\\nPrior to the sentencing hearing, Spreitz also submitted\\nseveral letters from friends, family, and jail personnel.\\n\\n On December 21, 1994, the trial court conducted a\\nsentencing hearing. Spreitz, Spreitz’s counsel, the victim’s\\n\\n SPREITZ V. RYAN 13\\n\\nsister, and the prosecutor each addressed the court. 7 The\\nprosecutor first disputed the mitigating evidence presented\\nby Spreitz at the mitigation-aggravation hearing. With\\nrespect to Spreitz’s alcohol abuse, the prosecutor argued that\\nSpreitz’s intoxication at the time of the crime did not meet\\nthe statutory definition for mitigation under § 13-703(G)(1)\\nbecause his conduct both during the murder and afterwards\\ndemonstrated that “he knew what he was doing.” The\\nprosecutor emphasized that the evidence revealed that Reid\\nhad been forcibly abducted and stuffed into the trunk of a\\ncar, thus forcing her to spend time “contemplat[ing] the\\nuncertainty of her fate.” The prosecutor argued that the\\nevidence revealed signs of struggle and that the serious\\ninjuries to Reid’s body belied any notion that Spreitz did not\\nknow what he was doing. The prosecutor further argued that\\nthe fact that Reid defecated on herself revealed that she was\\nterrified prior to her murder. In sum, the prosecutor argued\\nthat given both the heinous nature of the crime and the\\nmanner in which Reid suffered, a finding in aggravation that\\nReid was murdered in an “especially cruel manner” was\\nwarranted.\\n\\n After a short recess, the sentencing judge rendered oral\\nfindings addressing the aggravation and mitigation issues.\\nHe then imposed a sentence of death. 8 Following the\\n\\n\\n\\n 7\\n The trial court did not entertain argument at the conclusion of the\\naggravation-mitigation hearing. As a result, counsel addressed the\\naggravation-mitigation evidence and sentence options at the subsequent\\nsentencing hearing.\\n 8\\n Ring v. Arizona, 536 U.S. 584 (2002), the Supreme Court decision\\nholding judge-sentencing in capital cases unconstitutional, had yet to be\\ndecided.\\n\\n14 SPREITZ V. RYAN\\n\\nhearing, the sentencing judge filed a judgment setting forth\\nwritten findings.\\n\\n The judge first found one aggravating circumstance—\\nthat the offense was committed in an “especially cruel\\nmanner.” Ariz. Rev. Stat. Ann. § 13-703(F)(6). The judge\\nproceeded to review the “many factors [that were] submitted\\nin mitigation,” ultimately finding five mitigating\\ncircumstances of various levels of significance:\\n\\n [1] A mitigating circumstance defense\\n submitted was that the defendant had an\\n extremely disruptive childhood . . . . The\\n court finds the home was sub-normal, not\\n even a minimally healthy one for developing\\n children; and it is obvious the defendant\\n suffered a disruptive middle childhood—had\\n a punitive, controlling, cold mother, who he\\n could not please, no matter what he did.\\n\\n The defendant in his life turned to substance\\n abuse—alcohol and some suggestion he was\\n using cocaine and other drugs. However, the\\n court does not find such is a mitigating\\n circumstance that impaired his ability to\\n make a judgment on whether he was acting\\n rightfully or wrongfully in the death of the\\n victim.\\n\\n The defendant’s history of intoxication is\\n longstanding. He had been abusing\\n substances for close to ten years of his life at\\n the time of this offense when he was twenty-\\n two. Again, the court does not believe that\\n the substance abuse or intoxication impaired\\n\\n SPREITZ V. RYAN 15\\n\\n the defendant’s ability and capacity to\\n appreciate the wrongfulness of his conduct to\\n any significant degree. . . . The court does not\\n believe intoxication is any sort of mitigating\\n circumstance.\\n\\n ...\\n\\n [2] The court acknowledges that the\\n defendant has begun a process of\\n improvement and emotional growth while\\n confined at the Pima County Jail, where he\\n has taken part in education programs.\\n Correctional officers have testified he was a\\n prisoner who caused no problems.\\n\\n [3] The age of the defendant at the time of the\\n offense (twenty-two) is not a mitigating\\n circumstance in and of itself. Immaturity\\n probably is, but the court does not believe\\n immaturity was a significant mitigating\\n circumstance.\\n\\n [4] The court finds that Mr. Spreitz has no\\n criminal history of a felony nature—there is\\n no history or propensity for acts of violence.\\n\\n [5] The court believes the defendant is\\n capable of being rehabilitated. The court\\n does not know whether he has a good\\n prognosis for the future, but the court\\n believes he can be rehabilitated.\\n\\nIn conclusion, the judge found that “the mitigating\\ncircumstances [were] not sufficient to balance the\\n\\n16 SPREITZ V. RYAN\\n\\naggravating circumstances, nor [were] they sufficiently\\nsubstantial to call for leniency.” He thus imposed a death\\nsentence as to the first-degree murder conviction.\\n\\n B. Spreitz’s Appeal and Post-Conviction Proceedings\\n\\n Spreitz appealed his conviction and sentence to the\\nArizona Supreme Court with the assistance of new counsel.\\nThe Arizona Supreme Court affirmed Spreitz’s conviction\\nand sentence after conducting its own independent review of\\nthe record. 9 In reviewing the mitigating evidence, the court\\nfound four of the five mitigating circumstances that the\\nsentencing judge had found, declining to find Spreitz’s good\\nbehavior while in jail awaiting sentencing as a mitigating\\ncircumstance. Spreitz I, 945 P.2d at 1280–81. The court\\nexplained:\\n\\n [1] “We agree with the sentencing judge that\\n defendant’s upbringing was subnormal.” Id.\\n at 1280.\\n\\n [2] “We also find that the sentencing judge\\n . . . properly found that his emotional\\n immaturity was not a significant mitigating\\n factor.” Id. at 1281.\\n\\n [3] “We agree that the record supports the\\n sentencing judge’s findings that defendant\\n had no previous adult felony convictions, no\\n prior acts of violence, and\\n\\n\\n 9\\n See note 4, supra. Because the Arizona Supreme Court conducted\\na de novo review, we focus on its analysis, rather than, as the dissent\\ndoes in part, the analysis by the trial court.\\n\\n SPREITZ V. RYAN 17\\n\\n [4] that the defendant is capable of\\n rehabilitation.” Id.\\n\\nThe Arizona Supreme Court also added that it considered\\nremorse as an additional nonstatutory mitigating factor. Id.\\n\\n The court addressed Spreitz’s history of alcohol and\\nsubstance abuse as follows:\\n\\n The record demonstrates defendant’s\\n longtime substance abuse problems. We\\n note, however, that defendant’s general\\n problems with substance abuse are not\\n essential to our decision here. We therefore\\n decline to conclude that defendant was\\n impaired by alcohol consumption to an extent\\n that it interfered with his “capacity to\\n appreciate the wrongfulness of his conduct or\\n to conform his conduct to the requirements of\\n the law.” A[riz]. R[ev]. S[tat]. [Ann.]\\n § 13-703(G)(1); see also State v. Medrano,\\n 185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)\\n (citing [State v.] Stokley, 182 Ariz. [505,]\\n 520, 898 P.2d [454,] 469 [1995]).\\n\\nId. at 1280–81. In the only other portion of its opinion\\naddressing Spreitz’s history of alcohol and substance abuse,\\nthe court said:\\n\\n The sentencing judge found that defendant’s\\n ability to appreciate the wrongfulness of his\\n conduct was not impaired on the night of the\\n murder to any significant extent by substance\\n abuse, emotional disorders, situational stress,\\n or by a combination of these. Our review of\\n\\n18 SPREITZ V. RYAN\\n\\n the record convinces us that the trial court’s\\n finding was proper.\\n\\nId. at 1281.\\n\\n The Arizona Supreme Court agreed with the trial court’s\\naggravation finding that the murder was done in an\\nespecially cruel manner, id. at 1279, and then proceeded to\\nreweigh the applicable aggravating and mitigating factors,\\nfinding that “the aggravating circumstance of especial\\ncruelty in defendant’s murder of Ruby Reid outweigh[ed] all\\nfactors mitigating in favor of leniency.” Id. at 1282. It\\naffirmed his death sentence. Id. at 1283. Spreitz filed a\\npetition for a writ of certiorari, which the United States\\nSupreme Court denied. Spreitz v. Arizona, 523 U.S. 1027\\n(1998) (mem).\\n\\n In March 2000, represented again by new counsel,\\nSpreitz filed a petition for post-conviction relief in the\\nArizona Superior Court (“PCR court”). See Ariz. R. Crim.\\nP. 32.1. Spreitz alleged, inter alia, that the sentencing judge\\nand the Arizona Supreme Court committed error by failing\\nto consider his history of alcohol and substance abuse as a\\nnonstatutory mitigating factor apart from its causal\\nconnection to the murder—i.e., Eddings error. He also\\nalleged that his appellate counsel was ineffective for failing\\nto raise the sentencing judge’s Eddings error on appeal to the\\nArizona Supreme Court. Spreitz additionally raised several\\nnew claims of trial-counsel-ineffectiveness.\\n\\n The PCR court denied all of Spreitz’s claims and entered\\nan order dismissing his petition for post-conviction relief. In\\nPart 3 of its order, the PCR court discussed Spreitz’s claims\\nof “nexus-error”—that both the trial court and the Arizona\\nSupreme Court erred when they failed to consider Spreitz’s\\nlongstanding alcohol and substance abuse on the basis that\\n\\n SPREITZ V. RYAN 19\\n\\nSpreitz had failed to establish a causal nexus between the\\nlong-term substance abuse and the murder. 10 The PCR court\\ndismissed the claim 11 as waived, reasoning that Spreitz had\\nfailed to raise the issue on direct appeal. Nonetheless, the\\nPCR court addressed the merits of the nexus-error claim in\\nthe course of analyzing Spreitz’s argument that his appellate\\ncounsel was ineffective for failing to raise it. In doing so,\\nthe PCR court explained that the claim failed because:\\n\\n [I]t must be demonstrated, under A[riz].\\n R[ev]. S[tat]. [Ann.] § 13-703(G)(l), that\\n there is a causal link between the history of\\n alcohol or substance abuse and the offense\\n itself. E.g., State v. Stokley, [] 182 Ariz.\\n [505,] 523 [898 P.2d 454, 472 (Ariz. 1995)].\\n Without some basis for explaining or\\n defining the individual’s behavior at the time\\n of the offense, the Petitioner’s history of\\n alcohol or substance abuse would be\\n inconsequential (which is exactly what the\\n trial court and Supreme Court concluded).\\n State v. Kayer, 194 Ariz. 423, 984 P.2d 81\\n ([Ariz.] 1999).\\n\\n At times, the court can and should consider\\n an individual’s long-term alcoholism and\\n\\n 10\\n For ease of reference, we refer to the Arizona courts’ alleged error\\nas “nexus-error.”\\n\\n 11\\n Although Spreitz raised, and the PCR court recognized, two\\ndistinct claims—one with respect to the sentencing court and one with\\nrespect to the Arizona Supreme Court—the PCR court’s analysis treats\\nthe claims as one and the same and refers to them in the singular. To\\nreflect accurately the PCR court’s discussion, our summary here likewise\\nrefers to a singular “claim.”\\n\\n20 SPREITZ V. RYAN\\n\\n substance abuse, usually in conjunction with\\n other factors or diagnosis, as non-statutory\\n mitigation. However, the impact or effect of\\n the alcoholism or substance abuse must be\\n substantial and of such severity that it\\n provides a sufficient basis for explaining the\\n defendant’s conduct, character or ability to\\n control his behavior at the time of the offense.\\n [Citations omitted].\\n\\n As previously discussed, there is no evidence\\n in Petitioner’s case to suggest that he suffered\\n any long-term effects from his alcohol or\\n drug abuse that precluded him from\\n controlling his behavior. Petitioner did not\\n suffer from any cognitive or emotional\\n deficits that rendered him incapable of\\n controlling his conduct. Therefore, the trial\\n court did not err in failing to find Petitioner’s\\n history of alcohol or substance abuse as a\\n separate, non-statutory mitigating factor.\\n [Citation omitted].\\n\\n Spreitz filed a petition for review in the Arizona\\nSupreme Court challenging the PCR court’s judgment. The\\nArizona Supreme Court summarily affirmed the PCR court’s\\nmerits determination with respect to Spreitz’s Eddings\\nclaims. State v. Spreitz (Spreitz II), 39 P.3d 525, 527 (Ariz.\\n2002).\\n\\n Spreitz filed his federal habeas petition in February\\n2003. He alleged in claim seven that the sentencing judge\\nand the Arizona Supreme Court had both committed nexus-\\nerror with respect to his long-time alcohol and substance\\nabuse, and that appellate counsel was ineffective for failing\\n\\n SPREITZ V. RYAN 21\\n\\nto argue the sentencing judge’s nexus-error on direct appeal\\nto the Arizona Supreme Court.\\n\\n The district court denied all of Spreitz’s claims. With\\nrespect to the alleged nexus-error by both the sentencing\\ncourt and the Arizona Supreme Court, the district court\\nfound the claims procedurally barred, relying on the PCR\\ncourt’s determination that Spreitz had waived the claims\\nbecause he could have raised them on direct appeal but failed\\nto do so. The district court concluded that Spreitz had\\nproperly exhausted his ineffective assistance of appellate\\ncounsel claim, and thus addressed the alleged nexus-error in\\nthat context, ultimately concluding it was not meritorious.\\nAccordingly, the district court concluded that “appellate\\ncounsel’s failure to raise this issue on appeal does not\\nconstitute ineffectiveness.”\\n\\n Spreitz timely appealed.\\n\\n II. Jurisdiction and Standard of Review\\n\\n We have jurisdiction pursuant to 28 U.S.C. §§ 1291\\nand 2253(a), and review de novo the district court’s denial\\nof a writ of habeas corpus. Poyson v. Ryan, 879 F.3d 875,\\n887 (9th Cir. 2018). Because Spreitz filed his federal habeas\\npetition after April 24, 1996, he must satisfy the\\nAntiterrorism and Effective Death Penalty Act (“AEDPA”).\\nUnder AEDPA, we may not grant habeas relief unless the\\nstate’s adjudication of Spreitz’s claim (1) “was contrary to\\n. . . clearly established federal law[] as determined by the\\nSupreme Court,” (2) “involved an unreasonable application\\nof” such law, or (3) “was based on an unreasonable\\ndetermination of the facts in light of the evidence presented\\nin the State court proceeding.” (28 U.S.C. § 2254(d)). “In\\nmaking this determination, we look to the last state court\\ndecision to address the claim,” White v. Ryan, 895 F.3d 641,\\n\\n22 SPREITZ V. RYAN\\n\\n665 (9th Cir. 2018) (citing Wilson v. Sellers, 138 S. Ct. 1188,\\n1192 (2018)), which for Spreitz’s nexus-error claim is that\\nof the PCR court.\\n\\n Spreitz argues that the PCR court’s decision was\\n“contrary to” Eddings v. Oklahoma, 455 U.S. 104 (1982).\\n“A state court’s decision is contrary to clearly established\\nfederal law if it ‘applies a rule that contradicts the governing\\nlaw set forth in [U.S. Supreme Court] cases’ or arrives at a\\ndifferent result in a case that ‘confronts a set of facts that are\\nmaterially indistinguishable from a decision of [the\\nSupreme] Court.’” Castellanos v. Small, 766 F.3d 1137,\\n1146 (9th Cir. 2014) (alteration in original) (quoting\\nWilliams v. Taylor, 529 U.S. 362, 405–06 (2000)). “If the\\nstate court applies a legal standard that contradicts clearly\\nestablished [Supreme Court] law, we review de novo the\\napplicant’s claims, applying the correct legal standard to\\ndetermine whether the applicant is entitled to relief.” Id.\\n(citation omitted).\\n\\n III. Discussion\\n\\n As discussed supra, the precise question we must decide\\nis whether the Arizona Supreme Court applied its\\nunconstitutional causal nexus test in violation of Eddings\\nwhen it affirmed Spreitz’s death sentence. To answer that\\nquestion, we must first determine whether the claim is\\nproperly before us. After concluding that Spreitz’s claim\\ncould not have been procedurally defaulted, we turn to the\\nlevel of deference we must accord the PCR court’s ruling on\\nSpreitz’s Eddings claim under AEDPA. Because the PCR\\ncourt’s decision was contrary to clearly established Supreme\\nCourt precedent, we accord that decision no deference and\\nreview Spreitz’s Eddings claim de novo. We conclude that\\nthe Arizona Supreme Court violated Eddings by\\nimpermissibly requiring that Spreitz establish a causal\\n\\n SPREITZ V. RYAN 23\\n\\nconnection between his longstanding substance abuse and\\nthe murder before considering and weighing the evidence as\\na nonstatutory mitigating factor. Finally, having determined\\nthat the Arizona Supreme Court applied an unconstitutional\\ncausal nexus test in its sentencing procedure, we turn to\\nwhether the error was harmless under Brecht v. Abrahamson,\\n507 U.S. 619 (1993). We conclude that the error was not\\nharmless and therefore reverse the district court’s judgment\\nwith respect to Spreitz’s sentence.\\n\\n A.\\n\\n Under the doctrine of procedural default, a federal\\nhabeas court “will not review the merits of claims, including\\nconstitutional claims, that a state court declined to hear\\nbecause the prisoner failed to abide by a state procedural\\nrule.” Martinez v. Ryan, 566 U.S. 1, 9 (2012). “We review\\nde novo a district court’s conclusion that a claim is\\nprocedurally defaulted.” Cooper v. Neven, 641 F.3d 322,\\n326 (9th Cir. 2011).\\n\\n Here, the district court concluded that both of Spreitz’s\\nEddings claims—that the sentencing court and the Arizona\\nSupreme Court applied an unconstitutional causal nexus\\ntest—were procedurally defaulted, because the PCR court\\nfound each claim waived for failure to raise them on direct\\nappeal. This conclusion, however, is erroneous with respect\\nto Spreitz’s claim that the Arizona Supreme Court violated\\nEddings. After all, “[t]he Arizona Supreme Court reviews\\ncapital sentences de novo, making its own determination of\\nwhat constitute[s] legally relevant aggravating and\\nmitigating factors, and then weighing those factors\\nindependently. Ariz. Rev. Stat. Ann. § 13-755. The Arizona\\nSupreme Court ‘conducts a thorough and independent\\nreview of the record and of the aggravating and mitigating\\nevidence to determine whether the sentence is justified.’”\\n\\n24 SPREITZ V. RYAN\\n\\nMcKinney, 813 F.3d at 819 (emphasis removed) (quoting\\nState v. McKinney, 917 P.2d 1214, 1225 (Ariz. 1996)).\\nSpreitz could not have raised on direct appeal his claim that\\nthe Arizona Supreme Court violated Eddings in performing\\nits de novo review of Spreitz’s death sentence. The first\\nopportunity he had to raise that claim was before the PCR\\ncourt, at which time he did so. Thus, Spreitz’s claim is not\\nprocedurally defaulted and is properly before us. 12\\n\\n B.\\n\\n Because Spreitz’s claim that the Arizona Supreme Court\\napplied an unconstitutional causal nexus test to his\\nnonstatutory mitigating evidence of long-time alcohol and\\n 12\\n The district court declined to issue a certificate of appealability\\n(“COA”) on Spreitz’s substantive Eddings claim, which he raised as\\nclaim 1.7 on pages 108 through 111 in his federal habeas petition.\\nAlthough Spreitz did not present his Eddings claim as a separate\\nuncertified issue, we may grant a COA regarding any uncertified issue\\ndiscussed in a petitioner’s opening brief. See 9th Cir. R. 22-1(e).\\nSpreitz’s opening brief discusses the Arizona courts’ Eddings errors. We\\ntherefore may exercise our discretion to expand the COA to encompass\\nthe Eddings claim.\\n\\n To obtain a COA under 28 U.S.C. § 2253(c), Spreitz must\\ndemonstrate that “reasonable jurists could debate whether . . . the\\npetition should have been resolved in a different manner or that the issues\\npresented were adequate to deserve encouragement to proceed further.”\\nSlack v. McDaniel, 529 U.S. 473, 484 (2000). Reasonable jurists could\\ndebate the merits of Spreitz’s Eddings claim. We thus exercise our\\ndiscretion and expand the COA to encompass Spreitz’s substantive\\nEddings claim with respect to the Arizona Supreme Court. We\\nacknowledge that under Ninth Circuit Rule 22-1(f), the State would\\nnormally be afforded an opportunity to submit supplemental briefing on\\nthe uncertified Eddings issue. Here, however, the Eddings issue has been\\nexhaustively briefed, in the context of Spreitz’s appellate counsel’s\\nalleged ineffectiveness before the Arizona Supreme Court. Additional\\nbriefing is unwarranted.\\n\\n SPREITZ V. RYAN 25\\n\\nsubstance abuse is properly before us, we next determine the\\nlevel of deference to accord the PCR court’s adjudication of\\nthat claim. As discussed above, although the PCR court first\\ndeclared the claim waived, it proceeded to adjudicate the\\nclaim on the merits. In denying the claim, the PCR court\\nexplained:\\n\\n Without some basis for explaining or\\n defining the individual’s behavior at the time\\n of the offense, [Spreitz]’s history of alcohol\\n or substance abuse would be inconsequential\\n (which is exactly what the trial court and\\n Supreme Court concluded).\\n\\n At times, the court can and should consider\\n an individual’s long-term alcoholism and\\n substance abuse, usually in conjunction with\\n other factors or diagnosis, as non-statutory\\n mitigation. However, the impact or effect of\\n the alcoholism or substance abuse must be\\n substantial and of such severity that it\\n provides a sufficient basis for explaining the\\n defendant’s conduct, character, or ability to\\n control his behavior at the time of the offense.\\n\\n ....\\n\\n As previously discussed, there is no evidence\\n in [Spreitz]’s case to suggest that he suffered\\n any long-term effects from his alcohol or\\n drug abuse that precluded him, from\\n controlling his behavior. Petitioner did not\\n suffer from any cognitive or emotional\\n deficits that rendered him incapable of\\n controlling his conduct. Therefore, the trial\\n\\n26 SPREITZ V. RYAN\\n\\n court did not err in failing to find [Spreitz]’s\\n history of alcohol or substance abuse as a\\n separate, non-statutory mitigating factor.\\n\\n In other words, the PCR court concluded that Spreitz’s\\nlongstanding alcohol and substance abuse could only be\\nconsidered a nonstatutory mitigating factor if it “provide[d]\\na sufficient basis for explaining the defendant’s conduct,\\ncharacter, or ability to control his behavior at the time of the\\noffense.” This discussion both accurately described the\\ncausal nexus test and approved of it.\\n\\n Such reasoning is contrary to clearly established federal\\nlaw. As explained in McKinney, “the causal nexus test\\nclearly violates Eddings.” 813 F.3d at 810. Therefore,\\n“[b]ecause the state court used the wrong standard, we need\\nnot defer to that decision.” Hardy v. Chappell, 849 F.3d 803,\\n820 (9th Cir. 2016).\\n\\n C.\\n\\n Because the PCR court’s decision was “contrary to”\\nclearly established law, we review de novo the merits of\\nSpreitz’s Eddings claim. 28 U.S.C. § 2254(d)(1). We\\ntherefore consider whether the Arizona Supreme Court\\napplied its unconstitutional causal nexus test in reviewing\\nand affirming Spreitz’s death sentence. In so doing, “we\\nlook only to the decision of th[e Arizona Supreme Court],\\n. . . only [considering] . . . the decision of the sentencing\\njudge . . . to the degree it was adopted or substantially\\nincorporated by the Arizona Supreme Court.” McKinney,\\n813 F.3d at 819.\\n\\n SPREITZ V. RYAN 27\\n\\n 1.\\n\\n Our decision in McKinney frames our consideration of\\nthe Arizona Supreme Court’s decision and Spreitz’s claim.\\nMcKinney makes clear that the Arizona Supreme Court\\nconsistently articulated and applied its unconstitutional\\ncausal nexus test during the period in which Spreitz’s death\\nsentence was litigated in the trial court and reviewed by the\\nArizona Supreme Court. Id. at 824. Moreover, McKinney\\nmakes plain that although on habeas review, we generally\\napply a “presumption that state courts know and follow the\\nlaw,” “the Arizona Supreme Court’s consistent articulation\\nand application of its causal nexus test . . . make such a\\ncourse impossible.” Id. at 803 (internal quotation marks and\\ncitation omitted). In appeals heard from the late 1980s until\\nat least 2002, see State v. Canez, 42 P.3d 564, 594 (Ariz.\\n2002), it was unmistakably clear that “the Arizona Supreme\\nCourt did not know and follow federal law.” Id. (internal\\nquotation marks omitted). 13 Therefore, just as in McKinney\\nand subsequent cases applying McKinney, see, e.g., Poyson,\\n879 F.3d at 889, “the presumption” that the Arizona\\nSupreme Court knew and followed the law “is rebutted” here\\nas well. McKinney, 813 F.3d at 804.\\n\\n That said, McKinney does not dispose of Spreitz’s claim.\\nWe must still review the Arizona Supreme Court’s decision\\nto determine whether it did, in fact, apply its unconstitutional\\ncausal nexus test in Spreitz’s case. In doing so, we are\\nmindful of the fact that at the time the Arizona Supreme\\nCourt decided Spreitz’s appeal, “if there is to be a\\npresumption, it is that the Arizona Supreme Court violated\\n\\n\\n 13\\n In 2005, the Arizona Supreme Court repudiated its earlier nexus\\ntest. See State v. Anderson, 111 P.3d 369, 391–92 (Ariz. 2005).\\n\\n28 SPREITZ V. RYAN\\n\\nthe dictates of Lockett and Eddings.” Greenway v. Ryan,\\n866 F.3d 1094, 1095 (9th Cir. 2017).\\n\\n 2.\\n\\n Against this backdrop, we are convinced that the Arizona\\nSupreme Court applied its causal nexus test with respect to\\nSpreitz’s evidence of longstanding alcohol and substance\\nabuse in violation of Eddings.\\n\\n We quote again the Arizona Supreme Court’s discussion\\nof Spreitz’s history of alcohol and substance abuse:\\n\\n The record demonstrates defendant’s\\n longtime substance abuse problems. We\\n note, however, that defendant’s general\\n problems with substance abuse are not\\n essential to our decision here. We therefore\\n decline to conclude that defendant was\\n impaired by alcohol consumption to an extent\\n that it interfered with his “capacity to\\n appreciate the wrongfulness of his conduct or\\n to conform his conduct to the requirements of\\n the law.” A[riz]. R[ev]. S[tat]. [Ann.] § 13-\\n 703(G)(1); see also State v. Medrano,\\n 185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)\\n (citing Stokley, 182 Ariz. at 520, 898 P.2d at\\n 469).\\n\\n ....\\n\\n The sentencing judge found that defendant’s\\n ability to appreciate the wrongfulness of his\\n conduct was not impaired on the night of the\\n murder to any significant extent by substance\\n abuse, emotional disorders, situational stress,\\n\\n SPREITZ V. RYAN 29\\n\\n or by a combination of these. Our review of\\n the record convinces us that the trial court’s\\n finding was proper.\\n\\nSpreitz I, 945 P.2d at 1280–81. This discussion\\ndemonstrates that the Arizona Supreme Court primarily\\nconcluded that Spreitz failed to show statutory mitigation\\nunder Ariz. Rev. Stat. Ann. § 13-703(G)(1) based on his\\nintoxication on the night of the murder. “When applied\\nsolely in the context of statutory mitigation under\\n§ 13-703(G)(1), the causal nexus test does not violate\\nEddings.” McKinney, 813 F.3d at 810.\\n\\n The plain text of the Arizona Supreme Court’s decision,\\nhowever, also clearly demonstrates that the court did not\\nsolely apply its causal nexus test to Spreitz’s evidence of\\nstatutory mitigation. The court initially recognized Spreitz’s\\n“general problems with substance abuse”—evidence that\\nshould have been relevant as nonstatutory mitigation—but\\nconcluded those problems were “not essential to [its]\\ndecision” because they did not show Spreitz was “impaired\\non the night of the murder.” Spreitz I, 945 P.2d at 1280–81.\\nIn other words, the court did not acknowledge the relevance\\nof longtime substance abuse as nonstatutory mitigation in the\\nabsence of a causal relationship to the crime. Instead, the\\ncourt held that because there was no causal relationship,\\nSpreitz’s long-term alcohol abuse was not of significance—\\ni.e. “not essential”—to the court’s overall determination of\\neither statutory or nonstatutory mitigation.\\n\\n The Supreme Court, however, has been clear that “‘full\\nconsideration of evidence that mitigates against the death\\npenalty is essential if the sentencer is to give a reasoned\\nmoral response to the defendant’s background, character,\\nand crime.’” McKinney, 813 F.3d at 812 (first emphasis\\n\\n30 SPREITZ V. RYAN\\n\\nadded) (quoting Penry I, 492 U.S. at 319, 328).\\nAccordingly, “Eddings requires that ‘[t]he sentencer must\\nalso be able to consider and give effect’” to any relevant\\nmitigation evidence proffered by the defendant, which\\nincludes evidence going toward nonstatutory mitigation. Id.\\n(quoting Penry I, 492 U.S. at 319, 328). The Arizona\\nSupreme Court’s refusal to consider Spreitz’s alcohol and\\nsubstance abuse beyond its connection, if any, to Reid’s\\nmurder constitutes application of its unconstitutional causal\\nnexus test to relevant nonstatutory mitigation evidence.\\n\\n Our understanding of the Arizona Supreme Court’s\\nopinion is bolstered by the fact that in its discussion of\\nSpreitz’s longstanding alcohol and substance abuse, the\\ncourt cited to a prior decision that likewise applied the\\nunconstitutional causal nexus test, State v. Medrano,\\n914 P.2d 255 (Ariz. 1996). In McKinney, we explained that\\nin Medrano, the Arizona Supreme Court upheld the\\nsentencing judge’s determination that the defendant’s\\ncocaine use failed as a nonstatutory mitigating circumstance\\nbecause he did not show that it contributed to his conduct on\\nthe night of the murder. McKinney, 813 F.3d at 825–26. In\\nso doing, “[t]he Arizona Supreme Court applied the causal\\nnexus test.” Id. at 825. Although the Arizona Supreme\\nCourt’s citation to Medrano is not dispositive, it certainly\\ncorroborates our understanding of the court’s opinion. See\\nalso Poyson, 879 F.3d at 890 (emphasizing, among other\\nconsiderations, that the Arizona Supreme Court cited a\\npassage from a case specifically identified in McKinney as\\napplying the unconstitutional causal nexus test in order to\\nconclude that the state court also applied that test to Poyson’s\\nevidence). Moreover, as we explained in McKinney, “the\\nArizona Supreme Court applied its unconstitutional nexus\\ntest consistently” for fifteen years because it had “a mistaken\\nunderstanding of Eddings.” 813 F.3d at 826. It is thus\\n\\n SPREITZ V. RYAN 31\\n\\nentirely logical that in citing the exact portion of Medrano\\nwhere it previously applied its causal nexus test, the Arizona\\nSupreme Court was applying the same test in Spreitz’s case.\\nSee id. at 821; see also Poyson, 879 F.3d at 890.\\n\\n We find further support for our understanding of the\\nArizona Supreme Court’s unconstitutional treatment of\\nSpreitz’s alcohol and substance abuse when we compare it\\nwith the manner in which the court discussed other\\nmitigating factors. As detailed earlier, in evaluating the\\nevidence of Spreitz’s general history of alcohol and\\nsubstance abuse, the court stated that this evidence was “not\\nessential” to its decision. Spreitz I, 945 P.2d at 1280. By\\ncontrast, when the court considered other mitigating factors,\\nit specifically discussed the weight it would give those\\nfactors, as opposed to simply dismissing them outright. For\\nexample, when discussing Spreitz’s subnormal upbringing,\\nthe court stated, “Although we recognize defendant’s\\nupbringing as a mitigating circumstance, we accord it little\\nweight.” Id. at 1280 (emphasis added). As another example,\\nafter “finding that since his arrest, [Spreitz] has\\ndemonstrated remorse,” the court stated that his “remorse for\\nhis actions does little to counterbalance especial cruelty as a\\nserious aggravating circumstance.” Id. at 1281 (emphasis\\nadded). And finally, when discussing the sentencing judge’s\\ndetermination regarding Spreitz’s emotional immaturity, the\\ncourt stated, “We also find that the sentencing judge . . .\\nproperly found that his emotional immaturity was not a\\nsignificant mitigating factor.” Id. (emphasis added). In\\nother words, when the Arizona Supreme Court wanted to\\nassign weight to a given factor, it said so. This pattern stands\\nin stark contrast to the court’s conclusion that anything\\nrelated to Spreitz’s long-term alcohol or substance abuse that\\nfell short of the statutory definition could not serve as\\nmitigation. Id. at 1280–81.\\n\\n32 SPREITZ V. RYAN\\n\\n Finally, as also discussed earlier, the PCR court\\ndistinctly interpreted the Arizona Supreme Court’s decision\\nas having applied a causal nexus test. The PCR court\\nexplained that “[w]ithout some basis for explaining or\\ndefining the individual’s behavior at the time of the offense,\\n[Spreitz]’s history of alcohol or substance abuse would be\\ninconsequential (which is exactly what the trial court and\\nSupreme Court concluded).” Moreover, in articulating how\\nthe Arizona Supreme Court had applied its causal nexus test\\nin Spreitz’s direct appeal, the PCR court, as had the Arizona\\nSupreme Court, cited to a decision of the Arizona Supreme\\nCourt indisputably applying an unconstitutional nexus test,\\nState v. Kayer, 984 P.2d 31 (Ariz. 1999). See McKinney,\\n813 F.3d at 816 (explaining that Kayer held “that the\\ndefendant’s mental impairment ‘was not established as a\\nnonstatutory mitigating factor’ in part because ‘defendant\\noffered no evidence to show the requisite causal nexus that\\nmental impairment affected his judgment or his actions at the\\ntime of the murder’” (quoting Kayer, 984 P.2d at 46)). The\\nfact that the PCR court interpreted the Arizona Supreme\\nCourt’s opinion as we do—and then itself applied the\\nconstitutionally erroneous nexus test—lends further support\\nto our reading.\\n\\n We recognize that the Arizona Supreme Court’s decision\\non the particular issue before us was framed by language that\\nmight, in a decisional vacuum, suggest the court knew,\\nunderstood, and applied the law with respect to Eddings.\\nThe court acknowledged that Spreitz argued that his history\\nof alcohol and drug abuse served as both a statutory and\\nnonstatutory mitigating factor. Spreitz I, 945 P.2d at 1279.\\nIt then stated that it “must consider any aspect of the\\ndefendant’s character or record . . . relevant to determining\\nwhether the death penalty should be imposed.” Id. And\\nfinally, in reaching the conclusion to affirm Spreitz’s death\\n\\n SPREITZ V. RYAN 33\\n\\nsentence, the court explained that it had “examin[ed] the\\nentire record and reweigh[ed] the applicable aggravating and\\nmitigating factors.” Id. at 1281–82.\\n\\n The Arizona Supreme Court’s acknowledgment of what\\nSpreitz argued does not indicate that it relied on or accepted\\nthat argument as correct. Further, we cannot, given\\nMcKinney, Poyson, and Greenway, proceed in a decisional\\nvacuum.\\n\\n Considered in light of the baseline that the Arizona\\nSupreme Court was consistently applying its\\nunconstitutional causal nexus test in death penalty cases at\\nthe time Spreitz’s appeal was decided, as well as its citations\\nto cases held in McKinney to be applying that standard, we\\nare confident that the manner in which the Arizona Supreme\\nCourt disposed of Spreitz’s evidence of his longstanding\\nalcohol and substance abuse confirms that it applied the\\ncausal nexus test it was using during that period.\\n\\n Styers v. Schriro, 547 F.3d 1026 (9th Cir. 2008) supports\\nthis conclusion. Styers stated that the mere fact that “the\\nArizona Supreme Court stated that it had ‘considered all of\\nthe proffered mitigation,’” is not sufficient to contradict the\\nfact that “its analysis prior to this statement indicated\\notherwise.” Id. at 1035 (internal citation omitted). In Styers,\\nnotwithstanding assertions to the contrary, the Arizona\\nSupreme Court refused to consider evidence of the\\ndefendant’s Post Traumatic Stress Disorder (“PTSD”)\\nbecause he had failed to connect it causally to his crimes. Id.\\nHere, too, the Arizona Supreme Court’s actual analysis of\\nSpreitz’s longstanding alcohol and substance abuse reveals\\nthat it refused to consider this particular aspect of Spreitz’s\\ncharacter free from any connection to the crime. Rather, the\\ncourt’s discussion demonstrates that in evaluating Spreitz’s\\nlong-term alcohol and substance abuse, it applied a causal\\n\\n34 SPREITZ V. RYAN\\n\\nnexus test in violation of Eddings and therefore refused, as a\\nmatter of law, to consider it in assessing Spreitz’s evidence\\nof nonstatutory mitigating circumstances.\\n\\n To the extent the state argues the Arizona Supreme\\nCourt’s opinion is ambiguous as to whether it applied an\\nunconstitutional causal nexus test or not—and we believe\\nthat it is not—McKinney requires resolving that ambiguity\\nin favor of Spreitz. McKinney held it clear that the Arizona\\nSupreme Court “consistently articulated and applied its\\ncausal nexus test” at the time the trial court imposed\\nSpreitz’s death sentence and the Arizona Supreme Court\\nreviewed and affirmed that sentence. McKinney, 813 F.3d\\nat 803 (emphasis in original). Again, “if there is to be a\\npresumption, it is that the Arizona Supreme Court violated\\nthe dictates of Lockett and Eddings.” Greenway, 866 at\\n1095.\\n\\n In sum, because (1) the Arizona Supreme Court clearly\\nstated that Spreitz’s longstanding alcohol and substance\\nabuse was “not essential” to its decision, which was in line\\nwith the court’s consistent practice of applying a causal\\nnexus test; (2) the Arizona Supreme Court cited to Medrano,\\nin which it had previously articulated a causal nexus test;\\n(3) the Arizona Supreme Court carefully laid out the weight\\nit would accord other mitigating factors as opposed to simply\\nignoring them; and (4) the PCR court likewise interpreted\\nthe Arizona Supreme Court’s opinion as applying a causal\\nnexus test, we conclude that the Arizona Supreme Court\\nheld, as a matter of law, that Spreitz’s long-term alcohol and\\nsubstance abuse was not a nonstatutory mitigating factor and\\ntherefore refused to consider it. This holding was in clear\\nviolation of Eddings.\\n\\n SPREITZ V. RYAN 35\\n\\n 3.\\n\\n In light of our conclusion that the Arizona Supreme\\nCourt failed to comply with Eddings, we must next decide\\nwhether Spreitz was prejudiced by this error. On federal\\nhabeas review, relief is warranted if an error “had substantial\\nand injurious effect or influence” on the challenged decision.\\nSee Brecht v. Abrahamson, 507 U.S. 619, 623 (1993). The\\npetitioner is “not entitled to habeas relief” unless he can\\nestablish that the error “resulted in actual prejudice.” Davis\\nv. Ayala, 135 S. Ct. 2187, 2197 (2015) (internal quotation\\nmarks omitted). Here, because we are “in grave doubt about\\nwhether [the Arizona Supreme Court’s] error of federal law\\nhad substantial and injurious effect or influence in\\ndetermining [whether to affirm Spreitz’s death sentence],\\nthat error is not harmless. And, [Spreitz] must win.”\\nMcKinney, 813 F.3d at 822 (quoting O’Neal v. McAninch,\\n513 U.S. 432, 435 (1995)).\\n\\n We previously have required resentencing based on\\nprejudicial Eddings error where the Arizona Supreme Court\\nrefused to consider, as a matter of law, mitigation evidence\\n“central to” the petitioner’s “plea for leniency.” See id. at\\n823; see also Coleman v. Calderon, 210 F.3d 1047, 1051\\n(9th Cir. 2000) (concluding that a penalty-phase jury\\ninstruction error was not harmless under Brecht because “it\\nundermined the very core of Coleman’s plea for life”). 14 For\\nthe reasons that follow, it is clear that Spreitz’s long-term\\n\\n 14\\n We do not suggest that Brecht prejudice may only be established\\nif inappropriately considered mitigating evidence was at “the very core\\nof [a defendant’s] plea for life.” Coleman, 210 F.3d at 1051. It is\\nconceivable that other forms of Eddings error could have “a substantial\\nand injurious effect” on a defendant’s sentence if they would be\\nsignificant to the sentencer, despite the defendant’s failure to rely on\\nthem as central to his or her argument. Brecht, 507 U.S. at 623.\\n\\n36 SPREITZ V. RYAN\\n\\nalcohol and substance abuse was important mitigation\\nevidence, and was so closely related as to become central to\\nhis plea for leniency. The Arizona Supreme Court’s refusal\\nto consider that evidence had a substantial and injurious\\neffect on its sentencing determination that requires\\nresentencing.\\n\\n While a defendant’s evidence of longstanding alcohol\\nand/or substance abuse can be probative mitigation evidence\\non its own, a “history of substance abuse [is] substantially\\nmitigating when . . . combined with other mitigating\\nevidence.” Henry v. Ryan, 720 F.3d 1073, 1090 (9th Cir.\\n2013) (emphasis added) (listing cases and holding that the\\ndefendant’s alcoholism “would not have had a substantial\\nand injurious effect” because it “st[ood] alone, was similar\\nto evidence already considered by the sentencing courts, and\\nwas of limited probative value”) (footnote omitted). 15 Here,\\nSpreitz’s evidence regarding his history of alcohol and\\nsubstance abuse—spanning nearly half his life by the time\\nhe committed the crime at the age twenty-two—was the\\n\\n 15\\n The dissent makes much of Henry, suggesting it dictates the\\noutcome of our prejudice analysis. Dissent at 29. Not so. In that case,\\nthe state court had found that Henry’s intoxication on the day of the\\nmurder did constitute a statutory mitigating factor but was not\\nsufficiently substantial to outweigh the two aggravating factors. Henry,\\n720 F.3d at 1090. Henry argued that the state court erred by failing to\\nconsider evidence of his “historical alcoholism” in addition to his\\nintoxication on the day of the crime. Id. We concluded that any error\\nwas harmless because the history of substance abuse “was similar to\\nevidence already considered by the sentencing courts and was of limited\\nprobative value.” Id. In other words, because the state court had found\\ninsufficient evidence of alcohol use that was tied to the murder, we could\\nnot conclude that general alcohol use that was not tied to the murder\\ncould have been determinative. Here, by contrast, the Arizona Supreme\\nCourt found that Spreitz’s evidence of intoxication did not rise to the\\nlevel of statutory mitigation. Accordingly, there is not the same\\ncontextual harmlessness problem as in Henry.\\n\\n SPREITZ V. RYAN 37\\n\\nfactor on which much of his other mitigation evidence\\nturned.\\n\\n In both his report and his testimony, Dr. Flynn discussed\\nat length the extent to which alcohol consumed Spreitz’s life\\nin the decade leading up to Reid’s murder. Spreitz began\\ndrinking when he was twelve or thirteen; by fifteen, he was\\ndrinking steadily on the weekends and drinking vodka before\\nschool. According to Dr. Flynn, “[d]rug and alcohol abuse\\ndominated his teenage years.” After he left home, Spreitz’s\\ndrinking intensified: Spreitz experienced blackouts multiple\\ntimes and, a week before the murder, Spreitz was\\n“intoxicated a majority of the time.” Dr. Flynn concluded\\nthat “[i]t appears completely clear from the available\\ninformation that . . . Spreitz had a long-standing problem\\nwith alcohol which probably reached the level of physical\\ndependence.”\\n\\n Critically, Spreitz’s decade-long struggle with alcohol\\nand other addictive substances reflects the true gravity of his\\n“subnormal upbringing,” a mitigating circumstance\\nrecognized by the Arizona Supreme Court. See Spreitz I,\\n945 P.2d at 1280. Both Spreitz’s sentencing memorandum\\nand Dr. Flynn detailed how his difficult childhood led to\\nSpreitz’s early dependence on alcohol. For instance, Dr.\\nFlynn explained that Spreitz turned to alcohol as a coping\\nmechanism to deal with the “pervasive subabusive\\nemotional battering and neglect that he received from his\\nmother,” whom he described as “punitive, controlling, [and]\\nemotionally cold.” Dr. Flynn found that Spreitz “slipped\\nquietly into alcoholic numbness” after his “failure to live up\\nto his mother’s standards or earn her love.” Reflecting\\nparental neglect, Spreitz’s parents “failed to arrange . . . for\\nan appropriate rehabilitation program” even though\\n“[v]irtually everyone . . . who spent much time with him\\n\\n38 SPREITZ V. RYAN\\n\\ndescribed him as a heavy drinker.” “Given the intensity and\\nduration” of Spreitz’s reliance on alcohol, “the mother (and\\nstep-father) were either so uninvolved with him to have\\nfailed to notice the problem, or they were aware of it and\\nfailed to arrange (or attempt to arrange) for an appropriate\\nrehabilitation program.”\\n\\n Because Spreitz did not receive treatment during his\\nteenage years, his dependence on alcohol only deepened. In\\nhis sentencing memorandum, Spreitz, too, argued that his\\nlong-term alcohol abuse reflects the impact that his\\nupbringing by an emotionally unavailable, manipulative,\\npunitive mother had on him: it was “[n]o wonder that by 12\\nor 13 [h]e began drinking alcohol and smoking marijuana.”\\nAlthough the Arizona Supreme Court recognized that\\nSpreitz’s “subnormal” upbringing was “a mitigating\\ncircumstance,” it “accord[ed] it little weight.” Spreitz I,\\n945 P.2d at 1280. Had the court properly considered\\nSpreitz’s longstanding history of alcohol abuse in\\nconjunction with and as a manifestation of his disturbing\\nchildhood, both mitigating circumstances—and most\\nimportantly, the two in combination—would have been\\nmore compelling.\\n\\n The dissent argues that Spreitz’s long-term history of\\nalcohol abuse cannot be central to his plea for leniency\\nbecause much of his mitigation evidence focused on his\\n“abusive and dysfunctional childhood, upbringing, and\\ninteractions with his mother,” Dissent at 66, not his history\\nof alcohol and substance abuse. This argument, however,\\nmisses the point, as it overlooks the interlinked relationship\\nbetween Spreitz’s alcoholism and his subnormal childhood:\\nSpreitz’s history of alcohol and substance abuse is central to\\nevaluating the actual significance of his childhood as a\\nmitigating factor, and vice versa.\\n\\n SPREITZ V. RYAN 39\\n\\n In addition to being a symptom of his dysfunctional\\nchildhood, Spreitz’s alcohol and substance abuse can be seen\\nas linked to his emotional immaturity, another nonstatutory\\nmitigating circumstance recognized by the Arizona Supreme\\nCourt but described as not “significant.” Spreitz I, 945 P.2d\\nat 1282. In his report, Dr. Flynn explained that Spreitz was\\n“very likely to have been socially and emotionally\\nimmature” due, at least in part, to “[y]ears of alcoholism\\nintoxication [that] wipe[d] out many of the . . . healthy\\ndevelopmental processes requisite to age-appropriate social\\nand emotional maturity.” Similarly, in his sentencing\\nmemorandum, Spreitz connected “[h]is developmental\\nfailure” with his “[n]umbing by use of alcohol.” Again, had\\nthe Arizona Supreme Court properly considered Spreitz’s\\nlongstanding alcohol and substance abuse, another\\nmitigating factor—Spreitz’s emotional immaturity—would\\nhave been brought into proper perspective.\\n\\n Thus, by failing to consider Spreitz’s longstanding\\nalcohol and substance abuse, the Arizona Supreme Court\\nwas left with a critical void in Spreitz’s narrative: Spreitz’s\\nsubnormal childhood was so emotionally disturbing that it\\nled him to drink by the age of twelve or thirteen, which, in\\nturn, disrupted his normal development and contributed to\\nhis emotional immaturity. As both a symptom and cause of\\nother mitigating factors, Spreitz’s longstanding alcohol\\nabuse was thus central to his plea for leniency. 16 Without\\n\\n 16\\n Contrary to the dissent’s suggestion, Spreitz did not focus on his\\nintoxication at the time of the crime to the exclusion of his history of\\nalcohol abuse. Dissent at 22–24. Rather, both he and Dr. Flynn\\ndistinguished between Spreitz’s history of alcohol abuse as nonstatutory\\nmitigation and his intoxication at the time of the crime as statutory\\nmitigation. It is Spreitz’s history of alcohol abuse that underlies and\\namplifies much of his other mitigation evidence, not his intoxication at\\nthe time of Reid’s murder.\\n\\n40 SPREITZ V. RYAN\\n\\nconsideration of the evidence of his long-term substance and\\nalcohol abuse, the Arizona Supreme Court’s conclusions\\nregarding all of Spretiz’s mitigation evidence were\\nfundamentally flawed.\\n\\n We come to this conclusion fully recognizing, as the\\ndissent emphasizes, that “the severity of an aggravating\\ncircumstance must be considered when determining whether\\nan Eddings error is harmless.” Dissent at 71. The Arizona\\nSupreme Court described what Spreitz’s “confession and\\nphysical evidence” revealed: that “Spreitz beat and raped\\n[Reid] in a brutal assault that lasted many minutes before he\\ncrushed her skull.” Spreitz I, 945 P.2d at 1279. For this\\nreason, the Arizona Supreme Court affirmed the sentencing\\ncourt’s finding that Reid’s murder was especially cruel. Id.\\nEven so, we found prejudicial nexus-error in McKinney\\nwhere the murder was done in a cruel manner for pecuniary\\ngain as part of a double murder. McKinney, 813 F.3d at 823.\\n“When the choice is between life and death,” “the risk that\\nthe death penalty will be imposed in spite of facts which may\\ncall for a less severe penalty. . . . is unacceptable.” Abdul-\\nKabir v. Quarterman, 550 U.S. 233, 264 (2007) (quoting\\nLockett v. Ohio, 438 U.S. 586, 605 (1978)). Accordingly,\\nthe sentencer “must be able to give meaningful consideration\\nand effect to all mitigating evidence that might provide a\\nbasis for refusing to impose the death penalty on a particular\\nindividual, notwithstanding the severity of his crime.” Id. at\\n247 (emphasis added).\\n\\n Here, the Arizona Supreme Court failed to give any\\nmeaning to Spreitz’s longstanding alcohol and substance\\nabuse, an oversight which, in turn, minimized the value of\\nother mitigating evidence as well. Because the court’s\\nanalysis was missing a core component, we “cannot say,\\nwith fair assurance” that the Arizona Supreme Court’s\\n\\n SPREITZ V. RYAN 41\\n\\nEddings error did not affect its decision. McKinney,\\n813 F.3d at 822. As a result, the Arizona Supreme Court’s\\nrefusal to consider the evidence was not harmless under\\nBrecht.\\n\\n IV. Conclusion\\n\\n We reverse in part the district court’s judgment denying\\na writ of habeas corpus. We remand with instructions to\\ngrant the writ with respect to Spreitz’s sentence unless the\\nstate, within a reasonable period, either corrects the\\nconstitutional error in his death sentence or vacates the\\nsentence and imposes a lesser sentence consistent with the\\nlaw.\\n\\n REVERSED in part and REMANDED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4373822/\", \"author_raw\": \"PAEZ, Circuit Judge:\"}, {\"author\": \"TALLMAN, Circuit Judge, dissenting\", \"type\": \"dissent\", \"text\": \"TALLMAN, Circuit Judge, dissenting:\\n\\n The court declares that Christopher Spreitz’s death\\nsentence should be vacated because, applying de novo\\nreview, the sentencing court and the Arizona Supreme Court\\nviolated Eddings v. Oklahoma, 455 U.S. 104 (1982), and that\\nthis error was not harmless. I respectfully dissent.\\n\\n First, the record does not establish that either court\\nunconstitutionally refused to consider relevant mitigating\\nevidence. They were simply not persuaded by it. What the\\nrecord clearly shows is the Arizona courts did not find that\\nSpreitz’s proffered evidence outweighed the aggravating\\nfactor of cruelty in the way by which he brutally murdered\\nRuby Reid. Second, even if we determined that the Arizona\\ncourts did violate Eddings, Spreitz cannot show that this\\nerror had a “substantial and injurious effect or influence” on\\n\\n42 SPREITZ V. RYAN\\n\\nhis ultimate sentence. Brecht v. Abrahamson, 507 U.S. 619,\\n623 (1993). Given the savage brutality of his crimes and the\\nweakness of all the mitigation evidence he presented, any\\nremand for a redetermination of the appropriate penalty\\nought to result in the same decision.\\n\\n Nonetheless, the majority attempts unsuccessfully to\\nnavigate the course that Spreitz’s Eddings claim requires\\nhim to sail to warrant federal habeas relief. In doing so, the\\nmajority has once again foundered on the shoals of the\\nAntiterrorism and Effective Death Penalty Act of 1996\\n(AEDPA), specifically 28 U.S.C. § 2254—where so many\\nof our prior habeas corpus decisions are wrecked—failing to\\nafford the respect due the decisions of our sister state courts\\nand taunting the Supreme Court for another capsizing in lieu\\nof safe passage.\\n\\n I\\n\\n We review the Arizona post-conviction relief (PCR)\\ncourt’s decision because it is the last reasoned state court\\ndecision on Spreitz’s Eddings claim. See Barker v. Fleming,\\n423 F.3d 1085, 1091 (9th Cir. 2005). Under AEDPA we first\\nmust decide the level of deference to give to that decision by\\ndetermining whether it “was contrary to, or involved an\\nunreasonable application of, clearly established Federal law,\\nas determined by the Supreme Court of the United States.”\\nId. § 2254(d)(1).\\n\\n I do not quarrel with the majority’s determination that\\nthe PCR court held a mistaken view of what Eddings\\nrequires, and that we therefore must review Spreitz’s\\nEddings claim de novo. See Castellanos v. Small, 766 F.3d\\n1137, 1146 (9th Cir. 2014). However, “we may not grant\\nhabeas relief simply because of [the PCR court’s]\\n§ 2254(d)(1) error . . . .” Frantz v. Hazey, 533 F.3d 724, 735\\n\\n SPREITZ V. RYAN 43\\n\\n(9th Cir. 2008) (en banc); see also id. at 728, 735–37\\n(explaining our “approach to reviewing state court decisions\\nthat rely on legal principles contradicting clearly established\\nSupreme Court law but do not necessarily reach the wrong\\nresult”).\\n\\n On de novo review, Spreitz can meet neither of the two\\nrequirements to obtain habeas relief on his Eddings claim.\\nFirst, he cannot demonstrate that the state sentencing courts\\nactually violated Eddings. And second, even if there was a\\nviolation, he cannot show that any Eddings error by those\\nstate courts was prejudicial under Brecht.\\n\\n II\\n\\n Spreitz’s Eddings claim rests in large part on the\\nerroneous assumption that the sentencing court and the\\nArizona Supreme Court failed to consider his history of\\nalcohol and substance abuse, unrelated to the offense, as\\nnon-statutory mitigation when determining his sentence. He\\ncontends that, by failing to consider all mitigating\\ncircumstances, the state courts violated Eddings, which\\nprovides that in capital cases “the sentencer may not refuse\\nto consider or be precluded from considering ‘any relevant\\nmitigating evidence.’” Skipper v. South Carolina, 476 U.S.\\n1, 4 (1986) (quoting Eddings, 455 U.S. at 114). Eddings also\\nprovides that the sentencer “may determine the weight to be\\ngiven relevant mitigating evidence.” 455 U.S. at 114–15.\\n\\n Recently, we adopted an erroneous presumption that the\\nArizona Supreme Court repeatedly violated the dictates of\\nEddings by “consistently” applying an unconstitutional\\ncausal nexus test to all capital cases between 1989 and 2005.\\nSee McKinney v. Ryan, 813 F.3d 798, 803 (9th Cir. 2015)\\n\\n44 SPREITZ V. RYAN\\n\\n(en banc), cert. denied, 137 S. Ct. 39 (2016). 1 “We did not\\nsay, however, that the Arizona [Supreme Court] always\\napplied it.” See Greenway v. Ryan, 866 F.3d 1095, 1095 (9th\\nCir. 2017) (per curiam). Notably, in listing the cases in\\nwhich the Arizona courts erroneously applied the causal\\nnexus test, McKinney did not include Spreitz’s case. See\\n813 F.3d at 815–16, 824–26. And in McKinney, our holding\\nresolved only the “precise question” whether the state court\\nin that specific case had applied the causal nexus test. Id. at\\n804; see also Hedlund v. Ryan, 854 F.3d 557, 586 n.22 (9th\\nCir. 2017) (“We express no opinion as to how to apply\\nMcKinney in future Arizona capital cases from the suspect\\ntime period.”). Each case must be reviewed on its unique\\nfacts. We therefore must examine the state court decisions\\nin Spreitz’s case to determine whether they properly took\\ninto account all mitigating factors. 2\\n\\n\\n\\n\\n 1\\n The real damage inflicted by the en banc decision in McKinney is\\nthe broad assumption it adopts that now infects every death sentence\\nimposed by Arizona courts for horrendous murders adjudicated in the\\nsixteen-year period. It is regrettable that the United States Supreme\\nCourt denied certiorari in McKinney. But this case presents yet another\\nopportunity to right the error in Ninth Circuit death penalty habeas\\njurisprudence.\\n 2\\n In reviewing the last reasoned state court decision, we may\\nconsider “the record that was before the state court that adjudicated the\\nclaim on the merits.” Cannedy v. Adams, 706 F.3d 1148, 1156 (9th Cir.\\n2013) (citation omitted). The record before the PCR court in this case\\nincluded the sentencing judge’s decision and the Arizona Supreme\\nCourt’s direct appeal ruling. Thus, despite the majority’s preference for\\ncabining its view of the record to only the Arizona Supreme Court’s\\ndecision, we must consider both decisions here.\\n\\n SPREITZ V. RYAN 45\\n\\n A\\n\\n At all relevant times in this case, the operative Arizona\\ndeath penalty statute required a sentencing court to “impose\\na sentence of death if the court finds one or more of the\\naggravating circumstances enumerated in [the statute and\\ndetermines] that there are no mitigating circumstances\\nsufficiently substantial to call for leniency.” Ariz. Rev. Stat.\\n§ 13-703(E) (1993) (current version at Ariz. Rev. Stat.\\n§ 13-751(E) (2012)). Arizona law provided five statutory\\nmitigating factors, see § 13-703(G), including a modified\\nform of diminished capacity: “The defendant’s capacity to\\nappreciate the wrongfulness of his conduct or to conform his\\nconduct to the requirements of the law was significantly\\nimpaired, but not so impaired as to constitute a defense to\\nprosecution.” Id. § 13-703(G)(1). Even our en banc court\\nacknowledged, “When applied solely in the context of\\nstatutory mitigation under § 13-703(G)(1), the causal nexus\\ntest does not violate Eddings.” McKinney, 813 F.3d at 810.\\n\\n But Arizona law also provided for consideration of\\ncatch-all, non-statutory mitigating evidence that\\nencompassed “any factors proffered by the defendant or the\\nstate that are relevant in determining whether to impose a\\nsentence less than death, including any aspect of the\\ndefendant’s character, propensities or record and any of the\\ncircumstances of the offense.” Ariz. Rev. Stat. § 13-703(G).\\nThe dictates of Eddings apply to such non-statutory\\nmitigating evidence. See McKinney, 813 F.3d at 810\\n(“When applied in the context of nonstatutory mitigation, the\\ncausal nexus test clearly violates Eddings.”). Lastly,\\nArizona law requires the Arizona Supreme Court to\\n“independently review the trial court’s findings of\\naggravation and mitigation and the propriety of the death\\n\\n46 SPREITZ V. RYAN\\n\\nsentence.” Ariz. Rev. Stat. § 13-703.01(A) (1994) (current\\nversion at Ariz. Rev. Stat. § 13-755(A) (2009)).\\n\\n B\\n\\n Here, Spreitz cannot show that the sentencing judge, the\\nHonorable William N. Sherrill, violated Eddings in\\ndetermining Spreitz’s sentence. At sentencing, Spreitz\\nsubmitted a memorandum that specifically discussed his\\nhistory of substance abuse and cited Eddings; defense\\npsychologist Dr. Todd C. Flynn provided the court with a\\npsychological evaluation report and testimony specifically\\ndiscussing Spreitz’s history of substance abuse as both\\nstatutory and non-statutory mitigating factors; and Judge\\nSherrill differentiated between the two types of mitigation\\nand ultimately decided that the mitigating evidence did not\\noutweigh the aggravating factor of cruelty based on how\\nSpreitz murdered Ms. Reid.\\n\\n In his written judgment, Judge Sherrill acknowledged\\nthat “many factors have been submitted in mitigation.”\\nAmong the mitigating evidence, he observed that Spreitz “in\\nhis life turned to substance abuse—alcohol and some\\nsuggestion he was using cocaine and other drugs”—and that\\nSpreitz’s “history of intoxication [was] longstanding.” He\\nalso found that Spreitz “had been abusing substances for\\nclose to ten years of his life.” Judge Sherrill’s discussion of\\nSpreitz’s history of substance abuse unrelated to the crime\\nestablishes that he considered that evidence in his sentencing\\ndecision. See Williams v. Stewart, 441 F.3d 1030, 1057 (9th\\nCir. 2006) (stating that a sentencing court does not violate\\nEddings where it is clear that “the sentencing court\\nconsidered all relevant mitigating evidence that was\\noffered”).\\n\\n SPREITZ V. RYAN 47\\n\\n There is an important distinction between Judge\\nSherrill’s consideration of Spreitz’s evidence of his long-\\nterm substance abuse, and Judge Sherrill’s subsequent\\nfindings regarding the weight he ultimately decided to give\\nthat evidence. After discussing Spreitz’s proffered\\nmitigating evidence, Judge Sherrill correctly observed that\\nthe issue was, in weighing the imposition of capital\\npunishment, “whether any of these mitigating\\ncircumstances, individually or cumulatively, can balance or\\noutweigh the aggravating circumstances.” And, he finally\\ndetermined that “the mitigating circumstances [were] not\\nsufficient to balance the aggravating circumstances, nor\\n[were] they sufficiently substantial to call for leniency.” In\\nreaching that conclusion, Judge Sherrill specified that he did\\n“not believe that intoxication [was] any sort of mitigating\\ncircumstance.” (Emphasis added.) One should fairly read\\nthat conclusion to mean that Spreitz’s long history of\\nsubstance abuse was neither a statutory nor a non-statutory\\nmitigating factor “sufficiently substantial to call for\\nleniency.” Ariz. Rev. Stat. § 13-703(E).\\n\\n Under Eddings, Judge Sherrill was free to conclude that\\nSpreitz’s mitigating evidence, whether statutory or non-\\nstatutory, simply did not warrant leniency given the\\naggravating circumstance of how Spreitz cruelly murdered\\nRuby Reid. See Mann v. Ryan, 828 F.3d 1143, 1159–60 (9th\\nCir. 2016) (holding that defendant “failed to show that the\\nmitigating circumstances he had presented,” including a\\nhistory of substance abuse, “outweighed the aggravating\\ncircumstances of his crimes”), cert. denied, 137 S. Ct. 1091\\n(2017), 137 S. Ct. 1205 (2017); Henry v. Ryan, 720 F.3d\\n1073, 1090 n.11 (9th Cir. 2013) (recognizing that “historical\\nalcoholism might [be] considered aggravating as well as\\nmitigating, depending on the perspective of the sentencing\\ncourt”). “It is sufficient that a sentencing court state that it\\n\\n48 SPREITZ V. RYAN\\n\\nfound no mitigating circumstances that outweigh the\\naggravating circumstances.” Williams, 441 F.3d at 1057\\n(quotation omitted).\\n\\n Eddings also did not prevent Judge Sherrill from giving\\nless weight to Spreitz’s substance abuse history based on a\\nlack of nexus to the murder. See Hedlund, 854 F.3d at 587\\nn.23 (stating that under Eddings, “a court is free to assign\\nless weight to mitigating factors that did not influence a\\ndefendant’s conduct at the time of the crime”); Styers v.\\nRyan, 811 F.3d 292, 298–99 (9th Cir. 2015) (holding that the\\nArizona Supreme Court did not violate Eddings in assigning\\nlittle weight to the petitioner’s PTSD in the absence of a\\ncausal connection to the crime), cert. denied, 137 S. Ct. 1332\\n(2017). Nor did Eddings prevent Judge Sherrill from\\nassigning no weight to Spreitz’s long-term substance abuse\\nunrelated to the offense, so long as Judge Sherrill considered\\nthat evidence, which the record shows he definitely did. See\\nWilliams, 441 F.3d at 1057 (“Once mitigating evidence is\\nallowed in, a finding that there are ‘no mitigating\\ncircumstances’ does not violate the Constitution.”); Ortiz v.\\nStewart, 149 F.3d 923, 943 (9th Cir. 1998) (“[T]he\\nsentencing court did not refuse to consider mitigating\\nevidence; it considered the evidence and found it inadequate\\nto justify leniency. That assessment did not violate the\\nConstitution.”), overruled in part on other grounds by\\nMartinez v. Ryan, 566 U.S. 1 (2012).\\n\\n In addition to the above findings, Judge Sherrill also\\nadded that he did not “believe that [Spreitz’s] substance\\nabuse or intoxication impaired [his] ability and capacity to\\nappreciate the wrongfulness of his conduct to any significant\\ndegree” under § 13-703(G)(1). At the time, Arizona courts\\nwere required by law to examine whether evidence of\\nalcohol and drug use bore a causal connection to the crime.\\n\\n SPREITZ V. RYAN 49\\n\\nAriz. Rev. Stat. § 13-703(G)(1) (requiring the sentencer to\\nconsider leniency if the defendant proved that his “capacity\\nto appreciate the wrongfulness of his conduct or to conform\\nhis conduct to the requirements of law was significantly\\nimpaired, but not so impaired as to constitute a defense to\\nprosecution”). And, there is no question that Spreitz pursued\\na mitigation strategy at sentencing that attempted to link his\\nlong-term substance abuse with the kidnap/murder to show\\nstatutory mitigation under § 13-703(G)(1).\\n\\n For example, Spreitz’s sentencing memorandum stated\\nthat Dr. Flynn’s insights about him were of “the nature, type\\nand character contemplated by A.R.S. Section\\n13-703(G)(1),” and that “Dr. Flynn’s report [was], indeed,\\nrelevant and should be considered by the Court as statutory\\nmitigating circumstances.” Dr. Flynn’s report and expert\\ntestimony went to great lengths to argue that Spreitz’s\\nalcohol abuse was “very strongly correlated with violent\\nbehavior . . . .” Judge Sherrill could not respond to Spreitz’s\\nargument that § 13-703(G)(1) called for leniency without\\nanalyzing whether Spreitz had shown a connection between\\nhis substance abuse and the murder. We cannot infer from\\nthat causal nexus analysis that Judge Sherrill flatly refused\\nto consider any evidence of substance abuse that was not\\ncausally related to the crime.\\n\\n Rather, this record shows that Judge Sherrill\\nconstitutionally applied a causal nexus test in the context of\\nan Arizona statutory mitigating factor, § 13-703(G)(1). See\\nMcKinney, 813 F.3d at 810 (“When applied solely in the\\ncontext of statutory mitigation . . . , the causal nexus test\\ndoes not violate Eddings.”). Judge Sherrill’s other\\nstatements repeatedly demonstrate that he understood\\nEddings’ mandate and considered all of Spreitz’s proffered\\nmitigating evidence, but ultimately found that evidence\\n\\n50 SPREITZ V. RYAN\\n\\ninsufficient to warrant leniency. We are in the same\\nsituation here that recently confronted our Court in Hedlund.\\nSee 854 F.3d at 591 (Bea, J., concurring) (“[Although] Judge\\nSheldon constitutionally applied a causal-nexus test in the\\ncontext of an Arizona statutory mitigating factor[,\\n§ 13-703(G)(1), that] statement does not show that Judge\\nSheldon excluded mitigating evidence from his\\nconsideration, and Judge Sheldon’s other statements\\nrepeatedly demonstrate otherwise.”).\\n\\n Nor does the record before us support the conclusion that\\nJudge Sherrill failed to consider evidence of Spreitz’s long-\\nterm substance abuse as a non-statutory mitigating factor.\\nWhat we can say for sure is that he was not persuaded by\\nthat evidence. Judge Sherrill listened to testimony and\\narguments at trial and in the penalty phase, read everything\\nsubmitted, and then deliberated for three days before\\nconcluding that Spreitz’s mitigation evidence was not\\nsufficient to outweigh the evidence in aggravation\\nwarranting death. There is no Eddings error here.\\n\\n C\\n\\n Similarly, in its “detailed independent review” of\\nSpreitz’s sentence, the Arizona Supreme Court “examined\\nthe entire record to weigh and consider the aggravating and\\nmitigating circumstances.” State v. Spreitz, 945 P.2d 1260,\\n1278 (Ariz. 1997). Consistent with Eddings, the Arizona\\nSupreme Court recognized “that the sentencing judge must\\nconsider any aspect of the defendant’s character or record\\nand any circumstance of the offense relevant to determining\\nwhether the death penalty should be imposed.” Id. at 1279\\n(quotation omitted). And, it recognized “that the weight\\naccorded such evidence is within the sentencing judge’s\\ndiscretion.” Id.; see Greenway, 866 F.3d at 1097 (citing\\nsimilar language in the Arizona Supreme Court’s review of\\n\\n SPREITZ V. RYAN 51\\n\\na capital sentence to conclude that the court “did not reject\\nany mitigating factor, as a matter of law, on the theory that\\nit was not related to the commission of the crime”).\\n\\n In that review, the Arizona Supreme Court specifically\\nobserved that the “record demonstrates defendant’s longtime\\nsubstance abuse problems.” Spreitz, 945 P.2d at 1280.\\nNonetheless, the court found that Spreitz’s “general\\nproblems with substance abuse [were] not essential to [its]\\ndecision” because, ultimately, “the aggravating\\ncircumstance of especial cruelty in [Spreitz]’s murder of\\nRuby Reid outweigh[ed] all factors mitigating in favor of\\nleniency.” Id. at 1280, 1282. The Arizona Supreme Court\\nthus rejected, on the merits, Spreitz’s argument that his long-\\nterm substance abuse constituted either a statutory or non-\\nstatutory mitigating factor warranting leniency in his case.\\nThere is no mitigation evidence it refused to consider.\\n\\n To the extent that the majority cites Poyson v. Ryan,\\n879 F.3d 875 (9th Cir. 2018), for the proposition that\\nMcKinney, 813 F.3d at 802–03, adopts a per se rule that the\\nArizona Supreme Court applied an unconstitutional causal\\nnexus test in all capital cases over a 15-year period, I fully\\nagree with Judge Ikuta’s reluctant concurrence in Poyson.\\n879 F.3d at 897–900. McKinney was wrongly decided and\\nflips the “presumption that state courts know and follow the\\nlaw.” Id. at 897–98 (citing Woodford v. Visciotti, 537 U.S.\\n19, 24 (2002)). The record here rebukes even the McKinney\\npresumption and in that respect is factually distinguishable\\nfrom both McKinney and Poyson.\\n\\n The majority nonetheless concludes that the Arizona\\nSupreme Court’s use of the term “not essential” proves that\\nthe court “require[d] that Spreitz establish a causal\\nconnection between his longstanding substance abuse and\\nthe murder before considering and weighing the evidence as\\n\\n52 SPREITZ V. RYAN\\n\\na nonstatutory mitigating factor.” Op. at 22–23, 29. Not so.\\nWe can reasonably interpret “not essential” to mean\\ninconsequential, which we have previously defined in a\\nsimilar context as having little effect on the outcome. See\\nLandrigan v. Stewart, 272 F.3d 1221, 1230 (9th Cir. 2001)\\n(“At any rate, any error in failing to consider Landrigan’s\\nuse of alcohol and drugs would have been inconsequential;\\nit would have had no effect whatsoever on the outcome.”),\\nadopted sub nom. Landrigan v. Schriro, 501 F.3d 1147 (9th\\nCir. 2007) (en banc).\\n\\n The majority reasons that, unlike the Arizona Supreme\\nCourt’s consideration of Spreitz’s substance abuse history\\nunrelated to the murder, “when the court considered other\\nmitigating factors, it specifically discussed the weight it\\nwould give to those factors . . . .” Op. at 31. But the law has\\nnever imposed an obligation on the sentencer to explicitly\\ndisclose the value it assigns to every piece of mitigating\\nevidence. “The U.S. Supreme Court has ‘never held that a\\nspecific method for balancing mitigating and aggravating\\nfactors in a capital sentencing proceeding is constitutionally\\nrequired.’” Styers, 811 F.3d at 298 (quoting Kansas v.\\nMarsh, 548 U.S. 163, 175 (2006)). Nor did McKinney\\nabrogate the principle that a sentencing court is not required\\nto “itemize and discuss every piece of evidence offered in\\nmitigation.” Jeffers v. Lewis, 38 F.3d 411, 418 (9th Cir.\\n1994) (en banc). “We have determined that a sentencing\\ncourt need not specifically discuss each individual item of\\nmitigating evidence so long as it appears to have considered\\nall relevant evidence.” Ortiz, 149 F.3d at 943. This is\\nparticularly important when reviewing Eddings claims\\nbecause otherwise we could infer that a sentencing court\\nrefused to consider any piece of evidence to which it did not\\nexpressly assign a value, even though it sufficiently\\ndiscussed that proffered evidence. That is not the law.\\n\\n SPREITZ V. RYAN 53\\n\\n For example, in Jeffers, our en banc court held that the\\nstate court’s “failure to list and discuss each item of evidence\\n[that the defendant] offered in mitigation” did not\\ndemonstrate that the “court neglected to consider all of his\\nmitigation evidence.” 38 F.3d at 417–18. Instead, we held\\nthat there was no clear requirement under either Arizona law\\nor the federal Constitution that the sentencer itemize and\\ndiscuss each piece of evidence offered in mitigation. Id. at\\n418. And, we held that the sentencing court there clearly\\nconsidered all such evidence. Id. In addition, a sentencing\\ncourt “need not exhaustively analyze each mitigating factor\\nas long as a reviewing federal court can discern from the\\nrecord that the state court did indeed consider all mitigating\\nevidence offered by the defendant.” Moormann v. Schriro,\\n426 F.3d 1044, 1055 (9th Cir. 2005) (quotation omitted); see\\nalso Lopez v. Schriro, 491 F.3d 1029, 1039 (9th Cir. 2007)\\n(observing that there is no “clearly established Supreme\\nCourt precedent setting forth the record the sentencing court\\nmust make in order to permit sufficient appellate review”\\n(quotation omitted)). We can easily make that determination\\nhere.\\n\\n The majority also focuses on the Arizona Supreme\\nCourt’s additional finding that Spreitz “was [not] impaired\\nby alcohol consumption to an extent that it interfered with\\nhis ‘capacity to appreciate the wrongfulness of his conduct\\nor to conform his conduct to the requirements of the law.’”\\nOp. at 28 (quoting Spreitz, 945 P.2d at 1280–81). But this\\nstatement does not mean that the Arizona Supreme Court\\nrefused to consider Spreitz’s long-term substance abuse\\nunrelated to the murder. On the contrary, as the majority\\nrecognizes, the Arizona Supreme Court’s “decision . . .\\nsuggest[ed] the court knew, understood, and applied the law\\nwith respect to Eddings.” Id. at 32. For example, the\\nArizona Supreme Court stated: “In our review, we have\\n\\n54 SPREITZ V. RYAN\\n\\nbeen mindful that the sentencing judge must consider any\\naspect of the defendant’s character or record and any\\ncircumstance of the offense relevant to determining whether\\nthe death penalty should be imposed.” Spreitz, 945 P.2d at\\n1279 (quotations omitted) (emphasis added). After\\nindependently “weighing the mitigating circumstances both\\nindividually and cumulatively against the aggravating\\ncircumstance,” it ultimately found that “the aggravator of\\nespecial cruelty outweighed all other circumstances.” Id. at\\n1280. Those findings do not violate Eddings. They reflect\\na careful consideration of all of the evidence, ultimately\\nconcluding that the death sentence was warranted here\\ndespite what Spreitz offered in mitigation. Thus, there was\\nno Eddings violation.\\n\\n III\\n\\n Even if we were to determine that the state courts\\ncommitted Eddings error by failing to consider Spreitz’s\\nlong-term substance abuse unrelated to the offense, Spreitz\\ncannot show that this error was prejudicial because the\\nviolent manner in which he abducted and murdered Ruby\\nReid trenchantly outweighed all evidence proffered in\\nmitigation. See Greenway, 866 F.3d at 1100 (denying\\nhabeas relief on petitioner’s Eddings claim because the\\nArizona Supreme Court’s Eddings error was harmless).\\nThus, even assuming that the state courts unconstitutionally\\nexcluded that evidence, their factual determinations make\\nclear that the evidence would have been entitled to little\\nmitigating weight in their ultimate sentencing\\ndeterminations.\\n\\n “The harmless-error standard on habeas review provides\\nthat ‘relief must be granted’ only if the error ‘had substantial\\nand injurious effect or influence in determining the [sentence\\nimposed].’” McKinney, 813 F.3d at 822 (quoting Brecht,\\n\\n SPREITZ V. RYAN 55\\n\\n507 U.S. at 623). The United States Supreme Court has\\ndistinguished this harmless error standard from the standard\\napplicable to cases on direct review, which require merely a\\n“reasonable possibility” that the error affected the outcome\\nof the proceedings. See Brecht, 507 U.S. at 634–38 (citing\\nChapman v. California, 386 U.S. 18, 24 (1967)). Unlike\\ncases on direct review, “granting habeas relief merely\\nbecause there is a ‘reasonable possibility’ that [the] error\\ncontributed to the [outcome of the proceedings] is at odds\\nwith the historic meaning of habeas corpus.” Id. at 637\\n(citation and quotations omitted). As such, “an error that\\nmay justify reversal on direct appeal will not necessarily\\nsupport a collateral attack on a final judgment.” Id. at 634\\n(citation and quotations omitted).\\n\\n In McKinney, we clarified that a state court’s Eddings\\nviolation is harmless if we can “say[] with fair assurance”\\nthat “the judgment was not substantially swayed by the\\nerror.” 813 F.3d at 822 (quoting Kotteakos v. United States,\\n328 U.S. 750, 765 (1946)) (emphasis added). On the other\\nhand, an Eddings error is not harmless if we are in “grave\\ndoubt” about whether it had a substantial and injurious effect\\nor influence in determining the sentence imposed. Id.\\n(quoting O’Neal v. McAninch, 513 U.S. 432, 436 (1995)).\\nMcKinney held that the excluded evidence there “was\\nimportant mitigating evidence” because it was “central to\\n[McKinney’s] plea for leniency.” 813 F.3d at 823 (emphasis\\nadded). It cited Coleman v. Calderon, 210 F.3d 1047 (9th\\nCir. 2000), where we held that an unconstitutional jury\\ninstruction at sentencing was not harmless because “it\\nundermined the very core of Coleman’s plea for life.” Id. at\\n1051 (emphasis added).\\n\\n Here, based on what the Arizona courts repeatedly said,\\nthere is no “grave doubt” as to whether any Eddings error\\n\\n56 SPREITZ V. RYAN\\n\\nsubstantially influenced or swayed Spreitz’s ultimate\\nsentence. The courts at every level consistently found that\\n“the aggravating circumstance of especial cruelty in\\n[Spreitz’s] murder of Ruby Reid outweigh[ed] all factors\\nmitigating in favor of leniency.” Spreitz, 945 P.2d at 1282.\\nIn considering the entire record, the Arizona Supreme Court\\nexplained:\\n\\n We agree with the sentencing judge that\\n defendant’s upbringing was subnormal. The\\n record supports the judge’s conclusion that\\n defendant’s home life was sadly lacking and\\n that his mother’s erratic behavior toward\\n defendant inhibited his emotional\\n development and social skills . . . . Although\\n we recognize defendant’s upbringing as a\\n mitigating circumstance, we accord it little\\n weight. While defendant’s inadequate\\n upbringing may have contributed to his\\n emotional immaturity and undeveloped\\n humanitarian skills, we concur with\\n defendant’s statement at his sentencing\\n hearing that “people that have had as bad a\\n background or worse haven’t killed. And I\\n don’t want what everyone has said about my\\n background to be an excuse for what’s\\n happened.”\\n\\n ....\\n\\n The record demonstrates defendant’s\\n longtime substance abuse problems. We\\n note, however, that defendant’s general\\n problems with substance abuse are not\\n essential to our decision here. We therefore\\n\\n SPREITZ V. RYAN 57\\n\\ndecline to conclude that defendant was\\nimpaired by alcohol consumption to an extent\\nthat it interfered with his “capacity to\\nappreciate the wrongfulness of his conduct or\\nto conform his conduct to the requirements of\\nthe law.” A.R.S. § 13-703(G)(1) . . . .\\n\\n[W]e find that defendant expressed remorse\\nfor the victim’s death on more than one\\noccasion . . . . We recognize remorse as a\\nnonstatutory mitigating factor ....\\nHowever, defendant’s remorse for his actions\\ndoes little to counterbalance especial cruelty\\nas a serious aggravating circumstance in Ms.\\nReid’s murder. According to defendant’s\\nconfession, when he left Ms. Reid in the\\ndesert early the morning of May 19, 1989, he\\ndid not know whether she was alive or dead.\\nHe confessed that he rode his bicycle out to\\nthe murder site several days later to see if her\\nbody was still there, hoping that it would not\\nbe, that she was still alive. We would find\\ndefendant’s remorse a more compelling\\nmitigating factor if, for example, it had\\nprompted him to report his actions toward\\nMs. Reid to the authorities.\\n\\nThe sentencing judge found that defendant’s\\nability to appreciate the wrongfulness of his\\nconduct was not impaired on the night of the\\nmurder to any significant extent by substance\\nabuse, emotional disorders, situational stress,\\nor by a combination of these. Our review of\\nthe record convinces us that the trial court’s\\nfinding was proper . . . .\\n\\n58 SPREITZ V. RYAN\\n\\n We agree that the record supports the\\n sentencing judge’s findings that defendant\\n had no previous adult felony convictions, no\\n prior record of acts of violence, and that\\n defendant is capable of rehabilitation. We\\n also find that the sentencing judge correctly\\n rejected defendant’s age of twenty-two as a\\n mitigating circumstance and properly found\\n that his emotional immaturity was not a\\n significant mitigating factor.\\n\\n After examining the entire record and\\n reweighing the applicable aggravating and\\n mitigating factors, we find that the\\n aggravating circumstance of especial cruelty\\n in defendant’s murder of Ruby Reid\\n outweighs all factors mitigating in favor of\\n leniency.\\n\\n ....\\n\\n We have conducted an independent review of\\n defendant’s aggravating and mitigating\\n circumstances as required by A.R.S. section\\n 13-703.01 and find that the mitigating\\n circumstances cumulatively are not\\n sufficiently substantial to warrant leniency in\\n relation to the aggravating circumstance of\\n cruelty.\\n\\nId. at 1280–83.\\n\\n SPREITZ V. RYAN 59\\n\\n Despite this explanatory language summarizing its\\nanalysis of all of the evidence from the sentencing hearing,\\nthe majority nonetheless insists that the Arizona Supreme\\nCourt refused to consider Spreitz’s long-term substance\\nabuse as a non-statutory mitigating factor, and concludes this\\nrefusal “had a substantial and injurious effect on [Spreitz’s\\nsentence].” Op. at 35–36. My colleagues reason that\\n“Spreitz’s long-term alcohol and substance abuse was\\nimportant mitigation evidence . . . central to his plea for\\nleniency.” Id. The record belies that characterization.\\n\\n A\\n\\n Unlike McKinney and Coleman, Spreitz’s history of\\nsubstance abuse unrelated to the murder was not “central” to\\nhis plea for leniency or at the very core of his plea for life.\\nInstead, Spreitz argued at sentencing that he deserved\\nleniency because of, among other things, (1) his intoxication,\\nage, and emotional immaturity at the time of the crime;\\n(2) his disruptive childhood; (3) his abusive home\\nenvironment; (4) his emotionally cold mother; (5) his poor\\nsocial adjustment; (6) a persistent pattern of rejection; (7) the\\nabsence of a healthy role model; (8) a lack of economic and\\nemotional support; (9) his lack of a high school education;\\n(10) the absence of adult convictions and violent behavior;\\n(11) his low risk of future danger; (12) his good character;\\n(13) his good behavior while in jail; and (14) his\\nrehabilitative nature. See, e.g., Spreitz, 945 P.2d at 1279\\n(“At sentencing, defendant argued as statutory and\\nnonstatutory mitigating factors: (1) his dysfunctional family\\nlife and lack of socialization; (2) a history of alcohol and\\ndrug abuse; (3) his expressions of remorse; (4) his impaired\\ncapacity to appreciate the wrongfulness of his conduct,\\nA.R.S. § 13-703(G)(1); (5) his good behavior while\\nincarcerated; (6) his lack of adult convictions; (7) no prior\\n\\n60 SPREITZ V. RYAN\\n\\nrecord of violent tendencies; and (8) his age at the time of\\nthe murder, A.R.S. § 13-703(G)(5).”). Specifically,\\nSpreitz’s sentencing memorandum posited that:\\n\\n The Defendant’s background and upbringing\\n are ripe with numerous mitigating factors:\\n\\n 1. He faced a disruptive middle childhood.\\n\\n 2. Punitive and abusive behavior toward\\n [Spreitz].\\n\\n 3. An emotionally cold mother.\\n\\n 4. Poor social adjustment with peers.\\n\\n 5. No healthy role model.\\n\\n 6. Drug and alcohol abuse.\\n\\n 7. [That] Spreitz was forced to live on his\\n own without economic or emotional support.\\n\\n 8. Persistent pattern of rejection.\\n\\n In addition, Spreitz devoted a large portion of his\\nsentencing memorandum to explaining how his childhood\\nabuse, dysfunctional upbringing, developmental immaturity,\\nand emotional problems contributed to his commission of\\nthe murder:\\n\\n SPREITZ V. RYAN 61\\n\\nFor Chris, as with other chronically rejected,\\nneglected and devalued children, there was\\nno escaping [his] deep seated anger and\\nresentment. The desperate, quiet, battle\\nwrestling inside of Chris as he sought to\\nplease [his mother] and was rejected could\\nnot be numbed forever by alcohol. His\\ndevelopmental failure included the non-\\ndevelopment of personal insight and conflict\\nresolution skills. He could not control,\\nunderstand or deal with the anger. Numbing\\nby use of alcohol and trysts with older women\\nwere but stop-gap measures. His fragile\\nfacade was ready to crack. Each failure\\nbrought Chris closer to the edge. He crossed\\nthat point with the victim. The result was\\ninevitable [as] Chris had no way to\\ncompensate for a lifetime of misdirected\\ndevelopment, through a dysfunctional life\\nand family, especially when combined with\\narrested and distorted development and\\nalcoholism . . . .\\n\\nChris’s background is a relevant and\\npowerful mitigating factor with a\\ndeterminative factor culminating in the\\ninstant offense. When a person’s background\\nhas an impact [on] his behavior beyond his\\ncontrol, as here, such a circumstance is\\nclearly mitigation entitled to great\\n[deference]. Add to that the family and\\nabusive environment and [this] mitigation\\nclearly outweighs any possible claim of\\naggravation . . . . The situational stress build\\nup, including rejection by his mother, his\\n\\n62 SPREITZ V. RYAN\\n\\n girlfriend, and again that night, coupled with\\n poorly developed coping skills and alcohol\\n certainly mitigates the act.\\n\\n(Emphasis added.) 3\\n\\n But the majority nonetheless wishes to recast the defense\\ntheories and reweigh the facts, as if we were conducting a\\ndirect review. The majority singles out Dr. Flynn’s\\nobservations that (1) Spreitz began drinking alcohol and\\nsmoking marijuana at age twelve or thirteen, (2) Spreitz\\ndrank steadily by age fifteen, (3) Spreitz’s “alcohol abuse\\ncontinued to intensify after he left home,” (4) Spreitz’s\\nalcoholism reached the level of physical dependence, and\\n(5) a “variety of persons” described Spreitz as a heavy\\ndrinker, as intoxicated most of the time, and as having\\nblackouts while drinking alcohol. Those observations,\\nhowever, do not make Spreitz’s evidence of his longstanding\\nalcoholism, with no connection to the crime, central to his\\nplea for leniency.\\n\\n Instead, Spreitz presented his long-term substance abuse\\nprimarily to establish the statutory mitigating factors under\\n§ 13-703(G)(1) and § 13-703(G)(5). As to § 13-703(G)(1),\\nDr. Flynn repeatedly emphasized that Spreitz’s lifetime of\\nsubstance abuse contributed to his murder of Ms. Reid.\\nNotably, Dr. Flynn’s report opined that:\\n\\n The research on violent behavior shows a\\n strong correlation with alcohol intoxication.\\n 3\\n Likewise, Spreitz claims in this appeal that his lawyer was\\nineffective at sentencing by failing to present witnesses who had\\nextensive evidence about his early development, family history, head\\ninjuries, substance abuse, and emotional problems. We have rejected\\nthat claim in a contemporaneously filed memorandum disposition.\\n\\n SPREITZ V. RYAN 63\\n\\nGiven [Spreitz’s] history, I can see a\\nperceived rejection by two women in the\\nsame evening as especially disturbing to him,\\nparticularly when intoxicated, to the point at\\nwhich he lost aggressive control, disinhibited\\nby the alcohol intoxication . . . .\\n\\nMy best guess is that he was drunk, hurt and\\nangry at the unwillingness of his girlfriend to\\nlet him into her apartment, startled into an\\nalcoholic rage at the angry confrontation by\\nthe victim, and ended up venting the years of\\nstored up rage at her.\\n\\nAnd although several factors that normally\\npredict violence were not present, there is\\nalcoholism—one of the strongest correlates\\nof violent behavior. And, there is the\\nemotional, sexual maladjustment that comes\\nfrom his lifetime of misdirected development\\nin a dysfunctional family, combined with\\narrested and/or distorted adult development\\nbecause of the alcoholism . . . .\\n\\n1. Given the history of alcoholism . . . a\\nsignificant but unknown degree of alcohol\\nintoxication is likely. Alcohol intoxication\\nhas a well-documented disinhibiting effect\\nwhich frequently includes losses of control of\\nangry emotions and aggressive behavior.\\n\\n2. [Spreitz’s] history strongly suggests years\\nof early experiences likely to have caused a\\nbuild-up of pent-up angry, aggressive\\nfeelings toward women generally (and older\\n\\n64 SPREITZ V. RYAN\\n\\n women especially) which may have burst\\n forth with uncontrollable intensity with or\\n without alcohol intoxication. Only trivial\\n provocation is required for this type of\\n aggression explosion, termed an Intermittent\\n Explosive Disorder by DSM-IV.\\n\\n 3. Still more likely is that a combination of\\n 1. and 2. above contributed to an\\n uncontrollable outburst of aggression.\\n\\n(Emphasis added.)\\n\\n Dr. Flynn also gave extensive testimony on how\\nSpreitz’s substance abuse contributed to the impairment of\\nhis mental state “in terms of conforming his behavior to his\\nunderstanding of the law and right and wrong generally.” At\\nthe aggravation-mitigation hearing, Dr. Flynn went to great\\nlengths to opine that Spreitz’s alcohol abuse was “very\\nstrongly correlated with violent behavior.” Dr. Flynn also\\ntestified that Spreitz’s parents’ failure to treat his alcoholism\\nduring his teenage years was significant because it increased\\nhis violence potential: “[A]lcohol is probably the most\\ncommon, strongest correlative . . . with violent behavior\\n. . . , doubling the risk of violent behavior and I think the\\nabsence of successful treatment for alcoholism is likely to\\nhave been a significant and maybe a sine qua non that went\\non in this offense.” (Emphasis added.)\\n\\n In light of the evidence presented by Dr. Flynn, it is hard\\nto say that the Arizona courts did not consider the effect of\\nboth his intoxication and his history of substance abuse as\\nrequired under § 13-703(G)(1). They just weren’t persuaded\\nby it. Rather, they reasonably and properly found that\\nSpreitz’s history of alcohol abuse was not sufficiently\\n\\n SPREITZ V. RYAN 65\\n\\ninfluential on his ability to conform his conduct to the law or\\nhis ability to appreciate the consequences at the time he\\nviciously murdered Ms. Reid. Contrary to the majority’s\\nsuggestion, Op. at 28–29, it cannot be said that the Arizona\\ncourts never considered his longstanding alcoholism under\\n§ 13-703(G)(1) in the first place.\\n\\n As to the statutory mitigating factor under\\n§ 13-703(G)(5), Dr. Flynn explained how Spreitz’s history\\nof substance abuse contributed to his emotional immaturity:\\n“When you spend most of your teenage years drunk or\\nstoned, it has not only been my observation in criminal\\npopulations but my observations clinically . . . that virtually\\nall heavily drugged teenagers [are] socially and emotionally\\nimmature as having substance abuse problems.” (Emphasis\\nadded.) Dr. Flynn also testified that the lack of a treatment\\neffort in Spreitz’s home to deal with his alcoholism was\\nsignificant because it showed a lack of proper supervision\\nand parenting during his developing years, which\\ncontributed to his emotional immaturity under\\n§ 13-703(G)(5):\\n\\n [The lack of any treatment effort for Spreitz’s\\n alcoholism] is consistent with an uncaring\\n parent to have a child which is stealing booze\\n out of the house from the early teenage years,\\n who shows up drunk in his early teenage life\\n and everybody in the person’s life recognizes\\n there’s a serious alcoholic problem and yet\\n the parent either does one of two things,\\n doesn’t care enough to notice or doesn’t care\\n enough to do anything about it. That’s\\n important in and of itself in my opinion from\\n the standpoint of an evaluation of parenting.\\n\\n66 SPREITZ V. RYAN\\n\\n Thus, where Spreitz did not present his long-term\\nsubstance abuse to explain his conduct at the time of the\\nmurder under § 13-703(G)(1), he did so to plead for leniency\\nbased on his emotional immaturity under § 13-703(G)(5).\\nAny evidence that Spreitz presented of his long-term\\nsubstance abuse as non-statutory mitigation was cursory and\\nonly ancillary to the considerable evidence he adduced to\\nexplain how his past addiction affected his conduct and\\nimmaturity at the time of the murder.\\n\\n Moreover, the significant evidence presented regarding\\nSpreitz’s abusive and dysfunctional childhood, upbringing,\\nand interactions with his mother stands as further support\\nthat Spreitz’s history of alcohol and substance abuse was not\\ncentral to his plea for leniency. Dr. Flynn testified to those\\nmatters at length:\\n\\n [Spreitz grew up in] a fully pathological\\n home environment, [which] include[d]\\n physical and emotional neglect, physical\\n abuse and emotional neglect and possibly\\n emotional abuse by the mother according to\\n which there was no hint of acceptance of him\\n in spite of ongoing efforts on his part to\\n please her. In fact of all the people that I\\n interviewed and all the interviews that I read,\\n the only favorable, the only person who had\\n a single favorable thing to say about the\\n mother was Mr. Spreitz himself. From all\\n other persons they described her in the most\\n negative terms. His sister described herself\\n as being saved from the pathological effects\\n of the home by the fact she ran away and was\\n placed in a group home. Mr. and Mrs. Spreitz\\n described most of the mother’s behavior in a\\n\\n SPREITZ V. RYAN 67\\n\\n manner that would be considered pathogenic\\n by anyone who knew anything about\\n psychology and parenting . . . .\\n\\n [There is] massive evidence from a\\n psychological point . . . that there was a cold\\n uncaring mother who was anything but\\n sensitive to his needs, whom he was never\\n able to please. By this hypothesis it is most\\n common for children so treated to harbor a\\n store of anger, and it is consistent with this\\n train of thought that the anger was breaking\\n down partially by chronicity and partially by\\n series of emotionally unsuccessful\\n relationships with women, a tendency to\\n orient toward older women in the months\\n leading up to the offense[.] [S]hortly before\\n the offense . . . there had been a rejection by\\n a current girlfriend . . . . [T]his by itself and\\n probably especially this under the\\n disinhibiting influence of alcohol possibly\\n result[ed] in an explosion of aggressive\\n behavior especially toward a woman and\\n especially toward an older woman. The\\n recently pent-up rage accumulated over the\\n year aggravated the more recent months\\n which may be considered by the Court to\\n constitute an outburst of behavior that means\\n an inability to conform one’s behavior [to the\\n requirements of the law].\\n\\n(Emphasis added.)\\n\\n And, when Judge Sherrill asked Dr. Flynn what he\\nconsidered were non-statutory mitigating factors in this\\n\\n68 SPREITZ V. RYAN\\n\\ncase, Dr. Flynn responded: “[T]he likely relationship\\nbetween the alcoholism and the violence generally and this\\noffense specifically and . . . the deprived pathogenic home\\nenvironment [are what] I would consider non-statutory\\nfactors.” (Emphasis added.) In the face of this record, I am\\nat a loss to understand how my colleagues can say that\\nSpreitz’s history of substance abuse as a non-statutory\\nmitigating factor was central to or at the very core of\\nSpreitz’s plea for leniency. 4\\n\\n B\\n\\n Two other cases where we have found Arizona courts’\\nEddings errors to be harmless, Henry v. Ryan, 720 F.3d 1073\\n(9th Cir. 2013), and Stokley v. Ryan, 705 F.3d 401 (9th Cir.\\n2012), are instructive here.\\n\\n 1\\n\\n In Henry, the Arizona courts found a statutory mitigating\\nfactor under § 13-703(G)(1)—namely, that “Henry’s\\ncapacity to appreciate the wrongfulness of his conduct or to\\nconform his conduct to the requirements of law was\\nsignificantly impaired . . . because of his consumption of\\nalcohol on the day of the murder.” 720 F.3d at 1089–90.\\nNonetheless, the courts imposed the death penalty because\\nthey “agreed that the mitigation was not sufficiently\\n\\n\\n\\n 4\\n The majority does not take issue with the fact that the Arizona\\ncourts considered Spreitz’s substance abuse history in the statutory\\nmitigating context or that they found it insufficient to outweigh the\\naggravating factor of cruelty. “When applied solely in the context of\\nstatutory mitigation,” the “causal nexus test does not violate Eddings.”\\nMcKinney, 813 F.3d at 810.\\n\\n SPREITZ V. RYAN 69\\n\\nsubstantial to call for leniency” in light of two aggravating\\nfactors. Id. at 1090.\\n\\n As with Spreitz here, on habeas review, we assumed that\\nthe state courts violated Eddings by failing to consider\\nHenry’s history of alcohol abuse as a non-statutory\\nmitigating factor. However, we denied habeas relief because\\nHenry failed to show that the Eddings error had a\\n“substantial and injurious effect or influence” on his\\nsentence. Id. at 1087–89 (quoting Brecht, 507 U.S. at 623).\\nWe reasoned that, “in imposing the death penalty, the\\nArizona courts already considered Henry’s intoxication at\\nthe time of the murder as a mitigating factor” and concluded\\nthat it was “not sufficient to call for leniency.” Id. at 1090.\\nAnd:\\n\\n Given the similar nature of the mitigation, the\\n additional evidence of Henry’s historical\\n alcoholism would have had minimal\\n mitigating value. If the state courts\\n concluded that intoxication with a causal\\n connection to the crime was not sufficient to\\n call for leniency, it is highly doubtful that\\n they would have considered alcoholism\\n without a causal connection to be sufficient.\\n\\nId. 5\\n\\n Similarly, here, the Arizona courts agreed that Spreitz’s\\nsubstance abuse at the time of the offense was not sufficient\\nto call for leniency. In addition, they found that Spreitz\\ncould not show statutory mitigation under § 13-703(G)(1)\\n\\n 5\\n Although Henry involved different facts, as the majority points out,\\nits reasoning applies fully to Spreitz’s case. Op. at 36, n.15.\\n\\n70 SPREITZ V. RYAN\\n\\nbecause his “ability to appreciate the wrongfulness of his\\nconduct was not impaired on the night of the murder to any\\nsignificant extent by substance abuse.” Spreitz, 945 P.2d at\\n1281. The majority does not take issue with these findings.\\nAs in Henry, the Arizona courts’ consideration of Spreitz’s\\nlong-term substance abuse unrelated to the offense would\\nhave had minimal mitigating value because of its similarity\\nto the evidence already rejected. Henry, 720 F.3d at 1090;\\nsee also Landrigan, 272 F.3d at 1230 & n.9 (holding that,\\nwhere the sentencing court rejected the petitioner’s “alleged\\nintoxication and past history of drug use as a statutory\\nmitigating factor” and “did not go on to consider them as a\\nnonstatutory mitigating factor,” any “error in failing to\\nconsider Landrigan’s use of alcohol and drugs would have\\nbeen inconsequential; it would have had no effect\\nwhatsoever on the outcome”); Wong v. Belmontes, 558 U.S.\\n15, 23 (2009) (“The sentencing jury was thus ‘well\\nacquainted’ with Belmontes’ background and potential\\nhumanizing features. Additional evidence on these points\\nwould have offered an insignificant benefit, if any at all.”\\n(citation omitted)).\\n\\n The majority thus errs by failing to apply Henry’s\\nreasoning to conclude that any Eddings error was harmless\\nhere. See Hart v. Massanari, 266 F.3d 1155, 1172 (9th Cir.\\n2001) (“[I]f a controlling precedent is determined to be on\\npoint, it must be followed.”). If direct evidence of\\nsubstance-induced intoxication on the night of the murder\\nwas insufficient to outweigh the aggravated circumstances\\nof the crime, a long history of drug and alcohol abuse would\\nfare no better.\\n\\n 2\\n\\n In Stokley, we assumed that the Arizona Supreme Court\\ncommitted Eddings error by failing to consider Stokley’s\\n\\n SPREITZ V. RYAN 71\\n\\nabusive childhood and good behavior in jail because of a\\nlack of causal nexus. 705 F.3d at 403–04. We held,\\nhowever, that this error was harmless because there were\\nthree aggravating circumstances, including that Stokley—\\nlike Spreitz—had “committed the offense in an especially\\nheinous, cruel, and depraved manner.” Id. at 404. We noted\\nthe gruesome details of Stokley’s offense, including (1) that\\nhe had sexual intercourse with his victims; (2) that the\\nvictims’ bodies (two young girls) were violently beaten; and\\n(3) that the victims’ bodies were dragged to and thrown\\ndown a mineshaft. Id. at 405 n.1. We also observed that\\nboth the sentencing court and the Arizona Supreme Court\\nconsistently determined that “leniency was inappropriate.”\\nId. at 405. We found no likelihood “that, but for a failure to\\nfully consider Stokley’s family history or his good behavior\\nin jail during pre-trial incarceration, the Arizona courts\\nwould have come to a different conclusion” as to Stokley’s\\nsentence. Id.\\n\\n Although Stokley’s facts were relatively more egregious\\nthan the facts here—Stokley involved three aggravating\\ncircumstances instead of one and Stokley’s victims were\\nunder the age of fifteen—Stokley nonetheless informs us that\\nthe severity of an aggravating circumstance must be\\nconsidered when determining whether an Eddings error is\\nharmless. Here, both the sentencing court and the Arizona\\nSupreme Court consistently ruled that leniency was\\ninappropriate because of the especially cruel manner in\\nwhich Spreitz kidnapped and murdered Ruby Reid. Spreitz,\\n945 P.2d at 1266, 1278–79 (citing Ariz. Rev. Stat. § 13-\\n703(F)(6)). The crime scene photographs alone give\\ncompelling support to that ruling, showing Ms. Reid’s body\\nbeaten beyond recognition by heavy rocks. The pathologist\\nunderscored the extensive damage to her body revealed on\\nautopsy.\\n\\n72 SPREITZ V. RYAN\\n\\n Arizona courts consider five factors in “determining\\nwhether a murder was especially heinous, cruel or depraved:\\n(1) relishing the murder, (2) infliction of gratuitous violence,\\n(3) needless mutilation of the victim, (4) senselessness of the\\ncrime, and (5) helplessness of the victim.” Gulbrandson v.\\nRyan, 738 F.3d 976, 984 (9th Cir. 2013) (citing State v.\\nGretzler, 659 P.2d 1, 11–12 (Ariz. 1983) (en banc)). “A\\nfinding of cruelty is warranted when the defendant inflicts\\non the victim mental anguish or physical abuse before the\\nvictim’s death.” Spreitz, 945 P.2d at 1278. “Cruelty is found\\nwhen the victim is conscious at the time of the offense in\\norder to suffer pain and distress.” Id. (citation and alterations\\nomitted).\\n\\n Here, the Arizona courts found that Ms. Reid suffered\\ntremendous mental and physical pain when she was forced\\ninto Spreitz’s car trunk and transported to the desert where\\nshe was beaten, sexually assaulted, and eventually murdered.\\nId. at 1265, 1278. Judge Sherrill observed: “At the scene of\\nthe murder the victim was thoroughly beaten. The autopsy\\nshowed, in addition to the death-causing blows to the head,\\nfive ribs broken, eighteen bruises and grab marks, internal\\nbruising and bleeding, lacerations[,] and a broken jaw.”\\nSpreitz did not confess to kidnapping Ms. Reid, but did\\nadmit to beating her as she fought back, removing her\\nclothes, having intercourse with her, and smashing her in the\\nhead with a rock when she would not stop yelling. Id.\\nPhysical evidence found at the scene of the murder\\ncorroborated Spreitz’s confession: Ms. Reid’s clothing,\\nincluding her torn brassiere, was strewn in one area of the\\nscene, and photographs of the area vividly depicted drag\\nmarks running from that area to the spot where Ms. Reid’s\\nbody and the bloody rocks were found. According to\\nSpreitz’s confession, when he left Ms. Reid lying in the\\ndesert on the morning of May 19, 1989, he did not know\\n\\n SPREITZ V. RYAN 73\\n\\nwhether she was still alive or dead. Also significant to Judge\\nSherrill’s finding of mental anguish was that Ms. Reid had\\ndefecated in and on her clothing. See id. at 1265, 1278–79.\\n\\n In its independent review of these facts, the Arizona\\nSupreme Court agreed that Spreitz had murdered Ms. Reid\\nin an especially cruel manner. Id. at 1278. The Arizona\\nSupreme Court highlighted Spreitz’s “own admission that he\\nbeat her as she fought back and hit her with the rock when\\nshe would not stop yelling,” which it found was “clear\\nevidence of [Ms. Reid’s] conscious suffering.” Id. at 1279.\\nThe court found that Spreitz’s confession and the physical\\nevidence all pointed to the conclusion that Spreitz “beat and\\nraped [Ms. Reid] in a brutal assault that lasted many minutes\\nbefore he crushed her skull.” Id. My colleagues in the\\nmajority do not quarrel with that finding.\\n\\n Thus, even assuming that the Arizona courts failed to\\nconsider Spreitz’s evidence of long-term substance abuse\\nunrelated to the crime, it is unlikely their consideration of\\nthat evidence would have substantially altered their ultimate\\ndecision to sentence Spreitz to death based on the aggravated\\nmanner in which Ms. Reid met her death. And we cannot\\nsay “with fair assurance” that the sentence they imposed was\\nsubstantially swayed by their failure to consider that\\nevidence. McKinney, 813 F.3d at 822 (quoting Kotteakos,\\n328 U.S. at 765).\\n\\n Further, Eddings allows the Arizona courts “to consider\\nthe absence of a causal connection when assessing the\\nquality and strength of [mitigating] evidence.” Henry,\\n720 F.3d at 1090. It does not prevent the Arizona courts\\nfrom “assign[ing] less weight to mitigating factors that did\\nnot influence a defendant’s conduct at the time of the crime.”\\nHedlund, 854 F.3d at 587 n.23. The Arizona Supreme Court\\nhas recognized this principle in its more recent cases. See,\\n\\n74 SPREITZ V. RYAN\\n\\ne.g., State v. Newell, 132 P.3d 833, 849 (Ariz. 2006) (en\\nbanc) (“We do not require that a nexus between the\\nmitigating factors and the crime be established before we\\nconsider the mitigation evidence. But the failure to establish\\nsuch a causal connection may be considered in assessing the\\nquality and strength of the mitigation evidence.” (citation\\nomitted)). Thus, we can confidently say that if Spreitz were\\nresentenced, the Arizona courts would give even less weight\\nto his history of substance abuse unrelated to the crimes in\\nbalancing that evidence against what he did to Ruby Reid.\\n\\n C\\n\\n The majority cites Henry for the broad proposition that\\n“a ‘history of substance abuse [is] substantially mitigating\\nwhen . . . combined with other mitigating evidence.’” Op.\\nat 36 (emphasis omitted) (quoting Henry, 720 F.3d at 1090).\\nBut the majority has not offered up any authority that reaches\\nthis conclusion under factually-similar circumstances.\\nIndeed, unlike the three cases Henry relies upon in making\\nthis statement, Spreitz’s defense counsel presented\\nconsiderable evidence at sentencing of his lifelong history of\\nsubstance abuse, his mental and emotional problems, and his\\nfamily dysfunction. Counsel retained Dr. Flynn, a\\npsychotherapist and forensic consultant, who personally\\nevaluated Spreitz, performed psychological assessments,\\nand interviewed people who were familiar with him. Dr.\\nFlynn submitted a report and testified extensively regarding\\nSpreitz’s alcohol and drug addictions, his family history of\\nsubstance abuse, his disruptive childhood, and his emotional\\nproblems. Thus, the majority’s suggestion that Spreitz’s\\ncase is comparable to those in which we have found evidence\\nof longstanding substance abuse to be substantially\\nmitigating is misplaced. This is particularly so in light of the\\nhorrific way in which Ruby Reid died, the impact of which\\n\\n SPREITZ V. RYAN 75\\n\\nwas not lost on all of the Arizona judges who reviewed this\\nrecord.\\n\\n D\\n\\n Lastly, Spreitz’s own arguments on appeal belie the\\nconclusion that his long-term substance abuse unrelated to\\nthe crime was central to his plea for leniency. Spreitz argues\\nthat whether he “was intoxicated at the time of the murder\\nwas critical” to his case, and that “the issue of [his]\\nintoxication [at the time of the crime] was critically\\nimportant.” He contends that he “was highly intoxicated at\\nthe time of the murder” and that “this was the primary cause\\nof him committing the murder.” And, he urges that there\\n“was ample evidence that [he] was highly intoxicated at the\\ntime of the murder and that his intoxication contributed\\ndirectly to his actions.” “This was highly relevant\\nmitigation,” he says: “[h]ad the judge heard the evidence\\ndescribed herein, it is highly probable that he would not have\\nsentenced [Spreitz] to death.” But the judge did hear all of\\nthat evidence. He just didn’t buy the argument based on the\\ntestimony of officers who interacted with Spreitz before and\\nafter he killed Ms. Reid. That credibility determination is\\nnot assailable on appeal.\\n\\n Spreitz’s defense theories have shifted on collateral\\nappeal. Contrary to his position in supplemental briefing\\nordered by this Court post-McKinney, Spreitz does not make\\nany similar arguments about his long-term substance abuse\\nas a non-statutory mitigating factor. Instead, he claims that\\nevidence of his childhood head injuries that his counsel\\nfailed to present were vital to his mitigation evidence.\\nSpreitz argues that the “importance of counsels’ failure to\\ninvestigate [his] history of head injuries cannot be\\noverstated;” that the “connection between childhood head\\ninjuries and impulsive murder” was “well established” at the\\n\\n76 SPREITZ V. RYAN\\n\\ntime of sentencing; that “there was a correlation between\\n[his] head injuries as a child and his impulsive behavior in\\nthe instant case;” and that “[expert] testimony could have\\nenlightened the court as to the connection between childhood\\nhead injuries and impulsive behavior in murderers in\\ngeneral, and could have developed the link between\\n[Spreitz]’s own head injuries and his impulsive behavior\\nduring this murder in particular.” He contends that:\\n\\n The murder in this case was a classic\\n impulsive act. It was not thought out or\\n premeditated beforehand. Appellant beat the\\n victim with a rock; he did not even bring a\\n weapon. The victim refused to have sex with\\n him, fought him, and he became enraged and\\n impulsively killed her. That is the classic\\n behavior of a murderer who suffered\\n childhood head injuries.\\n\\n(Emphasis added.) Spreitz’s shifting defense theories\\nundermine his latest argument that long-term substance\\nabuse was central to his plea for leniency all along.\\n\\n We will never know how Ruby Reid, forced into the\\ntrunk of Spreitz’s car and pondering her fate as he drove her\\ninto the desert, would have answered any of Spreitz’s\\ncontentions. But the forensic evidence of the way in which\\nshe died stands in silent, but compelling refutation of that\\ndefense. Simply put, Spreitz cannot show on this record that\\nhis long-term substance abuse, as a non-statutory mitigating\\nfactor and without any connection to the crime, was either\\n“central to his plea for leniency,” McKinney, 813 F.3d at\\n823, or at “the very core of [his] plea for life,” Coleman,\\n210 F.3d at 1051. Therefore, he cannot demonstrate that the\\nArizona courts’ alleged failure to consider that evidence\\n\\n SPREITZ V. RYAN 77\\n\\nsubstantially influenced or swayed his ultimate sentence.\\nNor can he show that, had the Arizona courts further\\nconsidered its marginal mitigating weight, the evidence\\nwould have substantially affected the outcome of his\\nsentence. See Brecht, 507 U.S. at 623. Because Spreitz\\ncannot show that the error he alleges was not harmless, the\\ndistrict court’s denial of habeas relief on Spreitz’s Eddings\\nclaim should be affirmed.\\n\\n From the majority’s conclusion otherwise, I respectfully\\ndissent.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4373822/\", \"author_raw\": \"TALLMAN, Circuit Judge, dissenting\"}]}"},"judge1":{"kind":"string","value":"RICHARD A PAEZ"},"judge2":{"kind":"string","value":"MARSHA S BERZON"},"judge3":{"kind":"string","value":"RICHARD C TALLMAN"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":2,"string":"2"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":1,"string":"1"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4596569/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Christopher Spreitz v. Charles Ryan\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-04\nOpinion type: majority\nHas dissenting opinion: Yes\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nCHRISTOPHER J. SPREITZ, No. 09-99006\n Petitioner-Appellant,\n D.C. No.\n v. 4:02-CV-00121-JMR\n\nCHARLES L. RYAN, Warden,\n Respondent-Appellee. OPINION\n\n\n\n Appeal from the United States District Court\n for the District of Arizona\n John M. Roll, District Judge, Presiding\n\n Argued July 11, 2013\n Submitted March 4, 2019\n San Francisco, California\n\n Filed March 4, 2019\n\n Before: Richard A. Paez, Marsha S. Berzon,\n and Richard C. Tallman, Circuit Judges.\n\n Opinion by Judge Paez;\n Dissent by Judge Tallman\n\n2 SPREITZ V. RYAN\n\n SUMMARY *\n\n\n Habeas Corpus / Death Penalty\n\n The panel reversed the district court’s denial of habeas\ncorpus relief with respect to Christopher J. Spreitz’s death\nsentence, and remanded, in a case in which Spreitz argued\nthat the Arizona Supreme Court violated Eddings v.\nOklahoma, 455 U.S. 104 (1982), by refusing to consider, as\na matter of law, mitigating evidence of Spreitz’s\nlongstanding alcohol and substance abuse on the ground that\nhe did not establish a causal connection between this\nmitigating evidence and the crime.\n\n The panel held that the district court erred in concluding\nthat Spreitz’s claim that the Arizona Supreme Court violated\nEddings is procedurally defaulted. The panel explained that\nthe first opportunity Spreitz had to raise that claim was\nbefore the post-conviction-relief (PCR) court, at which time\nhe did so.\n\n Because the decision of the PCR court – which first\ndeclared the claim waived, but proceeded to adjudicate the\nclaim on the merits – was contrary to clearly established\nSupreme Court precedent, the panel accorded that decision\nno deference and reviewed Spreitz’s Eddings claim de novo.\nThe panel concluded that the Arizona Supreme Court\nviolated Eddings by impermissibly requiring that Spreitz\nestablish a causal connection between his longstanding\nsubstance abuse and the murder before considering and\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\n SPREITZ V. RYAN 3\n\nweighing the evidence as a nonstatutory mitigating factor.\nThe panel concluded that the error was not harmless.\n\n The panel affirmed the district court’s judgment denying\nrelief with respect to Spreitz’s conviction in a concurrently\nfiled memorandum disposition.\n\n Dissenting, Judge Tallman wrote that the record does not\nestablish that either the sentencing court or the Arizona\nSupreme Court unconstitutionally refused to consider\nrelevant mitigating evidence; and that even if the Arizona\ncourts did violate Eddings, Spreitz cannot show that this\nerror had a “substantial and injurious effect or influence” on\nhis ultimate sentence.\n\n\n COUNSEL\n\nTimothy M. Gabrielsen (argued), Assistant Federal Public\nDefender; John M. Sands, Federal Public Defender; Office\nof the Federal Public Defender, Tucson, Arizona; Susan B.\nFox and Sean Bruner, Law Office of Sean Bruner Ltd.,\nTucson, Arizona; for Petitioner-Appellant.\n\nLacey Stover Gard (argued) and Jeffrey A. Zick, Section\nChief Counsel; Kent E. Cattani, Chief Counsel, Criminal\nAppeals/Capital Litigation Section; Mark Brnovich,\nAttorney General; Office of the Attorney General, Tucson,\nArizona; for Respondents-Appellees.\n\n4 SPREITZ V. RYAN\n\n OPINION\n\nPAEZ, Circuit Judge:\n\n In 1994, an Arizona jury convicted Christopher J. Spreitz\n(“Spreitz”) of first-degree murder. The victim was thirty-\nnine year old Ruby Reid (“Reid”). Finding that the cruelty\nof the murder outweighed any mitigating circumstances, the\ntrial judge sentenced Spreitz to death. Spreitz appeals the\ndistrict court’s denial of his petition for a writ of habeas\ncorpus challenging his conviction and sentence. We affirm\nthe district court’s judgment with respect to Spreitz’s\nconviction, 1 and reverse with respect to his sentence. 2\n\n In challenging his sentence, Spreitz argues that the\nArizona Supreme Court unconstitutionally affirmed his\ndeath sentence by failing to consider mitigating evidence of\nhis longstanding alcohol and substance abuse. He contends\nthat the state court refused to consider, as a matter of law,\nthis evidence in mitigation because he did not establish a\ncausal connection between the crime and his long-term\nalcohol and substance abuse. In Eddings v. Oklahoma,\n455 U.S. 104, 110 (1982), the Supreme Court held that under\nboth the Eighth and Fourteenth Amendments, a sentencer in\na capital case may not “refuse to consider, as a matter of law,\n\n 1\n We affirm the judgment with respect to Spreitz’s conviction in a\nconcurrently filed memorandum disposition.\n\n 2\n On February 13, 2015, we vacated submission of Spreitz’s case\npending final resolution of the en banc proceedings in McKinney v. Ryan,\n730 F.3d 903 (9th Cir. 2013). In December 2015, the en banc court\nissued an opinion in McKinney. 813 F.3d 798 (9th Circ. 2015) (en banc).\nAfter the Supreme Court denied the State’s petition for a writ of\ncertiorari, Ryan v. McKinney, 137 S. Ct. 39 (2016) (mem), we ordered\nsupplemental briefing on the applicability of McKinney to Spreitz’s case.\n\n SPREITZ V. RYAN 5\n\nany relevant mitigating evidence” offered by the defendant.\nId. at 114. Although a sentencer “may determine the weight\nto be given relevant mitigating evidence . . . they may not\ngive it no weight by excluding such evidence from their\nconsideration.” Id. at 114–15 (footnote omitted). In\ninterpreting and applying Eddings, the Supreme Court has\nexplained that “full consideration of evidence that mitigates\nagainst the death penalty is essential if the [sentencer] is to\ngive a reasoned moral response to the defendant’s\nbackground, character, and crime.” Penry v. Lynaugh\n(Penry I), 492 U.S. 302, 328 (1989), abrogated on other\ngrounds by Atkins v. Virginia, 536 U.S. 304 (2002) (internal\nquotation marks and citation omitted). Moreover, the\nSupreme Court has been clear: requiring a defendant to\nprove a causal nexus between his mitigating evidence and\nthe crime is “a test we never countenanced and now have\nunequivocally rejected.” Smith v. Texas, 543 U.S. 37, 45\n(2004) (per curiam).\n\n At the time of Spreitz’s sentencing, Arizona Revised\nStatute Annotated § 13-703(G)(1994) 3 listed five mitigating\nfactors, and Arizona case law additionally recognized\nnonstatutory mitigating factors, including, for example, a\ndefendant’s difficult family background or mental condition\nnot severe enough to qualify as a statutory mitigating factor.\nIn an en banc decision of our court, McKinney v. Ryan,\n813 F.3d 798 (9th Cir. 2015), cert denied, 137 S. Ct. 39\n(2016) (mem), we explained:\n\n For a period of a little over 15 years in capital\n cases, in clear violation of Eddings, the\n\n 3\n Arizona has since revised its death penalty sentencing scheme. All\nreferences to Arizona’s Revised Statute Annotated are to those\nprovisions in effect at the time of Spreitz’s sentencing.\n\n6 SPREITZ V. RYAN\n\n Supreme Court of Arizona articulated and\n applied a “causal nexus” test for nonstatutory\n mitigation that forbade as a matter of law\n giving weight to mitigating evidence, such as\n family background or mental condition,\n unless the background or mental condition\n was causally connected to the crime.\n\nId. at 802. As a result, we held in McKinney that\n“[a]pplication of the causal nexus test to nonstatutory\nmitigating factors violated Eddings, for it resulted in Arizona\ncourts being entirely forbidden, as a matter of state law, to\ntreat as a mitigating factor a family background or a mental\ncondition that was not causally connected to a defendant’s\ncrime.” Id. Spreitz argues that the Arizona Supreme Court\napplied its causal nexus test in his case, refusing to consider\nevidence of his long-term substance and alcohol abuse\nbecause he did not adequately establish a causal connection\nbetween that history of abuse and his crime.\n\n As in McKinney, “the precise question before us is\nwhether the Arizona Supreme Court applied its\nunconstitutional causal nexus test in affirming [Spreitz]’s\ndeath sentence on de novo review.” 4 Id. at 804 (emphasis\n\n\n 4\n The Arizona Supreme Court is required by statute to undertake an\nindependent review of a death sentence. Ariz. Rev. Stat. § 13-703.01(A).\nIn conducting that review for crimes that occurred before 2002, the court\n“independently review[s] the trial court’s findings of aggravation and\nmitigation and the propriety of the death sentence. In doing so, [the\ncourt] review[s] the record de novo, considering the quality and the\nstrength, not simply the number, of aggravating and mitigating factors.”\nState v. Lynch, 357 P.3d 119, 141 (Ariz. 2015) (internal quotation marks\nand citations omitted), rev’d on other grounds, Lynch v. Arizona, 136 S.\nCt. 1818 (2016).\n\n SPREITZ V. RYAN 7\n\nand internal quotation marks omitted). For the reasons that\nfollow, we conclude that it did.\n\n I.\n\n A. Spreitz’s Crimes, Conviction, and Sentence\n\n On May 25, 1989, the police arrested Spreitz after\ndiscovering Ruby Reid’s body in the desert. Upon\nquestioning, Spreitz confessed to murdering Reid. We\nbriefly provide the facts of the murder.\n\n On the evening of May 18, after drinking heavily and\nbeing rejected by the woman he was dating, Spreitz “picked\nup” Reid at a convenience store. State v. Spreitz (Spreitz I),\n945 P.2d 1260, 1264–65 (Ariz. 1997). 5 In his confession,\nSpreitz claimed that Reid voluntarily left with him and that\nhis understanding was that they would have sex later that\nevening. Id. at 1265. Spreitz further claimed that he drove\nher out to the desert, where Reid decided she no longer\nwanted to have sex. Id. The two fought as a result. Id.\nSpreitz explained that Reid slapped him and that he\nresponded by punching her in the mouth. Id. Spreitz then\nsexually assaulted Reid—“remov[ing] her clothing and\nha[ving] vaginal intercourse with her.” Id. Spreitz also\nrecounted that he hit Reid in the head multiple times with a\nrock “to make her stop yelling.” Id. He explained that he\nleft Reid without knowing whether she was alive or dead.\nId.\n\n Shortly after leaving Reid in the desert, Spreitz was\nstopped by a Tucson police department officer. Id. at 1264.\n\n 5\n Our recitation of the facts is adopted from the Arizona Supreme\nCourt’s opinion affirming Spreitz’s conviction and death sentence.\n\n8 SPREITZ V. RYAN\n\nThe officer observed that Spreitz had a ripped shirt, smelled\nof feces, and appeared to be covered in blood and fecal\nmatter. Id. In addition, when detectives later searched\nSpreitz’s car, they found blood spatter in the trunk, some of\nwhich was inconsistent with Spreitz’s blood characteristics.\nId. at 1265.\n\n On Monday morning, May 22, Reid’s naked and\ndecomposing body was discovered on the outskirts of\nTucson. Id. At trial, “the medical examiner testified that,\ndue to the advanced state of decomposition, he could not\ndetermine the full extent and nature of [Reid]’s injuries.” Id.\nEven so, he was able to observe “bruising on the legs, arms,\nand back; bruising and abrasions on the buttocks; several\nbroken ribs; internal bleeding; a broken jaw; several head\nlacerations; and a skull fracture where the skull had been\n‘shoved in.’” Id. The medical examiner concluded that Reid\nhad been killed by “blunt-force trauma to the head.” Id.\n\n In addition to finding Reid’s body “[a]t the scene, police\ndetectives observed tire tracks leading back to the pavement,\noil stains in the dirt, footprints, and drag marks in the dirt\nleading away from the body. They also found feces-stained\npants, tennis shoes, socks, a used tampon, and a torn\nbrassiere. Two blood-stained rocks lay next to the body.”\nId. A few days later, police arrested Spreitz. Id.\n\n On June 2, 1989, a grand jury indicted Spreitz for first-\ndegree murder, Ariz. Rev. Stat. Ann. §§ 13-1105, 13-703;\nsexual assault, Ariz. Rev. Stat. Ann. §§ 13-406(A) & (B);\nand kidnapping, Ariz. Rev. Stat. Ann. §§ 13-304(A)(3) &\n(B). Spreitz I, 945 P.2d at 1265. After five years of pre-trial\nproceedings mostly regarding the admissibility of DNA\nevidence, a seven-day jury trial began on August 9, 1994.\nId. at 1266. After the conclusion of the trial, the jury\nreturned guilty verdicts on all three counts: first-degree\n\n SPREITZ V. RYAN 9\n\nmurder (both premeditated and felony murder), sexual\nassault, and kidnapping. Id.\n\n Prior to both his aggravation-mitigation and sentencing\nhearings before the trial judge, Spreitz submitted evidence\nand a memorandum in support of certain mitigating\ncircumstances. As noted earlier, at the time of Spreitz’s\nsentencing, Arizona’s death penalty statutes provided a list\nof five specific mitigating factors; Arizona case law\nrecognized nonstatutory mitigating factors as well. See Ariz.\nRev. Stat. Ann. § 13-703(G); McKinney, 813 F.3d at 802.\nSpreitz argued as nonstatutory mitigating factors: “(1) his\ndysfunctional family life and lack of socialization; (2) a\nhistory of alcohol and drug abuse; (3) his expressions of\nremorse; [(4)] his good behavior while incarcerated; [(5)] his\nlack of adult convictions; [and (6)] no prior record of violent\ntendencies.” Spreitz I, 945 P.2d at 1279. Spreitz argued as\nstatutory mitigating factors: (1) his age at the time of the\nmurder, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and (2) that\nhis “capacity to appreciate the wrongfulness of his conduct\nor to conform his conduct to the requirements of law was\nsignificantly impaired [due to alcohol use], but not so\nimpaired as to constitute a defense to prosecution,” id.\n§ 13-703(G)(1); see Spreitz I, 945 P.2d at 1279.\n\n Spreitz provided evidence of and argued for all the\nforegoing mitigating circumstances but focused heavily on a\ncombination of his relationship with his mother and his long\nhistory of alcohol and substance abuse. To that end, Spreitz\nsubmitted a written report by and presented testimony from\nan examining psychologist, Dr. Todd Flynn, Ph.D. After\nconducting interviews and research, Dr. Flynn concluded\nthat Spreitz’s longstanding alcohol and substance abuse\nshould be considered as both a statutory and nonstatutory\nmitigating factor. In his report, which was admitted into\n\n10 SPREITZ V. RYAN\n\nevidence at the aggravation and mitigation hearing, Dr.\nFlynn repeatedly emphasized Spreitz’s longstanding\nsubstance abuse 6:\n\n By age twelve or thirteen, Chris Spreitz\n began drinking alcohol and smoking\n marijuana. By age 15, he drank steadily on\n weekends and would have a shot of vodka\n before school.\n\n The collateral information shows that the\n alcohol abuse continued to intensify after he\n left home. A variety of persons . . . described\n him as a heavy drinker. This includes a\n second cousin, Scott [Jouett], who saw him to\n be intoxicated, “a majority of the time,” when\n he was visiting Santa Barbara a week before\n the current offense. To the interviewing\n investigator, Mr. [Jouett] also described,\n “several different occasions when Chris has\n blackouts,” while drinking alcohol.\n\n It appears completely clear from the available\n information that Chris Spreitz had a long-\n standing problem with alcohol which\n probably reached the level of physical\n dependence. He described himself as\n drinking in the morning as early as age 15.\n Virtually everyone else who spent much time\n with him described him as a heavy drinker.\n\n\n\n 6\n Dr. Flynn testified consistently with his report. For ease of\nreference, we refer only to his report.\n\n SPREITZ V. RYAN 11\n\nUltimately, Dr. Flynn concluded that Spreitz’s alcohol\nabuse, childhood home life, and stunted development,\ncombined with rejection by his girlfriend on the day of the\ncrime, and intoxication at the time of the crime, led to the\nmurder.\n\n At the end of his report, Dr. Flynn summarized his\nfindings and opinions with respect to both statutory and\nnonstatutory mitigating factors. He opined that both\nSpreitz’s age, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and\nimpaired capacity to appreciate the wrongfulness of his\nconduct or conform his conduct to the law, Ariz. Rev. Stat.\nAnn. § 13-703(G)(1), were mitigating statutory factors. As\nto § 13-703(G)(5), specifically, Dr. Flynn opined that a\ncombination of a disturbing upbringing in a “pathogenic,\nemotionally neglectful home environment,” and “[y]ears of\nalcoholism intoxication” combined to cause major deficits in\nSpreitz’s social and emotional development. As to\n§ 13-703(G)(1), Dr. Flynn offered a similar conclusion. In\nlight of Spreitz’s “history of alcoholism . . . , a significant\nbut unknown degree of alcohol intoxication is likely” on the\nnight of the crime. In addition, Spreitz’s “history strongly\nsuggests years of early experiences likely to have caused a\nbuild-up of pent-up angry, aggressive feelings toward\nwomen generally (and older women especially) which may\nhave burst forth with uncontrollable intensity with or without\nalcohol intoxication.” Dr. Flynn concluded that Spreitz’s\nintoxication on the night of the crime coupled with his early\nchildhood experiences “likely . . . contributed to an\nuncontrollable outburst of aggression” and inability to\ncontrol his conduct.\n\n Dr. Flynn also concluded that certain nonstatutory\nmitigating factors were present. The nonstatutory mitigating\nfactors included, in his opinion: Spreitz’s low potential for\n\n12 SPREITZ V. RYAN\n\nfuture violence, the “failure of [Spreitz’s] parents to provide\ntreatment for alcohol abuse in [his] teenage years,” and the\n“emotionally deprived, physically punitive home\nenvironment” of Spreitz’s upbringing. Dr. Flynn also\nemphasized that even if both statutory mitigating factors\nfailed to satisfy the statutory threshold, they may still “be\nappropriately considered . . . as nonstatutory mitigation.”\n\n On November 28, 1994, the trial court conducted an\naggravation-mitigation hearing. At the hearing, Spreitz\ncalled three mitigation witnesses: Dr. Flynn and two\ncorrectional officers from the Pima County jail, where\nSpreitz was incarcerated. The State did not offer any\nwitnesses or evidence at the hearing. Dr. Flynn testified\nconsistently with his report, as detailed above.\n\n On the same date, a probation officer filed a Pre-\nSentence Report, which concluded:\n\n It appears the defendant became involved in\n the senseless commission of the instant\n offense due to his alcohol and drug abuse.\n After five years in custody, he now admits his\n substance abuse problem; however, this does\n not condone his involvement in the offense.\n It is unfortunate the victim died before the\n defendant had his revelation.\n\nPrior to the sentencing hearing, Spreitz also submitted\nseveral letters from friends, family, and jail personnel.\n\n On December 21, 1994, the trial court conducted a\nsentencing hearing. Spreitz, Spreitz’s counsel, the victim’s\n\n SPREITZ V. RYAN 13\n\nsister, and the prosecutor each addressed the court. 7 The\nprosecutor first disputed the mitigating evidence presented\nby Spreitz at the mitigation-aggravation hearing. With\nrespect to Spreitz’s alcohol abuse, the prosecutor argued that\nSpreitz’s intoxication at the time of the crime did not meet\nthe statutory definition for mitigation under § 13-703(G)(1)\nbecause his conduct both during the murder and afterwards\ndemonstrated that “he knew what he was doing.” The\nprosecutor emphasized that the evidence revealed that Reid\nhad been forcibly abducted and stuffed into the trunk of a\ncar, thus forcing her to spend time “contemplat[ing] the\nuncertainty of her fate.” The prosecutor argued that the\nevidence revealed signs of struggle and that the serious\ninjuries to Reid’s body belied any notion that Spreitz did not\nknow what he was doing. The prosecutor further argued that\nthe fact that Reid defecated on herself revealed that she was\nterrified prior to her murder. In sum, the prosecutor argued\nthat given both the heinous nature of the crime and the\nmanner in which Reid suffered, a finding in aggravation that\nReid was murdered in an “especially cruel manner” was\nwarranted.\n\n After a short recess, the sentencing judge rendered oral\nfindings addressing the aggravation and mitigation issues.\nHe then imposed a sentence of death. 8 Following the\n\n\n\n 7\n The trial court did not entertain argument at the conclusion of the\naggravation-mitigation hearing. As a result, counsel addressed the\naggravation-mitigation evidence and sentence options at the subsequent\nsentencing hearing.\n 8\n Ring v. Arizona, 536 U.S. 584 (2002), the Supreme Court decision\nholding judge-sentencing in capital cases unconstitutional, had yet to be\ndecided.\n\n14 SPREITZ V. RYAN\n\nhearing, the sentencing judge filed a judgment setting forth\nwritten findings.\n\n The judge first found one aggravating circumstance—\nthat the offense was committed in an “especially cruel\nmanner.” Ariz. Rev. Stat. Ann. § 13-703(F)(6). The judge\nproceeded to review the “many factors [that were] submitted\nin mitigation,” ultimately finding five mitigating\ncircumstances of various levels of significance:\n\n [1] A mitigating circumstance defense\n submitted was that the defendant had an\n extremely disruptive childhood . . . . The\n court finds the home was sub-normal, not\n even a minimally healthy one for developing\n children; and it is obvious the defendant\n suffered a disruptive middle childhood—had\n a punitive, controlling, cold mother, who he\n could not please, no matter what he did.\n\n The defendant in his life turned to substance\n abuse—alcohol and some suggestion he was\n using cocaine and other drugs. However, the\n court does not find such is a mitigating\n circumstance that impaired his ability to\n make a judgment on whether he was acting\n rightfully or wrongfully in the death of the\n victim.\n\n The defendant’s history of intoxication is\n longstanding. He had been abusing\n substances for close to ten years of his life at\n the time of this offense when he was twenty-\n two. Again, the court does not believe that\n the substance abuse or intoxication impaired\n\n SPREITZ V. RYAN 15\n\n the defendant’s ability and capacity to\n appreciate the wrongfulness of his conduct to\n any significant degree. . . . The court does not\n believe intoxication is any sort of mitigating\n circumstance.\n\n ...\n\n [2] The court acknowledges that the\n defendant has begun a process of\n improvement and emotional growth while\n confined at the Pima County Jail, where he\n has taken part in education programs.\n Correctional officers have testified he was a\n prisoner who caused no problems.\n\n [3] The age of the defendant at the time of the\n offense (twenty-two) is not a mitigating\n circumstance in and of itself. Immaturity\n probably is, but the court does not believe\n immaturity was a significant mitigating\n circumstance.\n\n [4] The court finds that Mr. Spreitz has no\n criminal history of a felony nature—there is\n no history or propensity for acts of violence.\n\n [5] The court believes the defendant is\n capable of being rehabilitated. The court\n does not know whether he has a good\n prognosis for the future, but the court\n believes he can be rehabilitated.\n\nIn conclusion, the judge found that “the mitigating\ncircumstances [were] not sufficient to balance the\n\n16 SPREITZ V. RYAN\n\naggravating circumstances, nor [were] they sufficiently\nsubstantial to call for leniency.” He thus imposed a death\nsentence as to the first-degree murder conviction.\n\n B. Spreitz’s Appeal and Post-Conviction Proceedings\n\n Spreitz appealed his conviction and sentence to the\nArizona Supreme Court with the assistance of new counsel.\nThe Arizona Supreme Court affirmed Spreitz’s conviction\nand sentence after conducting its own independent review of\nthe record. 9 In reviewing the mitigating evidence, the court\nfound four of the five mitigating circumstances that the\nsentencing judge had found, declining to find Spreitz’s good\nbehavior while in jail awaiting sentencing as a mitigating\ncircumstance. Spreitz I, 945 P.2d at 1280–81. The court\nexplained:\n\n [1] “We agree with the sentencing judge that\n defendant’s upbringing was subnormal.” Id.\n at 1280.\n\n [2] “We also find that the sentencing judge\n . . . properly found that his emotional\n immaturity was not a significant mitigating\n factor.” Id. at 1281.\n\n [3] “We agree that the record supports the\n sentencing judge’s findings that defendant\n had no previous adult felony convictions, no\n prior acts of violence, and\n\n\n 9\n See note 4, supra. Because the Arizona Supreme Court conducted\na de novo review, we focus on its analysis, rather than, as the dissent\ndoes in part, the analysis by the trial court.\n\n SPREITZ V. RYAN 17\n\n [4] that the defendant is capable of\n rehabilitation.” Id.\n\nThe Arizona Supreme Court also added that it considered\nremorse as an additional nonstatutory mitigating factor. Id.\n\n The court addressed Spreitz’s history of alcohol and\nsubstance abuse as follows:\n\n The record demonstrates defendant’s\n longtime substance abuse problems. We\n note, however, that defendant’s general\n problems with substance abuse are not\n essential to our decision here. We therefore\n decline to conclude that defendant was\n impaired by alcohol consumption to an extent\n that it interfered with his “capacity to\n appreciate the wrongfulness of his conduct or\n to conform his conduct to the requirements of\n the law.” A[riz]. R[ev]. S[tat]. [Ann.]\n § 13-703(G)(1); see also State v. Medrano,\n 185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)\n (citing [State v.] Stokley, 182 Ariz. [505,]\n 520, 898 P.2d [454,] 469 [1995]).\n\nId. at 1280–81. In the only other portion of its opinion\naddressing Spreitz’s history of alcohol and substance abuse,\nthe court said:\n\n The sentencing judge found that defendant’s\n ability to appreciate the wrongfulness of his\n conduct was not impaired on the night of the\n murder to any significant extent by substance\n abuse, emotional disorders, situational stress,\n or by a combination of these. Our review of\n\n18 SPREITZ V. RYAN\n\n the record convinces us that the trial court’s\n finding was proper.\n\nId. at 1281.\n\n The Arizona Supreme Court agreed with the trial court’s\naggravation finding that the murder was done in an\nespecially cruel manner, id. at 1279, and then proceeded to\nreweigh the applicable aggravating and mitigating factors,\nfinding that “the aggravating circumstance of especial\ncruelty in defendant’s murder of Ruby Reid outweigh[ed] all\nfactors mitigating in favor of leniency.” Id. at 1282. It\naffirmed his death sentence. Id. at 1283. Spreitz filed a\npetition for a writ of certiorari, which the United States\nSupreme Court denied. Spreitz v. Arizona, 523 U.S. 1027\n(1998) (mem).\n\n In March 2000, represented again by new counsel,\nSpreitz filed a petition for post-conviction relief in the\nArizona Superior Court (“PCR court”). See Ariz. R. Crim.\nP. 32.1. Spreitz alleged, inter alia, that the sentencing judge\nand the Arizona Supreme Court committed error by failing\nto consider his history of alcohol and substance abuse as a\nnonstatutory mitigating factor apart from its causal\nconnection to the murder—i.e., Eddings error. He also\nalleged that his appellate counsel was ineffective for failing\nto raise the sentencing judge’s Eddings error on appeal to the\nArizona Supreme Court. Spreitz additionally raised several\nnew claims of trial-counsel-ineffectiveness.\n\n The PCR court denied all of Spreitz’s claims and entered\nan order dismissing his petition for post-conviction relief. In\nPart 3 of its order, the PCR court discussed Spreitz’s claims\nof “nexus-error”—that both the trial court and the Arizona\nSupreme Court erred when they failed to consider Spreitz’s\nlongstanding alcohol and substance abuse on the basis that\n\n SPREITZ V. RYAN 19\n\nSpreitz had failed to establish a causal nexus between the\nlong-term substance abuse and the murder. 10 The PCR court\ndismissed the claim 11 as waived, reasoning that Spreitz had\nfailed to raise the issue on direct appeal. Nonetheless, the\nPCR court addressed the merits of the nexus-error claim in\nthe course of analyzing Spreitz’s argument that his appellate\ncounsel was ineffective for failing to raise it. In doing so,\nthe PCR court explained that the claim failed because:\n\n [I]t must be demonstrated, under A[riz].\n R[ev]. S[tat]. [Ann.] § 13-703(G)(l), that\n there is a causal link between the history of\n alcohol or substance abuse and the offense\n itself. E.g., State v. Stokley, [] 182 Ariz.\n [505,] 523 [898 P.2d 454, 472 (Ariz. 1995)].\n Without some basis for explaining or\n defining the individual’s behavior at the time\n of the offense, the Petitioner’s history of\n alcohol or substance abuse would be\n inconsequential (which is exactly what the\n trial court and Supreme Court concluded).\n State v. Kayer, 194 Ariz. 423, 984 P.2d 81\n ([Ariz.] 1999).\n\n At times, the court can and should consider\n an individual’s long-term alcoholism and\n\n 10\n For ease of reference, we refer to the Arizona courts’ alleged error\nas “nexus-error.”\n\n 11\n Although Spreitz raised, and the PCR court recognized, two\ndistinct claims—one with respect to the sentencing court and one with\nrespect to the Arizona Supreme Court—the PCR court’s analysis treats\nthe claims as one and the same and refers to them in the singular. To\nreflect accurately the PCR court’s discussion, our summary here likewise\nrefers to a singular “claim.”\n\n20 SPREITZ V. RYAN\n\n substance abuse, usually in conjunction with\n other factors or diagnosis, as non-statutory\n mitigation. However, the impact or effect of\n the alcoholism or substance abuse must be\n substantial and of such severity that it\n provides a sufficient basis for explaining the\n defendant’s conduct, character or ability to\n control his behavior at the time of the offense.\n [Citations omitted].\n\n As previously discussed, there is no evidence\n in Petitioner’s case to suggest that he suffered\n any long-term effects from his alcohol or\n drug abuse that precluded him from\n controlling his behavior. Petitioner did not\n suffer from any cognitive or emotional\n deficits that rendered him incapable of\n controlling his conduct. Therefore, the trial\n court did not err in failing to find Petitioner’s\n history of alcohol or substance abuse as a\n separate, non-statutory mitigating factor.\n [Citation omitted].\n\n Spreitz filed a petition for review in the Arizona\nSupreme Court challenging the PCR court’s judgment. The\nArizona Supreme Court summarily affirmed the PCR court’s\nmerits determination with respect to Spreitz’s Eddings\nclaims. State v. Spreitz (Spreitz II), 39 P.3d 525, 527 (Ariz.\n2002).\n\n Spreitz filed his federal habeas petition in February\n2003. He alleged in claim seven that the sentencing judge\nand the Arizona Supreme Court had both committed nexus-\nerror with respect to his long-time alcohol and substance\nabuse, and that appellate counsel was ineffective for failing\n\n SPREITZ V. RYAN 21\n\nto argue the sentencing judge’s nexus-error on direct appeal\nto the Arizona Supreme Court.\n\n The district court denied all of Spreitz’s claims. With\nrespect to the alleged nexus-error by both the sentencing\ncourt and the Arizona Supreme Court, the district court\nfound the claims procedurally barred, relying on the PCR\ncourt’s determination that Spreitz had waived the claims\nbecause he could have raised them on direct appeal but failed\nto do so. The district court concluded that Spreitz had\nproperly exhausted his ineffective assistance of appellate\ncounsel claim, and thus addressed the alleged nexus-error in\nthat context, ultimately concluding it was not meritorious.\nAccordingly, the district court concluded that “appellate\ncounsel’s failure to raise this issue on appeal does not\nconstitute ineffectiveness.”\n\n Spreitz timely appealed.\n\n II. Jurisdiction and Standard of Review\n\n We have jurisdiction pursuant to 28 U.S.C. §§ 1291\nand 2253(a), and review de novo the district court’s denial\nof a writ of habeas corpus. Poyson v. Ryan, 879 F.3d 875,\n887 (9th Cir. 2018). Because Spreitz filed his federal habeas\npetition after April 24, 1996, he must satisfy the\nAntiterrorism and Effective Death Penalty Act (“AEDPA”).\nUnder AEDPA, we may not grant habeas relief unless the\nstate’s adjudication of Spreitz’s claim (1) “was contrary to\n. . . clearly established federal law[] as determined by the\nSupreme Court,” (2) “involved an unreasonable application\nof” such law, or (3) “was based on an unreasonable\ndetermination of the facts in light of the evidence presented\nin the State court proceeding.” (28 U.S.C. § 2254(d)). “In\nmaking this determination, we look to the last state court\ndecision to address the claim,” White v. Ryan, 895 F.3d 641,\n\n22 SPREITZ V. RYAN\n\n665 (9th Cir. 2018) (citing Wilson v. Sellers, 138 S. Ct. 1188,\n1192 (2018)), which for Spreitz’s nexus-error claim is that\nof the PCR court.\n\n Spreitz argues that the PCR court’s decision was\n“contrary to” Eddings v. Oklahoma, 455 U.S. 104 (1982).\n“A state court’s decision is contrary to clearly established\nfederal law if it ‘applies a rule that contradicts the governing\nlaw set forth in [U.S. Supreme Court] cases’ or arrives at a\ndifferent result in a case that ‘confronts a set of facts that are\nmaterially indistinguishable from a decision of [the\nSupreme] Court.’” Castellanos v. Small, 766 F.3d 1137,\n1146 (9th Cir. 2014) (alteration in original) (quoting\nWilliams v. Taylor, 529 U.S. 362, 405–06 (2000)). “If the\nstate court applies a legal standard that contradicts clearly\nestablished [Supreme Court] law, we review de novo the\napplicant’s claims, applying the correct legal standard to\ndetermine whether the applicant is entitled to relief.” Id.\n(citation omitted).\n\n III. Discussion\n\n As discussed supra, the precise question we must decide\nis whether the Arizona Supreme Court applied its\nunconstitutional causal nexus test in violation of Eddings\nwhen it affirmed Spreitz’s death sentence. To answer that\nquestion, we must first determine whether the claim is\nproperly before us. After concluding that Spreitz’s claim\ncould not have been procedurally defaulted, we turn to the\nlevel of deference we must accord the PCR court’s ruling on\nSpreitz’s Eddings claim under AEDPA. Because the PCR\ncourt’s decision was contrary to clearly established Supreme\nCourt precedent, we accord that decision no deference and\nreview Spreitz’s Eddings claim de novo. We conclude that\nthe Arizona Supreme Court violated Eddings by\nimpermissibly requiring that Spreitz establish a causal\n\n SPREITZ V. RYAN 23\n\nconnection between his longstanding substance abuse and\nthe murder before considering and weighing the evidence as\na nonstatutory mitigating factor. Finally, having determined\nthat the Arizona Supreme Court applied an unconstitutional\ncausal nexus test in its sentencing procedure, we turn to\nwhether the error was harmless under Brecht v. Abrahamson,\n507 U.S. 619 (1993). We conclude that the error was not\nharmless and therefore reverse the district court’s judgment\nwith respect to Spreitz’s sentence.\n\n A.\n\n Under the doctrine of procedural default, a federal\nhabeas court “will not review the merits of claims, including\nconstitutional claims, that a state court declined to hear\nbecause the prisoner failed to abide by a state procedural\nrule.” Martinez v. Ryan, 566 U.S. 1, 9 (2012). “We review\nde novo a district court’s conclusion that a claim is\nprocedurally defaulted.” Cooper v. Neven, 641 F.3d 322,\n326 (9th Cir. 2011).\n\n Here, the district court concluded that both of Spreitz’s\nEddings claims—that the sentencing court and the Arizona\nSupreme Court applied an unconstitutional causal nexus\ntest—were procedurally defaulted, because the PCR court\nfound each claim waived for failure to raise them on direct\nappeal. This conclusion, however, is erroneous with respect\nto Spreitz’s claim that the Arizona Supreme Court violated\nEddings. After all, “[t]he Arizona Supreme Court reviews\ncapital sentences de novo, making its own determination of\nwhat constitute[s] legally relevant aggravating and\nmitigating factors, and then weighing those factors\nindependently. Ariz. Rev. Stat. Ann. § 13-755. The Arizona\nSupreme Court ‘conducts a thorough and independent\nreview of the record and of the aggravating and mitigating\nevidence to determine whether the sentence is justified.’”\n\n24 SPREITZ V. RYAN\n\nMcKinney, 813 F.3d at 819 (emphasis removed) (quoting\nState v. McKinney, 917 P.2d 1214, 1225 (Ariz. 1996)).\nSpreitz could not have raised on direct appeal his claim that\nthe Arizona Supreme Court violated Eddings in performing\nits de novo review of Spreitz’s death sentence. The first\nopportunity he had to raise that claim was before the PCR\ncourt, at which time he did so. Thus, Spreitz’s claim is not\nprocedurally defaulted and is properly before us. 12\n\n B.\n\n Because Spreitz’s claim that the Arizona Supreme Court\napplied an unconstitutional causal nexus test to his\nnonstatutory mitigating evidence of long-time alcohol and\n 12\n The district court declined to issue a certificate of appealability\n(“COA”) on Spreitz’s substantive Eddings claim, which he raised as\nclaim 1.7 on pages 108 through 111 in his federal habeas petition.\nAlthough Spreitz did not present his Eddings claim as a separate\nuncertified issue, we may grant a COA regarding any uncertified issue\ndiscussed in a petitioner’s opening brief. See 9th Cir. R. 22-1(e).\nSpreitz’s opening brief discusses the Arizona courts’ Eddings errors. We\ntherefore may exercise our discretion to expand the COA to encompass\nthe Eddings claim.\n\n To obtain a COA under 28 U.S.C. § 2253(c), Spreitz must\ndemonstrate that “reasonable jurists could debate whether . . . the\npetition should have been resolved in a different manner or that the issues\npresented were adequate to deserve encouragement to proceed further.”\nSlack v. McDaniel, 529 U.S. 473, 484 (2000). Reasonable jurists could\ndebate the merits of Spreitz’s Eddings claim. We thus exercise our\ndiscretion and expand the COA to encompass Spreitz’s substantive\nEddings claim with respect to the Arizona Supreme Court. We\nacknowledge that under Ninth Circuit Rule 22-1(f), the State would\nnormally be afforded an opportunity to submit supplemental briefing on\nthe uncertified Eddings issue. Here, however, the Eddings issue has been\nexhaustively briefed, in the context of Spreitz’s appellate counsel’s\nalleged ineffectiveness before the Arizona Supreme Court. Additional\nbriefing is unwarranted.\n\n SPREITZ V. RYAN 25\n\nsubstance abuse is properly before us, we next determine the\nlevel of deference to accord the PCR court’s adjudication of\nthat claim. As discussed above, although the PCR court first\ndeclared the claim waived, it proceeded to adjudicate the\nclaim on the merits. In denying the claim, the PCR court\nexplained:\n\n Without some basis for explaining or\n defining the individual’s behavior at the time\n of the offense, [Spreitz]’s history of alcohol\n or substance abuse would be inconsequential\n (which is exactly what the trial court and\n Supreme Court concluded).\n\n At times, the court can and should consider\n an individual’s long-term alcoholism and\n substance abuse, usually in conjunction with\n other factors or diagnosis, as non-statutory\n mitigation. However, the impact or effect of\n the alcoholism or substance abuse must be\n substantial and of such severity that it\n provides a sufficient basis for explaining the\n defendant’s conduct, character, or ability to\n control his behavior at the time of the offense.\n\n ....\n\n As previously discussed, there is no evidence\n in [Spreitz]’s case to suggest that he suffered\n any long-term effects from his alcohol or\n drug abuse that precluded him, from\n controlling his behavior. Petitioner did not\n suffer from any cognitive or emotional\n deficits that rendered him incapable of\n controlling his conduct. Therefore, the trial\n\n26 SPREITZ V. RYAN\n\n court did not err in failing to find [Spreitz]’s\n history of alcohol or substance abuse as a\n separate, non-statutory mitigating factor.\n\n In other words, the PCR court concluded that Spreitz’s\nlongstanding alcohol and substance abuse could only be\nconsidered a nonstatutory mitigating factor if it “provide[d]\na sufficient basis for explaining the defendant’s conduct,\ncharacter, or ability to control his behavior at the time of the\noffense.” This discussion both accurately described the\ncausal nexus test and approved of it.\n\n Such reasoning is contrary to clearly established federal\nlaw. As explained in McKinney, “the causal nexus test\nclearly violates Eddings.” 813 F.3d at 810. Therefore,\n“[b]ecause the state court used the wrong standard, we need\nnot defer to that decision.” Hardy v. Chappell, 849 F.3d 803,\n820 (9th Cir. 2016).\n\n C.\n\n Because the PCR court’s decision was “contrary to”\nclearly established law, we review de novo the merits of\nSpreitz’s Eddings claim. 28 U.S.C. § 2254(d)(1). We\ntherefore consider whether the Arizona Supreme Court\napplied its unconstitutional causal nexus test in reviewing\nand affirming Spreitz’s death sentence. In so doing, “we\nlook only to the decision of th[e Arizona Supreme Court],\n. . . only [considering] . . . the decision of the sentencing\njudge . . . to the degree it was adopted or substantially\nincorporated by the Arizona Supreme Court.” McKinney,\n813 F.3d at 819.\n\n SPREITZ V. RYAN 27\n\n 1.\n\n Our decision in McKinney frames our consideration of\nthe Arizona Supreme Court’s decision and Spreitz’s claim.\nMcKinney makes clear that the Arizona Supreme Court\nconsistently articulated and applied its unconstitutional\ncausal nexus test during the period in which Spreitz’s death\nsentence was litigated in the trial court and reviewed by the\nArizona Supreme Court. Id. at 824. Moreover, McKinney\nmakes plain that although on habeas review, we generally\napply a “presumption that state courts know and follow the\nlaw,” “the Arizona Supreme Court’s consistent articulation\nand application of its causal nexus test . . . make such a\ncourse impossible.” Id. at 803 (internal quotation marks and\ncitation omitted). In appeals heard from the late 1980s until\nat least 2002, see State v. Canez, 42 P.3d 564, 594 (Ariz.\n2002), it was unmistakably clear that “the Arizona Supreme\nCourt did not know and follow federal law.” Id. (internal\nquotation marks omitted). 13 Therefore, just as in McKinney\nand subsequent cases applying McKinney, see, e.g., Poyson,\n879 F.3d at 889, “the presumption” that the Arizona\nSupreme Court knew and followed the law “is rebutted” here\nas well. McKinney, 813 F.3d at 804.\n\n That said, McKinney does not dispose of Spreitz’s claim.\nWe must still review the Arizona Supreme Court’s decision\nto determine whether it did, in fact, apply its unconstitutional\ncausal nexus test in Spreitz’s case. In doing so, we are\nmindful of the fact that at the time the Arizona Supreme\nCourt decided Spreitz’s appeal, “if there is to be a\npresumption, it is that the Arizona Supreme Court violated\n\n\n 13\n In 2005, the Arizona Supreme Court repudiated its earlier nexus\ntest. See State v. Anderson, 111 P.3d 369, 391–92 (Ariz. 2005).\n\n28 SPREITZ V. RYAN\n\nthe dictates of Lockett and Eddings.” Greenway v. Ryan,\n866 F.3d 1094, 1095 (9th Cir. 2017).\n\n 2.\n\n Against this backdrop, we are convinced that the Arizona\nSupreme Court applied its causal nexus test with respect to\nSpreitz’s evidence of longstanding alcohol and substance\nabuse in violation of Eddings.\n\n We quote again the Arizona Supreme Court’s discussion\nof Spreitz’s history of alcohol and substance abuse:\n\n The record demonstrates defendant’s\n longtime substance abuse problems. We\n note, however, that defendant’s general\n problems with substance abuse are not\n essential to our decision here. We therefore\n decline to conclude that defendant was\n impaired by alcohol consumption to an extent\n that it interfered with his “capacity to\n appreciate the wrongfulness of his conduct or\n to conform his conduct to the requirements of\n the law.” A[riz]. R[ev]. S[tat]. [Ann.] § 13-\n 703(G)(1); see also State v. Medrano,\n 185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)\n (citing Stokley, 182 Ariz. at 520, 898 P.2d at\n 469).\n\n ....\n\n The sentencing judge found that defendant’s\n ability to appreciate the wrongfulness of his\n conduct was not impaired on the night of the\n murder to any significant extent by substance\n abuse, emotional disorders, situational stress,\n\n SPREITZ V. RYAN 29\n\n or by a combination of these. Our review of\n the record convinces us that the trial court’s\n finding was proper.\n\nSpreitz I, 945 P.2d at 1280–81. This discussion\ndemonstrates that the Arizona Supreme Court primarily\nconcluded that Spreitz failed to show statutory mitigation\nunder Ariz. Rev. Stat. Ann. § 13-703(G)(1) based on his\nintoxication on the night of the murder. “When applied\nsolely in the context of statutory mitigation under\n§ 13-703(G)(1), the causal nexus test does not violate\nEddings.” McKinney, 813 F.3d at 810.\n\n The plain text of the Arizona Supreme Court’s decision,\nhowever, also clearly demonstrates that the court did not\nsolely apply its causal nexus test to Spreitz’s evidence of\nstatutory mitigation. The court initially recognized Spreitz’s\n“general problems with substance abuse”—evidence that\nshould have been relevant as nonstatutory mitigation—but\nconcluded those problems were “not essential to [its]\ndecision” because they did not show Spreitz was “impaired\non the night of the murder.” Spreitz I, 945 P.2d at 1280–81.\nIn other words, the court did not acknowledge the relevance\nof longtime substance abuse as nonstatutory mitigation in the\nabsence of a causal relationship to the crime. Instead, the\ncourt held that because there was no causal relationship,\nSpreitz’s long-term alcohol abuse was not of significance—\ni.e. “not essential”—to the court’s overall determination of\neither statutory or nonstatutory mitigation.\n\n The Supreme Court, however, has been clear that “‘full\nconsideration of evidence that mitigates against the death\npenalty is essential if the sentencer is to give a reasoned\nmoral response to the defendant’s background, character,\nand crime.’” McKinney, 813 F.3d at 812 (first emphasis\n\n30 SPREITZ V. RYAN\n\nadded) (quoting Penry I, 492 U.S. at 319, 328).\nAccordingly, “Eddings requires that ‘[t]he sentencer must\nalso be able to consider and give effect’” to any relevant\nmitigation evidence proffered by the defendant, which\nincludes evidence going toward nonstatutory mitigation. Id.\n(quoting Penry I, 492 U.S. at 319, 328). The Arizona\nSupreme Court’s refusal to consider Spreitz’s alcohol and\nsubstance abuse beyond its connection, if any, to Reid’s\nmurder constitutes application of its unconstitutional causal\nnexus test to relevant nonstatutory mitigation evidence.\n\n Our understanding of the Arizona Supreme Court’s\nopinion is bolstered by the fact that in its discussion of\nSpreitz’s longstanding alcohol and substance abuse, the\ncourt cited to a prior decision that likewise applied the\nunconstitutional causal nexus test, State v. Medrano,\n914 P.2d 255 (Ariz. 1996). In McKinney, we explained that\nin Medrano, the Arizona Supreme Court upheld the\nsentencing judge’s determination that the defendant’s\ncocaine use failed as a nonstatutory mitigating circumstance\nbecause he did not show that it contributed to his conduct on\nthe night of the murder. McKinney, 813 F.3d at 825–26. In\nso doing, “[t]he Arizona Supreme Court applied the causal\nnexus test.” Id. at 825. Although the Arizona Supreme\nCourt’s citation to Medrano is not dispositive, it certainly\ncorroborates our understanding of the court’s opinion. See\nalso Poyson, 879 F.3d at 890 (emphasizing, among other\nconsiderations, that the Arizona Supreme Court cited a\npassage from a case specifically identified in McKinney as\napplying the unconstitutional causal nexus test in order to\nconclude that the state court also applied that test to Poyson’s\nevidence). Moreover, as we explained in McKinney, “the\nArizona Supreme Court applied its unconstitutional nexus\ntest consistently” for fifteen years because it had “a mistaken\nunderstanding of Eddings.” 813 F.3d at 826. It is thus\n\n SPREITZ V. RYAN 31\n\nentirely logical that in citing the exact portion of Medrano\nwhere it previously applied its causal nexus test, the Arizona\nSupreme Court was applying the same test in Spreitz’s case.\nSee id. at 821; see also Poyson, 879 F.3d at 890.\n\n We find further support for our understanding of the\nArizona Supreme Court’s unconstitutional treatment of\nSpreitz’s alcohol and substance abuse when we compare it\nwith the manner in which the court discussed other\nmitigating factors. As detailed earlier, in evaluating the\nevidence of Spreitz’s general history of alcohol and\nsubstance abuse, the court stated that this evidence was “not\nessential” to its decision. Spreitz I, 945 P.2d at 1280. By\ncontrast, when the court considered other mitigating factors,\nit specifically discussed the weight it would give those\nfactors, as opposed to simply dismissing them outright. For\nexample, when discussing Spreitz’s subnormal upbringing,\nthe court stated, “Although we recognize defendant’s\nupbringing as a mitigating circumstance, we accord it little\nweight.” Id. at 1280 (emphasis added). As another example,\nafter “finding that since his arrest, [Spreitz] has\ndemonstrated remorse,” the court stated that his “remorse for\nhis actions does little to counterbalance especial cruelty as a\nserious aggravating circumstance.” Id. at 1281 (emphasis\nadded). And finally, when discussing the sentencing judge’s\ndetermination regarding Spreitz’s emotional immaturity, the\ncourt stated, “We also find that the sentencing judge . . .\nproperly found that his emotional immaturity was not a\nsignificant mitigating factor.” Id. (emphasis added). In\nother words, when the Arizona Supreme Court wanted to\nassign weight to a given factor, it said so. This pattern stands\nin stark contrast to the court’s conclusion that anything\nrelated to Spreitz’s long-term alcohol or substance abuse that\nfell short of the statutory definition could not serve as\nmitigation. Id. at 1280–81.\n\n32 SPREITZ V. RYAN\n\n Finally, as also discussed earlier, the PCR court\ndistinctly interpreted the Arizona Supreme Court’s decision\nas having applied a causal nexus test. The PCR court\nexplained that “[w]ithout some basis for explaining or\ndefining the individual’s behavior at the time of the offense,\n[Spreitz]’s history of alcohol or substance abuse would be\ninconsequential (which is exactly what the trial court and\nSupreme Court concluded).” Moreover, in articulating how\nthe Arizona Supreme Court had applied its causal nexus test\nin Spreitz’s direct appeal, the PCR court, as had the Arizona\nSupreme Court, cited to a decision of the Arizona Supreme\nCourt indisputably applying an unconstitutional nexus test,\nState v. Kayer, 984 P.2d 31 (Ariz. 1999). See McKinney,\n813 F.3d at 816 (explaining that Kayer held “that the\ndefendant’s mental impairment ‘was not established as a\nnonstatutory mitigating factor’ in part because ‘defendant\noffered no evidence to show the requisite causal nexus that\nmental impairment affected his judgment or his actions at the\ntime of the murder’” (quoting Kayer, 984 P.2d at 46)). The\nfact that the PCR court interpreted the Arizona Supreme\nCourt’s opinion as we do—and then itself applied the\nconstitutionally erroneous nexus test—lends further support\nto our reading.\n\n We recognize that the Arizona Supreme Court’s decision\non the particular issue before us was framed by language that\nmight, in a decisional vacuum, suggest the court knew,\nunderstood, and applied the law with respect to Eddings.\nThe court acknowledged that Spreitz argued that his history\nof alcohol and drug abuse served as both a statutory and\nnonstatutory mitigating factor. Spreitz I, 945 P.2d at 1279.\nIt then stated that it “must consider any aspect of the\ndefendant’s character or record . . . relevant to determining\nwhether the death penalty should be imposed.” Id. And\nfinally, in reaching the conclusion to affirm Spreitz’s death\n\n SPREITZ V. RYAN 33\n\nsentence, the court explained that it had “examin[ed] the\nentire record and reweigh[ed] the applicable aggravating and\nmitigating factors.” Id. at 1281–82.\n\n The Arizona Supreme Court’s acknowledgment of what\nSpreitz argued does not indicate that it relied on or accepted\nthat argument as correct. Further, we cannot, given\nMcKinney, Poyson, and Greenway, proceed in a decisional\nvacuum.\n\n Considered in light of the baseline that the Arizona\nSupreme Court was consistently applying its\nunconstitutional causal nexus test in death penalty cases at\nthe time Spreitz’s appeal was decided, as well as its citations\nto cases held in McKinney to be applying that standard, we\nare confident that the manner in which the Arizona Supreme\nCourt disposed of Spreitz’s evidence of his longstanding\nalcohol and substance abuse confirms that it applied the\ncausal nexus test it was using during that period.\n\n Styers v. Schriro, 547 F.3d 1026 (9th Cir. 2008) supports\nthis conclusion. Styers stated that the mere fact that “the\nArizona Supreme Court stated that it had ‘considered all of\nthe proffered mitigation,’” is not sufficient to contradict the\nfact that “its analysis prior to this statement indicated\notherwise.” Id. at 1035 (internal citation omitted). In Styers,\nnotwithstanding assertions to the contrary, the Arizona\nSupreme Court refused to consider evidence of the\ndefendant’s Post Traumatic Stress Disorder (“PTSD”)\nbecause he had failed to connect it causally to his crimes. Id.\nHere, too, the Arizona Supreme Court’s actual analysis of\nSpreitz’s longstanding alcohol and substance abuse reveals\nthat it refused to consider this particular aspect of Spreitz’s\ncharacter free from any connection to the crime. Rather, the\ncourt’s discussion demonstrates that in evaluating Spreitz’s\nlong-term alcohol and substance abuse, it applied a causal\n\n34 SPREITZ V. RYAN\n\nnexus test in violation of Eddings and therefore refused, as a\nmatter of law, to consider it in assessing Spreitz’s evidence\nof nonstatutory mitigating circumstances.\n\n To the extent the state argues the Arizona Supreme\nCourt’s opinion is ambiguous as to whether it applied an\nunconstitutional causal nexus test or not—and we believe\nthat it is not—McKinney requires resolving that ambiguity\nin favor of Spreitz. McKinney held it clear that the Arizona\nSupreme Court “consistently articulated and applied its\ncausal nexus test” at the time the trial court imposed\nSpreitz’s death sentence and the Arizona Supreme Court\nreviewed and affirmed that sentence. McKinney, 813 F.3d\nat 803 (emphasis in original). Again, “if there is to be a\npresumption, it is that the Arizona Supreme Court violated\nthe dictates of Lockett and Eddings.” Greenway, 866 at\n1095.\n\n In sum, because (1) the Arizona Supreme Court clearly\nstated that Spreitz’s longstanding alcohol and substance\nabuse was “not essential” to its decision, which was in line\nwith the court’s consistent practice of applying a causal\nnexus test; (2) the Arizona Supreme Court cited to Medrano,\nin which it had previously articulated a causal nexus test;\n(3) the Arizona Supreme Court carefully laid out the weight\nit would accord other mitigating factors as opposed to simply\nignoring them; and (4) the PCR court likewise interpreted\nthe Arizona Supreme Court’s opinion as applying a causal\nnexus test, we conclude that the Arizona Supreme Court\nheld, as a matter of law, that Spreitz’s long-term alcohol and\nsubstance abuse was not a nonstatutory mitigating factor and\ntherefore refused to consider it. This holding was in clear\nviolation of Eddings.\n\n SPREITZ V. RYAN 35\n\n 3.\n\n In light of our conclusion that the Arizona Supreme\nCourt failed to comply with Eddings, we must next decide\nwhether Spreitz was prejudiced by this error. On federal\nhabeas review, relief is warranted if an error “had substantial\nand injurious effect or influence” on the challenged decision.\nSee Brecht v. Abrahamson, 507 U.S. 619, 623 (1993). The\npetitioner is “not entitled to habeas relief” unless he can\nestablish that the error “resulted in actual prejudice.” Davis\nv. Ayala, 135 S. Ct. 2187, 2197 (2015) (internal quotation\nmarks omitted). Here, because we are “in grave doubt about\nwhether [the Arizona Supreme Court’s] error of federal law\nhad substantial and injurious effect or influence in\ndetermining [whether to affirm Spreitz’s death sentence],\nthat error is not harmless. And, [Spreitz] must win.”\nMcKinney, 813 F.3d at 822 (quoting O’Neal v. McAninch,\n513 U.S. 432, 435 (1995)).\n\n We previously have required resentencing based on\nprejudicial Eddings error where the Arizona Supreme Court\nrefused to consider, as a matter of law, mitigation evidence\n“central to” the petitioner’s “plea for leniency.” See id. at\n823; see also Coleman v. Calderon, 210 F.3d 1047, 1051\n(9th Cir. 2000) (concluding that a penalty-phase jury\ninstruction error was not harmless under Brecht because “it\nundermined the very core of Coleman’s plea for life”). 14 For\nthe reasons that follow, it is clear that Spreitz’s long-term\n\n 14\n We do not suggest that Brecht prejudice may only be established\nif inappropriately considered mitigating evidence was at “the very core\nof [a defendant’s] plea for life.” Coleman, 210 F.3d at 1051. It is\nconceivable that other forms of Eddings error could have “a substantial\nand injurious effect” on a defendant’s sentence if they would be\nsignificant to the sentencer, despite the defendant’s failure to rely on\nthem as central to his or her argument. Brecht, 507 U.S. at 623.\n\n36 SPREITZ V. RYAN\n\nalcohol and substance abuse was important mitigation\nevidence, and was so closely related as to become central to\nhis plea for leniency. The Arizona Supreme Court’s refusal\nto consider that evidence had a substantial and injurious\neffect on its sentencing determination that requires\nresentencing.\n\n While a defendant’s evidence of longstanding alcohol\nand/or substance abuse can be probative mitigation evidence\non its own, a “history of substance abuse [is] substantially\nmitigating when . . . combined with other mitigating\nevidence.” Henry v. Ryan, 720 F.3d 1073, 1090 (9th Cir.\n2013) (emphasis added) (listing cases and holding that the\ndefendant’s alcoholism “would not have had a substantial\nand injurious effect” because it “st[ood] alone, was similar\nto evidence already considered by the sentencing courts, and\nwas of limited probative value”) (footnote omitted). 15 Here,\nSpreitz’s evidence regarding his history of alcohol and\nsubstance abuse—spanning nearly half his life by the time\nhe committed the crime at the age twenty-two—was the\n\n 15\n The dissent makes much of Henry, suggesting it dictates the\noutcome of our prejudice analysis. Dissent at 29. Not so. In that case,\nthe state court had found that Henry’s intoxication on the day of the\nmurder did constitute a statutory mitigating factor but was not\nsufficiently substantial to outweigh the two aggravating factors. Henry,\n720 F.3d at 1090. Henry argued that the state court erred by failing to\nconsider evidence of his “historical alcoholism” in addition to his\nintoxication on the day of the crime. Id. We concluded that any error\nwas harmless because the history of substance abuse “was similar to\nevidence already considered by the sentencing courts and was of limited\nprobative value.” Id. In other words, because the state court had found\ninsufficient evidence of alcohol use that was tied to the murder, we could\nnot conclude that general alcohol use that was not tied to the murder\ncould have been determinative. Here, by contrast, the Arizona Supreme\nCourt found that Spreitz’s evidence of intoxication did not rise to the\nlevel of statutory mitigation. Accordingly, there is not the same\ncontextual harmlessness problem as in Henry.\n\n SPREITZ V. RYAN 37\n\nfactor on which much of his other mitigation evidence\nturned.\n\n In both his report and his testimony, Dr. Flynn discussed\nat length the extent to which alcohol consumed Spreitz’s life\nin the decade leading up to Reid’s murder. Spreitz began\ndrinking when he was twelve or thirteen; by fifteen, he was\ndrinking steadily on the weekends and drinking vodka before\nschool. According to Dr. Flynn, “[d]rug and alcohol abuse\ndominated his teenage years.” After he left home, Spreitz’s\ndrinking intensified: Spreitz experienced blackouts multiple\ntimes and, a week before the murder, Spreitz was\n“intoxicated a majority of the time.” Dr. Flynn concluded\nthat “[i]t appears completely clear from the available\ninformation that . . . Spreitz had a long-standing problem\nwith alcohol which probably reached the level of physical\ndependence.”\n\n Critically, Spreitz’s decade-long struggle with alcohol\nand other addictive substances reflects the true gravity of his\n“subnormal upbringing,” a mitigating circumstance\nrecognized by the Arizona Supreme Court. See Spreitz I,\n945 P.2d at 1280. Both Spreitz’s sentencing memorandum\nand Dr. Flynn detailed how his difficult childhood led to\nSpreitz’s early dependence on alcohol. For instance, Dr.\nFlynn explained that Spreitz turned to alcohol as a coping\nmechanism to deal with the “pervasive subabusive\nemotional battering and neglect that he received from his\nmother,” whom he described as “punitive, controlling, [and]\nemotionally cold.” Dr. Flynn found that Spreitz “slipped\nquietly into alcoholic numbness” after his “failure to live up\nto his mother’s standards or earn her love.” Reflecting\nparental neglect, Spreitz’s parents “failed to arrange . . . for\nan appropriate rehabilitation program” even though\n“[v]irtually everyone . . . who spent much time with him\n\n38 SPREITZ V. RYAN\n\ndescribed him as a heavy drinker.” “Given the intensity and\nduration” of Spreitz’s reliance on alcohol, “the mother (and\nstep-father) were either so uninvolved with him to have\nfailed to notice the problem, or they were aware of it and\nfailed to arrange (or attempt to arrange) for an appropriate\nrehabilitation program.”\n\n Because Spreitz did not receive treatment during his\nteenage years, his dependence on alcohol only deepened. In\nhis sentencing memorandum, Spreitz, too, argued that his\nlong-term alcohol abuse reflects the impact that his\nupbringing by an emotionally unavailable, manipulative,\npunitive mother had on him: it was “[n]o wonder that by 12\nor 13 [h]e began drinking alcohol and smoking marijuana.”\nAlthough the Arizona Supreme Court recognized that\nSpreitz’s “subnormal” upbringing was “a mitigating\ncircumstance,” it “accord[ed] it little weight.” Spreitz I,\n945 P.2d at 1280. Had the court properly considered\nSpreitz’s longstanding history of alcohol abuse in\nconjunction with and as a manifestation of his disturbing\nchildhood, both mitigating circumstances—and most\nimportantly, the two in combination—would have been\nmore compelling.\n\n The dissent argues that Spreitz’s long-term history of\nalcohol abuse cannot be central to his plea for leniency\nbecause much of his mitigation evidence focused on his\n“abusive and dysfunctional childhood, upbringing, and\ninteractions with his mother,” Dissent at 66, not his history\nof alcohol and substance abuse. This argument, however,\nmisses the point, as it overlooks the interlinked relationship\nbetween Spreitz’s alcoholism and his subnormal childhood:\nSpreitz’s history of alcohol and substance abuse is central to\nevaluating the actual significance of his childhood as a\nmitigating factor, and vice versa.\n\n SPREITZ V. RYAN 39\n\n In addition to being a symptom of his dysfunctional\nchildhood, Spreitz’s alcohol and substance abuse can be seen\nas linked to his emotional immaturity, another nonstatutory\nmitigating circumstance recognized by the Arizona Supreme\nCourt but described as not “significant.” Spreitz I, 945 P.2d\nat 1282. In his report, Dr. Flynn explained that Spreitz was\n“very likely to have been socially and emotionally\nimmature” due, at least in part, to “[y]ears of alcoholism\nintoxication [that] wipe[d] out many of the . . . healthy\ndevelopmental processes requisite to age-appropriate social\nand emotional maturity.” Similarly, in his sentencing\nmemorandum, Spreitz connected “[h]is developmental\nfailure” with his “[n]umbing by use of alcohol.” Again, had\nthe Arizona Supreme Court properly considered Spreitz’s\nlongstanding alcohol and substance abuse, another\nmitigating factor—Spreitz’s emotional immaturity—would\nhave been brought into proper perspective.\n\n Thus, by failing to consider Spreitz’s longstanding\nalcohol and substance abuse, the Arizona Supreme Court\nwas left with a critical void in Spreitz’s narrative: Spreitz’s\nsubnormal childhood was so emotionally disturbing that it\nled him to drink by the age of twelve or thirteen, which, in\nturn, disrupted his normal development and contributed to\nhis emotional immaturity. As both a symptom and cause of\nother mitigating factors, Spreitz’s longstanding alcohol\nabuse was thus central to his plea for leniency. 16 Without\n\n 16\n Contrary to the dissent’s suggestion, Spreitz did not focus on his\nintoxication at the time of the crime to the exclusion of his history of\nalcohol abuse. Dissent at 22–24. Rather, both he and Dr. Flynn\ndistinguished between Spreitz’s history of alcohol abuse as nonstatutory\nmitigation and his intoxication at the time of the crime as statutory\nmitigation. It is Spreitz’s history of alcohol abuse that underlies and\namplifies much of his other mitigation evidence, not his intoxication at\nthe time of Reid’s murder.\n\n40 SPREITZ V. RYAN\n\nconsideration of the evidence of his long-term substance and\nalcohol abuse, the Arizona Supreme Court’s conclusions\nregarding all of Spretiz’s mitigation evidence were\nfundamentally flawed.\n\n We come to this conclusion fully recognizing, as the\ndissent emphasizes, that “the severity of an aggravating\ncircumstance must be considered when determining whether\nan Eddings error is harmless.” Dissent at 71. The Arizona\nSupreme Court described what Spreitz’s “confession and\nphysical evidence” revealed: that “Spreitz beat and raped\n[Reid] in a brutal assault that lasted many minutes before he\ncrushed her skull.” Spreitz I, 945 P.2d at 1279. For this\nreason, the Arizona Supreme Court affirmed the sentencing\ncourt’s finding that Reid’s murder was especially cruel. Id.\nEven so, we found prejudicial nexus-error in McKinney\nwhere the murder was done in a cruel manner for pecuniary\ngain as part of a double murder. McKinney, 813 F.3d at 823.\n“When the choice is between life and death,” “the risk that\nthe death penalty will be imposed in spite of facts which may\ncall for a less severe penalty. . . . is unacceptable.” Abdul-\nKabir v. Quarterman, 550 U.S. 233, 264 (2007) (quoting\nLockett v. Ohio, 438 U.S. 586, 605 (1978)). Accordingly,\nthe sentencer “must be able to give meaningful consideration\nand effect to all mitigating evidence that might provide a\nbasis for refusing to impose the death penalty on a particular\nindividual, notwithstanding the severity of his crime.” Id. at\n247 (emphasis added).\n\n Here, the Arizona Supreme Court failed to give any\nmeaning to Spreitz’s longstanding alcohol and substance\nabuse, an oversight which, in turn, minimized the value of\nother mitigating evidence as well. Because the court’s\nanalysis was missing a core component, we “cannot say,\nwith fair assurance” that the Arizona Supreme Court’s\n\n SPREITZ V. RYAN 41\n\nEddings error did not affect its decision. McKinney,\n813 F.3d at 822. As a result, the Arizona Supreme Court’s\nrefusal to consider the evidence was not harmless under\nBrecht.\n\n IV. Conclusion\n\n We reverse in part the district court’s judgment denying\na writ of habeas corpus. We remand with instructions to\ngrant the writ with respect to Spreitz’s sentence unless the\nstate, within a reasonable period, either corrects the\nconstitutional error in his death sentence or vacates the\nsentence and imposes a lesser sentence consistent with the\nlaw.\n\n REVERSED in part and REMANDED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Christopher Spreitz v. Charles Ryan\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-04\\nOpinion type: majority\\nHas dissenting opinion: Yes\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nCHRISTOPHER J. SPREITZ, No. 09-99006\\n Petitioner-Appellant,\\n D.C. No.\\n v. 4:02-CV-00121-JMR\\n\\nCHARLES L. RYAN, Warden,\\n Respondent-Appellee. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n John M. Roll, District Judge, Presiding\\n\\n Argued July 11, 2013\\n Submitted March 4, 2019\\n San Francisco, California\\n\\n Filed March 4, 2019\\n\\n Before: Richard A. Paez, Marsha S. Berzon,\\n and Richard C. Tallman, Circuit Judges.\\n\\n Opinion by Judge Paez;\\n Dissent by Judge Tallman\\n\\n2 SPREITZ V. RYAN\\n\\n SUMMARY *\\n\\n\\n Habeas Corpus / Death Penalty\\n\\n The panel reversed the district court’s denial of habeas\\ncorpus relief with respect to Christopher J. Spreitz’s death\\nsentence, and remanded, in a case in which Spreitz argued\\nthat the Arizona Supreme Court violated Eddings v.\\nOklahoma, 455 U.S. 104 (1982), by refusing to consider, as\\na matter of law, mitigating evidence of Spreitz’s\\nlongstanding alcohol and substance abuse on the ground that\\nhe did not establish a causal connection between this\\nmitigating evidence and the crime.\\n\\n The panel held that the district court erred in concluding\\nthat Spreitz’s claim that the Arizona Supreme Court violated\\nEddings is procedurally defaulted. The panel explained that\\nthe first opportunity Spreitz had to raise that claim was\\nbefore the post-conviction-relief (PCR) court, at which time\\nhe did so.\\n\\n Because the decision of the PCR court – which first\\ndeclared the claim waived, but proceeded to adjudicate the\\nclaim on the merits – was contrary to clearly established\\nSupreme Court precedent, the panel accorded that decision\\nno deference and reviewed Spreitz’s Eddings claim de novo.\\nThe panel concluded that the Arizona Supreme Court\\nviolated Eddings by impermissibly requiring that Spreitz\\nestablish a causal connection between his longstanding\\nsubstance abuse and the murder before considering and\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n SPREITZ V. RYAN 3\\n\\nweighing the evidence as a nonstatutory mitigating factor.\\nThe panel concluded that the error was not harmless.\\n\\n The panel affirmed the district court’s judgment denying\\nrelief with respect to Spreitz’s conviction in a concurrently\\nfiled memorandum disposition.\\n\\n Dissenting, Judge Tallman wrote that the record does not\\nestablish that either the sentencing court or the Arizona\\nSupreme Court unconstitutionally refused to consider\\nrelevant mitigating evidence; and that even if the Arizona\\ncourts did violate Eddings, Spreitz cannot show that this\\nerror had a “substantial and injurious effect or influence” on\\nhis ultimate sentence.\\n\\n\\n COUNSEL\\n\\nTimothy M. Gabrielsen (argued), Assistant Federal Public\\nDefender; John M. Sands, Federal Public Defender; Office\\nof the Federal Public Defender, Tucson, Arizona; Susan B.\\nFox and Sean Bruner, Law Office of Sean Bruner Ltd.,\\nTucson, Arizona; for Petitioner-Appellant.\\n\\nLacey Stover Gard (argued) and Jeffrey A. Zick, Section\\nChief Counsel; Kent E. Cattani, Chief Counsel, Criminal\\nAppeals/Capital Litigation Section; Mark Brnovich,\\nAttorney General; Office of the Attorney General, Tucson,\\nArizona; for Respondents-Appellees.\\n\\n4 SPREITZ V. RYAN\\n\\n OPINION\\n\\nPAEZ, Circuit Judge:\\n\\n In 1994, an Arizona jury convicted Christopher J. Spreitz\\n(“Spreitz”) of first-degree murder. The victim was thirty-\\nnine year old Ruby Reid (“Reid”). Finding that the cruelty\\nof the murder outweighed any mitigating circumstances, the\\ntrial judge sentenced Spreitz to death. Spreitz appeals the\\ndistrict court’s denial of his petition for a writ of habeas\\ncorpus challenging his conviction and sentence. We affirm\\nthe district court’s judgment with respect to Spreitz’s\\nconviction, 1 and reverse with respect to his sentence. 2\\n\\n In challenging his sentence, Spreitz argues that the\\nArizona Supreme Court unconstitutionally affirmed his\\ndeath sentence by failing to consider mitigating evidence of\\nhis longstanding alcohol and substance abuse. He contends\\nthat the state court refused to consider, as a matter of law,\\nthis evidence in mitigation because he did not establish a\\ncausal connection between the crime and his long-term\\nalcohol and substance abuse. In Eddings v. Oklahoma,\\n455 U.S. 104, 110 (1982), the Supreme Court held that under\\nboth the Eighth and Fourteenth Amendments, a sentencer in\\na capital case may not “refuse to consider, as a matter of law,\\n\\n 1\\n We affirm the judgment with respect to Spreitz’s conviction in a\\nconcurrently filed memorandum disposition.\\n\\n 2\\n On February 13, 2015, we vacated submission of Spreitz’s case\\npending final resolution of the en banc proceedings in McKinney v. Ryan,\\n730 F.3d 903 (9th Cir. 2013). In December 2015, the en banc court\\nissued an opinion in McKinney. 813 F.3d 798 (9th Circ. 2015) (en banc).\\nAfter the Supreme Court denied the State’s petition for a writ of\\ncertiorari, Ryan v. McKinney, 137 S. Ct. 39 (2016) (mem), we ordered\\nsupplemental briefing on the applicability of McKinney to Spreitz’s case.\\n\\n SPREITZ V. RYAN 5\\n\\nany relevant mitigating evidence” offered by the defendant.\\nId. at 114. Although a sentencer “may determine the weight\\nto be given relevant mitigating evidence . . . they may not\\ngive it no weight by excluding such evidence from their\\nconsideration.” Id. at 114–15 (footnote omitted). In\\ninterpreting and applying Eddings, the Supreme Court has\\nexplained that “full consideration of evidence that mitigates\\nagainst the death penalty is essential if the [sentencer] is to\\ngive a reasoned moral response to the defendant’s\\nbackground, character, and crime.” Penry v. Lynaugh\\n(Penry I), 492 U.S. 302, 328 (1989), abrogated on other\\ngrounds by Atkins v. Virginia, 536 U.S. 304 (2002) (internal\\nquotation marks and citation omitted). Moreover, the\\nSupreme Court has been clear: requiring a defendant to\\nprove a causal nexus between his mitigating evidence and\\nthe crime is “a test we never countenanced and now have\\nunequivocally rejected.” Smith v. Texas, 543 U.S. 37, 45\\n(2004) (per curiam).\\n\\n At the time of Spreitz’s sentencing, Arizona Revised\\nStatute Annotated § 13-703(G)(1994) 3 listed five mitigating\\nfactors, and Arizona case law additionally recognized\\nnonstatutory mitigating factors, including, for example, a\\ndefendant’s difficult family background or mental condition\\nnot severe enough to qualify as a statutory mitigating factor.\\nIn an en banc decision of our court, McKinney v. Ryan,\\n813 F.3d 798 (9th Cir. 2015), cert denied, 137 S. Ct. 39\\n(2016) (mem), we explained:\\n\\n For a period of a little over 15 years in capital\\n cases, in clear violation of Eddings, the\\n\\n 3\\n Arizona has since revised its death penalty sentencing scheme. All\\nreferences to Arizona’s Revised Statute Annotated are to those\\nprovisions in effect at the time of Spreitz’s sentencing.\\n\\n6 SPREITZ V. RYAN\\n\\n Supreme Court of Arizona articulated and\\n applied a “causal nexus” test for nonstatutory\\n mitigation that forbade as a matter of law\\n giving weight to mitigating evidence, such as\\n family background or mental condition,\\n unless the background or mental condition\\n was causally connected to the crime.\\n\\nId. at 802. As a result, we held in McKinney that\\n“[a]pplication of the causal nexus test to nonstatutory\\nmitigating factors violated Eddings, for it resulted in Arizona\\ncourts being entirely forbidden, as a matter of state law, to\\ntreat as a mitigating factor a family background or a mental\\ncondition that was not causally connected to a defendant’s\\ncrime.” Id. Spreitz argues that the Arizona Supreme Court\\napplied its causal nexus test in his case, refusing to consider\\nevidence of his long-term substance and alcohol abuse\\nbecause he did not adequately establish a causal connection\\nbetween that history of abuse and his crime.\\n\\n As in McKinney, “the precise question before us is\\nwhether the Arizona Supreme Court applied its\\nunconstitutional causal nexus test in affirming [Spreitz]’s\\ndeath sentence on de novo review.” 4 Id. at 804 (emphasis\\n\\n\\n 4\\n The Arizona Supreme Court is required by statute to undertake an\\nindependent review of a death sentence. Ariz. Rev. Stat. § 13-703.01(A).\\nIn conducting that review for crimes that occurred before 2002, the court\\n“independently review[s] the trial court’s findings of aggravation and\\nmitigation and the propriety of the death sentence. In doing so, [the\\ncourt] review[s] the record de novo, considering the quality and the\\nstrength, not simply the number, of aggravating and mitigating factors.”\\nState v. Lynch, 357 P.3d 119, 141 (Ariz. 2015) (internal quotation marks\\nand citations omitted), rev’d on other grounds, Lynch v. Arizona, 136 S.\\nCt. 1818 (2016).\\n\\n SPREITZ V. RYAN 7\\n\\nand internal quotation marks omitted). For the reasons that\\nfollow, we conclude that it did.\\n\\n I.\\n\\n A. Spreitz’s Crimes, Conviction, and Sentence\\n\\n On May 25, 1989, the police arrested Spreitz after\\ndiscovering Ruby Reid’s body in the desert. Upon\\nquestioning, Spreitz confessed to murdering Reid. We\\nbriefly provide the facts of the murder.\\n\\n On the evening of May 18, after drinking heavily and\\nbeing rejected by the woman he was dating, Spreitz “picked\\nup” Reid at a convenience store. State v. Spreitz (Spreitz I),\\n945 P.2d 1260, 1264–65 (Ariz. 1997). 5 In his confession,\\nSpreitz claimed that Reid voluntarily left with him and that\\nhis understanding was that they would have sex later that\\nevening. Id. at 1265. Spreitz further claimed that he drove\\nher out to the desert, where Reid decided she no longer\\nwanted to have sex. Id. The two fought as a result. Id.\\nSpreitz explained that Reid slapped him and that he\\nresponded by punching her in the mouth. Id. Spreitz then\\nsexually assaulted Reid—“remov[ing] her clothing and\\nha[ving] vaginal intercourse with her.” Id. Spreitz also\\nrecounted that he hit Reid in the head multiple times with a\\nrock “to make her stop yelling.” Id. He explained that he\\nleft Reid without knowing whether she was alive or dead.\\nId.\\n\\n Shortly after leaving Reid in the desert, Spreitz was\\nstopped by a Tucson police department officer. Id. at 1264.\\n\\n 5\\n Our recitation of the facts is adopted from the Arizona Supreme\\nCourt’s opinion affirming Spreitz’s conviction and death sentence.\\n\\n8 SPREITZ V. RYAN\\n\\nThe officer observed that Spreitz had a ripped shirt, smelled\\nof feces, and appeared to be covered in blood and fecal\\nmatter. Id. In addition, when detectives later searched\\nSpreitz’s car, they found blood spatter in the trunk, some of\\nwhich was inconsistent with Spreitz’s blood characteristics.\\nId. at 1265.\\n\\n On Monday morning, May 22, Reid’s naked and\\ndecomposing body was discovered on the outskirts of\\nTucson. Id. At trial, “the medical examiner testified that,\\ndue to the advanced state of decomposition, he could not\\ndetermine the full extent and nature of [Reid]’s injuries.” Id.\\nEven so, he was able to observe “bruising on the legs, arms,\\nand back; bruising and abrasions on the buttocks; several\\nbroken ribs; internal bleeding; a broken jaw; several head\\nlacerations; and a skull fracture where the skull had been\\n‘shoved in.’” Id. The medical examiner concluded that Reid\\nhad been killed by “blunt-force trauma to the head.” Id.\\n\\n In addition to finding Reid’s body “[a]t the scene, police\\ndetectives observed tire tracks leading back to the pavement,\\noil stains in the dirt, footprints, and drag marks in the dirt\\nleading away from the body. They also found feces-stained\\npants, tennis shoes, socks, a used tampon, and a torn\\nbrassiere. Two blood-stained rocks lay next to the body.”\\nId. A few days later, police arrested Spreitz. Id.\\n\\n On June 2, 1989, a grand jury indicted Spreitz for first-\\ndegree murder, Ariz. Rev. Stat. Ann. §§ 13-1105, 13-703;\\nsexual assault, Ariz. Rev. Stat. Ann. §§ 13-406(A) & (B);\\nand kidnapping, Ariz. Rev. Stat. Ann. §§ 13-304(A)(3) &\\n(B). Spreitz I, 945 P.2d at 1265. After five years of pre-trial\\nproceedings mostly regarding the admissibility of DNA\\nevidence, a seven-day jury trial began on August 9, 1994.\\nId. at 1266. After the conclusion of the trial, the jury\\nreturned guilty verdicts on all three counts: first-degree\\n\\n SPREITZ V. RYAN 9\\n\\nmurder (both premeditated and felony murder), sexual\\nassault, and kidnapping. Id.\\n\\n Prior to both his aggravation-mitigation and sentencing\\nhearings before the trial judge, Spreitz submitted evidence\\nand a memorandum in support of certain mitigating\\ncircumstances. As noted earlier, at the time of Spreitz’s\\nsentencing, Arizona’s death penalty statutes provided a list\\nof five specific mitigating factors; Arizona case law\\nrecognized nonstatutory mitigating factors as well. See Ariz.\\nRev. Stat. Ann. § 13-703(G); McKinney, 813 F.3d at 802.\\nSpreitz argued as nonstatutory mitigating factors: “(1) his\\ndysfunctional family life and lack of socialization; (2) a\\nhistory of alcohol and drug abuse; (3) his expressions of\\nremorse; [(4)] his good behavior while incarcerated; [(5)] his\\nlack of adult convictions; [and (6)] no prior record of violent\\ntendencies.” Spreitz I, 945 P.2d at 1279. Spreitz argued as\\nstatutory mitigating factors: (1) his age at the time of the\\nmurder, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and (2) that\\nhis “capacity to appreciate the wrongfulness of his conduct\\nor to conform his conduct to the requirements of law was\\nsignificantly impaired [due to alcohol use], but not so\\nimpaired as to constitute a defense to prosecution,” id.\\n§ 13-703(G)(1); see Spreitz I, 945 P.2d at 1279.\\n\\n Spreitz provided evidence of and argued for all the\\nforegoing mitigating circumstances but focused heavily on a\\ncombination of his relationship with his mother and his long\\nhistory of alcohol and substance abuse. To that end, Spreitz\\nsubmitted a written report by and presented testimony from\\nan examining psychologist, Dr. Todd Flynn, Ph.D. After\\nconducting interviews and research, Dr. Flynn concluded\\nthat Spreitz’s longstanding alcohol and substance abuse\\nshould be considered as both a statutory and nonstatutory\\nmitigating factor. In his report, which was admitted into\\n\\n10 SPREITZ V. RYAN\\n\\nevidence at the aggravation and mitigation hearing, Dr.\\nFlynn repeatedly emphasized Spreitz’s longstanding\\nsubstance abuse 6:\\n\\n By age twelve or thirteen, Chris Spreitz\\n began drinking alcohol and smoking\\n marijuana. By age 15, he drank steadily on\\n weekends and would have a shot of vodka\\n before school.\\n\\n The collateral information shows that the\\n alcohol abuse continued to intensify after he\\n left home. A variety of persons . . . described\\n him as a heavy drinker. This includes a\\n second cousin, Scott [Jouett], who saw him to\\n be intoxicated, “a majority of the time,” when\\n he was visiting Santa Barbara a week before\\n the current offense. To the interviewing\\n investigator, Mr. [Jouett] also described,\\n “several different occasions when Chris has\\n blackouts,” while drinking alcohol.\\n\\n It appears completely clear from the available\\n information that Chris Spreitz had a long-\\n standing problem with alcohol which\\n probably reached the level of physical\\n dependence. He described himself as\\n drinking in the morning as early as age 15.\\n Virtually everyone else who spent much time\\n with him described him as a heavy drinker.\\n\\n\\n\\n 6\\n Dr. Flynn testified consistently with his report. For ease of\\nreference, we refer only to his report.\\n\\n SPREITZ V. RYAN 11\\n\\nUltimately, Dr. Flynn concluded that Spreitz’s alcohol\\nabuse, childhood home life, and stunted development,\\ncombined with rejection by his girlfriend on the day of the\\ncrime, and intoxication at the time of the crime, led to the\\nmurder.\\n\\n At the end of his report, Dr. Flynn summarized his\\nfindings and opinions with respect to both statutory and\\nnonstatutory mitigating factors. He opined that both\\nSpreitz’s age, Ariz. Rev. Stat. Ann. § 13-703(G)(5), and\\nimpaired capacity to appreciate the wrongfulness of his\\nconduct or conform his conduct to the law, Ariz. Rev. Stat.\\nAnn. § 13-703(G)(1), were mitigating statutory factors. As\\nto § 13-703(G)(5), specifically, Dr. Flynn opined that a\\ncombination of a disturbing upbringing in a “pathogenic,\\nemotionally neglectful home environment,” and “[y]ears of\\nalcoholism intoxication” combined to cause major deficits in\\nSpreitz’s social and emotional development. As to\\n§ 13-703(G)(1), Dr. Flynn offered a similar conclusion. In\\nlight of Spreitz’s “history of alcoholism . . . , a significant\\nbut unknown degree of alcohol intoxication is likely” on the\\nnight of the crime. In addition, Spreitz’s “history strongly\\nsuggests years of early experiences likely to have caused a\\nbuild-up of pent-up angry, aggressive feelings toward\\nwomen generally (and older women especially) which may\\nhave burst forth with uncontrollable intensity with or without\\nalcohol intoxication.” Dr. Flynn concluded that Spreitz’s\\nintoxication on the night of the crime coupled with his early\\nchildhood experiences “likely . . . contributed to an\\nuncontrollable outburst of aggression” and inability to\\ncontrol his conduct.\\n\\n Dr. Flynn also concluded that certain nonstatutory\\nmitigating factors were present. The nonstatutory mitigating\\nfactors included, in his opinion: Spreitz’s low potential for\\n\\n12 SPREITZ V. RYAN\\n\\nfuture violence, the “failure of [Spreitz’s] parents to provide\\ntreatment for alcohol abuse in [his] teenage years,” and the\\n“emotionally deprived, physically punitive home\\nenvironment” of Spreitz’s upbringing. Dr. Flynn also\\nemphasized that even if both statutory mitigating factors\\nfailed to satisfy the statutory threshold, they may still “be\\nappropriately considered . . . as nonstatutory mitigation.”\\n\\n On November 28, 1994, the trial court conducted an\\naggravation-mitigation hearing. At the hearing, Spreitz\\ncalled three mitigation witnesses: Dr. Flynn and two\\ncorrectional officers from the Pima County jail, where\\nSpreitz was incarcerated. The State did not offer any\\nwitnesses or evidence at the hearing. Dr. Flynn testified\\nconsistently with his report, as detailed above.\\n\\n On the same date, a probation officer filed a Pre-\\nSentence Report, which concluded:\\n\\n It appears the defendant became involved in\\n the senseless commission of the instant\\n offense due to his alcohol and drug abuse.\\n After five years in custody, he now admits his\\n substance abuse problem; however, this does\\n not condone his involvement in the offense.\\n It is unfortunate the victim died before the\\n defendant had his revelation.\\n\\nPrior to the sentencing hearing, Spreitz also submitted\\nseveral letters from friends, family, and jail personnel.\\n\\n On December 21, 1994, the trial court conducted a\\nsentencing hearing. Spreitz, Spreitz’s counsel, the victim’s\\n\\n SPREITZ V. RYAN 13\\n\\nsister, and the prosecutor each addressed the court. 7 The\\nprosecutor first disputed the mitigating evidence presented\\nby Spreitz at the mitigation-aggravation hearing. With\\nrespect to Spreitz’s alcohol abuse, the prosecutor argued that\\nSpreitz’s intoxication at the time of the crime did not meet\\nthe statutory definition for mitigation under § 13-703(G)(1)\\nbecause his conduct both during the murder and afterwards\\ndemonstrated that “he knew what he was doing.” The\\nprosecutor emphasized that the evidence revealed that Reid\\nhad been forcibly abducted and stuffed into the trunk of a\\ncar, thus forcing her to spend time “contemplat[ing] the\\nuncertainty of her fate.” The prosecutor argued that the\\nevidence revealed signs of struggle and that the serious\\ninjuries to Reid’s body belied any notion that Spreitz did not\\nknow what he was doing. The prosecutor further argued that\\nthe fact that Reid defecated on herself revealed that she was\\nterrified prior to her murder. In sum, the prosecutor argued\\nthat given both the heinous nature of the crime and the\\nmanner in which Reid suffered, a finding in aggravation that\\nReid was murdered in an “especially cruel manner” was\\nwarranted.\\n\\n After a short recess, the sentencing judge rendered oral\\nfindings addressing the aggravation and mitigation issues.\\nHe then imposed a sentence of death. 8 Following the\\n\\n\\n\\n 7\\n The trial court did not entertain argument at the conclusion of the\\naggravation-mitigation hearing. As a result, counsel addressed the\\naggravation-mitigation evidence and sentence options at the subsequent\\nsentencing hearing.\\n 8\\n Ring v. Arizona, 536 U.S. 584 (2002), the Supreme Court decision\\nholding judge-sentencing in capital cases unconstitutional, had yet to be\\ndecided.\\n\\n14 SPREITZ V. RYAN\\n\\nhearing, the sentencing judge filed a judgment setting forth\\nwritten findings.\\n\\n The judge first found one aggravating circumstance—\\nthat the offense was committed in an “especially cruel\\nmanner.” Ariz. Rev. Stat. Ann. § 13-703(F)(6). The judge\\nproceeded to review the “many factors [that were] submitted\\nin mitigation,” ultimately finding five mitigating\\ncircumstances of various levels of significance:\\n\\n [1] A mitigating circumstance defense\\n submitted was that the defendant had an\\n extremely disruptive childhood . . . . The\\n court finds the home was sub-normal, not\\n even a minimally healthy one for developing\\n children; and it is obvious the defendant\\n suffered a disruptive middle childhood—had\\n a punitive, controlling, cold mother, who he\\n could not please, no matter what he did.\\n\\n The defendant in his life turned to substance\\n abuse—alcohol and some suggestion he was\\n using cocaine and other drugs. However, the\\n court does not find such is a mitigating\\n circumstance that impaired his ability to\\n make a judgment on whether he was acting\\n rightfully or wrongfully in the death of the\\n victim.\\n\\n The defendant’s history of intoxication is\\n longstanding. He had been abusing\\n substances for close to ten years of his life at\\n the time of this offense when he was twenty-\\n two. Again, the court does not believe that\\n the substance abuse or intoxication impaired\\n\\n SPREITZ V. RYAN 15\\n\\n the defendant’s ability and capacity to\\n appreciate the wrongfulness of his conduct to\\n any significant degree. . . . The court does not\\n believe intoxication is any sort of mitigating\\n circumstance.\\n\\n ...\\n\\n [2] The court acknowledges that the\\n defendant has begun a process of\\n improvement and emotional growth while\\n confined at the Pima County Jail, where he\\n has taken part in education programs.\\n Correctional officers have testified he was a\\n prisoner who caused no problems.\\n\\n [3] The age of the defendant at the time of the\\n offense (twenty-two) is not a mitigating\\n circumstance in and of itself. Immaturity\\n probably is, but the court does not believe\\n immaturity was a significant mitigating\\n circumstance.\\n\\n [4] The court finds that Mr. Spreitz has no\\n criminal history of a felony nature—there is\\n no history or propensity for acts of violence.\\n\\n [5] The court believes the defendant is\\n capable of being rehabilitated. The court\\n does not know whether he has a good\\n prognosis for the future, but the court\\n believes he can be rehabilitated.\\n\\nIn conclusion, the judge found that “the mitigating\\ncircumstances [were] not sufficient to balance the\\n\\n16 SPREITZ V. RYAN\\n\\naggravating circumstances, nor [were] they sufficiently\\nsubstantial to call for leniency.” He thus imposed a death\\nsentence as to the first-degree murder conviction.\\n\\n B. Spreitz’s Appeal and Post-Conviction Proceedings\\n\\n Spreitz appealed his conviction and sentence to the\\nArizona Supreme Court with the assistance of new counsel.\\nThe Arizona Supreme Court affirmed Spreitz’s conviction\\nand sentence after conducting its own independent review of\\nthe record. 9 In reviewing the mitigating evidence, the court\\nfound four of the five mitigating circumstances that the\\nsentencing judge had found, declining to find Spreitz’s good\\nbehavior while in jail awaiting sentencing as a mitigating\\ncircumstance. Spreitz I, 945 P.2d at 1280–81. The court\\nexplained:\\n\\n [1] “We agree with the sentencing judge that\\n defendant’s upbringing was subnormal.” Id.\\n at 1280.\\n\\n [2] “We also find that the sentencing judge\\n . . . properly found that his emotional\\n immaturity was not a significant mitigating\\n factor.” Id. at 1281.\\n\\n [3] “We agree that the record supports the\\n sentencing judge’s findings that defendant\\n had no previous adult felony convictions, no\\n prior acts of violence, and\\n\\n\\n 9\\n See note 4, supra. Because the Arizona Supreme Court conducted\\na de novo review, we focus on its analysis, rather than, as the dissent\\ndoes in part, the analysis by the trial court.\\n\\n SPREITZ V. RYAN 17\\n\\n [4] that the defendant is capable of\\n rehabilitation.” Id.\\n\\nThe Arizona Supreme Court also added that it considered\\nremorse as an additional nonstatutory mitigating factor. Id.\\n\\n The court addressed Spreitz’s history of alcohol and\\nsubstance abuse as follows:\\n\\n The record demonstrates defendant’s\\n longtime substance abuse problems. We\\n note, however, that defendant’s general\\n problems with substance abuse are not\\n essential to our decision here. We therefore\\n decline to conclude that defendant was\\n impaired by alcohol consumption to an extent\\n that it interfered with his “capacity to\\n appreciate the wrongfulness of his conduct or\\n to conform his conduct to the requirements of\\n the law.” A[riz]. R[ev]. S[tat]. [Ann.]\\n § 13-703(G)(1); see also State v. Medrano,\\n 185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)\\n (citing [State v.] Stokley, 182 Ariz. [505,]\\n 520, 898 P.2d [454,] 469 [1995]).\\n\\nId. at 1280–81. In the only other portion of its opinion\\naddressing Spreitz’s history of alcohol and substance abuse,\\nthe court said:\\n\\n The sentencing judge found that defendant’s\\n ability to appreciate the wrongfulness of his\\n conduct was not impaired on the night of the\\n murder to any significant extent by substance\\n abuse, emotional disorders, situational stress,\\n or by a combination of these. Our review of\\n\\n18 SPREITZ V. RYAN\\n\\n the record convinces us that the trial court’s\\n finding was proper.\\n\\nId. at 1281.\\n\\n The Arizona Supreme Court agreed with the trial court’s\\naggravation finding that the murder was done in an\\nespecially cruel manner, id. at 1279, and then proceeded to\\nreweigh the applicable aggravating and mitigating factors,\\nfinding that “the aggravating circumstance of especial\\ncruelty in defendant’s murder of Ruby Reid outweigh[ed] all\\nfactors mitigating in favor of leniency.” Id. at 1282. It\\naffirmed his death sentence. Id. at 1283. Spreitz filed a\\npetition for a writ of certiorari, which the United States\\nSupreme Court denied. Spreitz v. Arizona, 523 U.S. 1027\\n(1998) (mem).\\n\\n In March 2000, represented again by new counsel,\\nSpreitz filed a petition for post-conviction relief in the\\nArizona Superior Court (“PCR court”). See Ariz. R. Crim.\\nP. 32.1. Spreitz alleged, inter alia, that the sentencing judge\\nand the Arizona Supreme Court committed error by failing\\nto consider his history of alcohol and substance abuse as a\\nnonstatutory mitigating factor apart from its causal\\nconnection to the murder—i.e., Eddings error. He also\\nalleged that his appellate counsel was ineffective for failing\\nto raise the sentencing judge’s Eddings error on appeal to the\\nArizona Supreme Court. Spreitz additionally raised several\\nnew claims of trial-counsel-ineffectiveness.\\n\\n The PCR court denied all of Spreitz’s claims and entered\\nan order dismissing his petition for post-conviction relief. In\\nPart 3 of its order, the PCR court discussed Spreitz’s claims\\nof “nexus-error”—that both the trial court and the Arizona\\nSupreme Court erred when they failed to consider Spreitz’s\\nlongstanding alcohol and substance abuse on the basis that\\n\\n SPREITZ V. RYAN 19\\n\\nSpreitz had failed to establish a causal nexus between the\\nlong-term substance abuse and the murder. 10 The PCR court\\ndismissed the claim 11 as waived, reasoning that Spreitz had\\nfailed to raise the issue on direct appeal. Nonetheless, the\\nPCR court addressed the merits of the nexus-error claim in\\nthe course of analyzing Spreitz’s argument that his appellate\\ncounsel was ineffective for failing to raise it. In doing so,\\nthe PCR court explained that the claim failed because:\\n\\n [I]t must be demonstrated, under A[riz].\\n R[ev]. S[tat]. [Ann.] § 13-703(G)(l), that\\n there is a causal link between the history of\\n alcohol or substance abuse and the offense\\n itself. E.g., State v. Stokley, [] 182 Ariz.\\n [505,] 523 [898 P.2d 454, 472 (Ariz. 1995)].\\n Without some basis for explaining or\\n defining the individual’s behavior at the time\\n of the offense, the Petitioner’s history of\\n alcohol or substance abuse would be\\n inconsequential (which is exactly what the\\n trial court and Supreme Court concluded).\\n State v. Kayer, 194 Ariz. 423, 984 P.2d 81\\n ([Ariz.] 1999).\\n\\n At times, the court can and should consider\\n an individual’s long-term alcoholism and\\n\\n 10\\n For ease of reference, we refer to the Arizona courts’ alleged error\\nas “nexus-error.”\\n\\n 11\\n Although Spreitz raised, and the PCR court recognized, two\\ndistinct claims—one with respect to the sentencing court and one with\\nrespect to the Arizona Supreme Court—the PCR court’s analysis treats\\nthe claims as one and the same and refers to them in the singular. To\\nreflect accurately the PCR court’s discussion, our summary here likewise\\nrefers to a singular “claim.”\\n\\n20 SPREITZ V. RYAN\\n\\n substance abuse, usually in conjunction with\\n other factors or diagnosis, as non-statutory\\n mitigation. However, the impact or effect of\\n the alcoholism or substance abuse must be\\n substantial and of such severity that it\\n provides a sufficient basis for explaining the\\n defendant’s conduct, character or ability to\\n control his behavior at the time of the offense.\\n [Citations omitted].\\n\\n As previously discussed, there is no evidence\\n in Petitioner’s case to suggest that he suffered\\n any long-term effects from his alcohol or\\n drug abuse that precluded him from\\n controlling his behavior. Petitioner did not\\n suffer from any cognitive or emotional\\n deficits that rendered him incapable of\\n controlling his conduct. Therefore, the trial\\n court did not err in failing to find Petitioner’s\\n history of alcohol or substance abuse as a\\n separate, non-statutory mitigating factor.\\n [Citation omitted].\\n\\n Spreitz filed a petition for review in the Arizona\\nSupreme Court challenging the PCR court’s judgment. The\\nArizona Supreme Court summarily affirmed the PCR court’s\\nmerits determination with respect to Spreitz’s Eddings\\nclaims. State v. Spreitz (Spreitz II), 39 P.3d 525, 527 (Ariz.\\n2002).\\n\\n Spreitz filed his federal habeas petition in February\\n2003. He alleged in claim seven that the sentencing judge\\nand the Arizona Supreme Court had both committed nexus-\\nerror with respect to his long-time alcohol and substance\\nabuse, and that appellate counsel was ineffective for failing\\n\\n SPREITZ V. RYAN 21\\n\\nto argue the sentencing judge’s nexus-error on direct appeal\\nto the Arizona Supreme Court.\\n\\n The district court denied all of Spreitz’s claims. With\\nrespect to the alleged nexus-error by both the sentencing\\ncourt and the Arizona Supreme Court, the district court\\nfound the claims procedurally barred, relying on the PCR\\ncourt’s determination that Spreitz had waived the claims\\nbecause he could have raised them on direct appeal but failed\\nto do so. The district court concluded that Spreitz had\\nproperly exhausted his ineffective assistance of appellate\\ncounsel claim, and thus addressed the alleged nexus-error in\\nthat context, ultimately concluding it was not meritorious.\\nAccordingly, the district court concluded that “appellate\\ncounsel’s failure to raise this issue on appeal does not\\nconstitute ineffectiveness.”\\n\\n Spreitz timely appealed.\\n\\n II. Jurisdiction and Standard of Review\\n\\n We have jurisdiction pursuant to 28 U.S.C. §§ 1291\\nand 2253(a), and review de novo the district court’s denial\\nof a writ of habeas corpus. Poyson v. Ryan, 879 F.3d 875,\\n887 (9th Cir. 2018). Because Spreitz filed his federal habeas\\npetition after April 24, 1996, he must satisfy the\\nAntiterrorism and Effective Death Penalty Act (“AEDPA”).\\nUnder AEDPA, we may not grant habeas relief unless the\\nstate’s adjudication of Spreitz’s claim (1) “was contrary to\\n. . . clearly established federal law[] as determined by the\\nSupreme Court,” (2) “involved an unreasonable application\\nof” such law, or (3) “was based on an unreasonable\\ndetermination of the facts in light of the evidence presented\\nin the State court proceeding.” (28 U.S.C. § 2254(d)). “In\\nmaking this determination, we look to the last state court\\ndecision to address the claim,” White v. Ryan, 895 F.3d 641,\\n\\n22 SPREITZ V. RYAN\\n\\n665 (9th Cir. 2018) (citing Wilson v. Sellers, 138 S. Ct. 1188,\\n1192 (2018)), which for Spreitz’s nexus-error claim is that\\nof the PCR court.\\n\\n Spreitz argues that the PCR court’s decision was\\n“contrary to” Eddings v. Oklahoma, 455 U.S. 104 (1982).\\n“A state court’s decision is contrary to clearly established\\nfederal law if it ‘applies a rule that contradicts the governing\\nlaw set forth in [U.S. Supreme Court] cases’ or arrives at a\\ndifferent result in a case that ‘confronts a set of facts that are\\nmaterially indistinguishable from a decision of [the\\nSupreme] Court.’” Castellanos v. Small, 766 F.3d 1137,\\n1146 (9th Cir. 2014) (alteration in original) (quoting\\nWilliams v. Taylor, 529 U.S. 362, 405–06 (2000)). “If the\\nstate court applies a legal standard that contradicts clearly\\nestablished [Supreme Court] law, we review de novo the\\napplicant’s claims, applying the correct legal standard to\\ndetermine whether the applicant is entitled to relief.” Id.\\n(citation omitted).\\n\\n III. Discussion\\n\\n As discussed supra, the precise question we must decide\\nis whether the Arizona Supreme Court applied its\\nunconstitutional causal nexus test in violation of Eddings\\nwhen it affirmed Spreitz’s death sentence. To answer that\\nquestion, we must first determine whether the claim is\\nproperly before us. After concluding that Spreitz’s claim\\ncould not have been procedurally defaulted, we turn to the\\nlevel of deference we must accord the PCR court’s ruling on\\nSpreitz’s Eddings claim under AEDPA. Because the PCR\\ncourt’s decision was contrary to clearly established Supreme\\nCourt precedent, we accord that decision no deference and\\nreview Spreitz’s Eddings claim de novo. We conclude that\\nthe Arizona Supreme Court violated Eddings by\\nimpermissibly requiring that Spreitz establish a causal\\n\\n SPREITZ V. RYAN 23\\n\\nconnection between his longstanding substance abuse and\\nthe murder before considering and weighing the evidence as\\na nonstatutory mitigating factor. Finally, having determined\\nthat the Arizona Supreme Court applied an unconstitutional\\ncausal nexus test in its sentencing procedure, we turn to\\nwhether the error was harmless under Brecht v. Abrahamson,\\n507 U.S. 619 (1993). We conclude that the error was not\\nharmless and therefore reverse the district court’s judgment\\nwith respect to Spreitz’s sentence.\\n\\n A.\\n\\n Under the doctrine of procedural default, a federal\\nhabeas court “will not review the merits of claims, including\\nconstitutional claims, that a state court declined to hear\\nbecause the prisoner failed to abide by a state procedural\\nrule.” Martinez v. Ryan, 566 U.S. 1, 9 (2012). “We review\\nde novo a district court’s conclusion that a claim is\\nprocedurally defaulted.” Cooper v. Neven, 641 F.3d 322,\\n326 (9th Cir. 2011).\\n\\n Here, the district court concluded that both of Spreitz’s\\nEddings claims—that the sentencing court and the Arizona\\nSupreme Court applied an unconstitutional causal nexus\\ntest—were procedurally defaulted, because the PCR court\\nfound each claim waived for failure to raise them on direct\\nappeal. This conclusion, however, is erroneous with respect\\nto Spreitz’s claim that the Arizona Supreme Court violated\\nEddings. After all, “[t]he Arizona Supreme Court reviews\\ncapital sentences de novo, making its own determination of\\nwhat constitute[s] legally relevant aggravating and\\nmitigating factors, and then weighing those factors\\nindependently. Ariz. Rev. Stat. Ann. § 13-755. The Arizona\\nSupreme Court ‘conducts a thorough and independent\\nreview of the record and of the aggravating and mitigating\\nevidence to determine whether the sentence is justified.’”\\n\\n24 SPREITZ V. RYAN\\n\\nMcKinney, 813 F.3d at 819 (emphasis removed) (quoting\\nState v. McKinney, 917 P.2d 1214, 1225 (Ariz. 1996)).\\nSpreitz could not have raised on direct appeal his claim that\\nthe Arizona Supreme Court violated Eddings in performing\\nits de novo review of Spreitz’s death sentence. The first\\nopportunity he had to raise that claim was before the PCR\\ncourt, at which time he did so. Thus, Spreitz’s claim is not\\nprocedurally defaulted and is properly before us. 12\\n\\n B.\\n\\n Because Spreitz’s claim that the Arizona Supreme Court\\napplied an unconstitutional causal nexus test to his\\nnonstatutory mitigating evidence of long-time alcohol and\\n 12\\n The district court declined to issue a certificate of appealability\\n(“COA”) on Spreitz’s substantive Eddings claim, which he raised as\\nclaim 1.7 on pages 108 through 111 in his federal habeas petition.\\nAlthough Spreitz did not present his Eddings claim as a separate\\nuncertified issue, we may grant a COA regarding any uncertified issue\\ndiscussed in a petitioner’s opening brief. See 9th Cir. R. 22-1(e).\\nSpreitz’s opening brief discusses the Arizona courts’ Eddings errors. We\\ntherefore may exercise our discretion to expand the COA to encompass\\nthe Eddings claim.\\n\\n To obtain a COA under 28 U.S.C. § 2253(c), Spreitz must\\ndemonstrate that “reasonable jurists could debate whether . . . the\\npetition should have been resolved in a different manner or that the issues\\npresented were adequate to deserve encouragement to proceed further.”\\nSlack v. McDaniel, 529 U.S. 473, 484 (2000). Reasonable jurists could\\ndebate the merits of Spreitz’s Eddings claim. We thus exercise our\\ndiscretion and expand the COA to encompass Spreitz’s substantive\\nEddings claim with respect to the Arizona Supreme Court. We\\nacknowledge that under Ninth Circuit Rule 22-1(f), the State would\\nnormally be afforded an opportunity to submit supplemental briefing on\\nthe uncertified Eddings issue. Here, however, the Eddings issue has been\\nexhaustively briefed, in the context of Spreitz’s appellate counsel’s\\nalleged ineffectiveness before the Arizona Supreme Court. Additional\\nbriefing is unwarranted.\\n\\n SPREITZ V. RYAN 25\\n\\nsubstance abuse is properly before us, we next determine the\\nlevel of deference to accord the PCR court’s adjudication of\\nthat claim. As discussed above, although the PCR court first\\ndeclared the claim waived, it proceeded to adjudicate the\\nclaim on the merits. In denying the claim, the PCR court\\nexplained:\\n\\n Without some basis for explaining or\\n defining the individual’s behavior at the time\\n of the offense, [Spreitz]’s history of alcohol\\n or substance abuse would be inconsequential\\n (which is exactly what the trial court and\\n Supreme Court concluded).\\n\\n At times, the court can and should consider\\n an individual’s long-term alcoholism and\\n substance abuse, usually in conjunction with\\n other factors or diagnosis, as non-statutory\\n mitigation. However, the impact or effect of\\n the alcoholism or substance abuse must be\\n substantial and of such severity that it\\n provides a sufficient basis for explaining the\\n defendant’s conduct, character, or ability to\\n control his behavior at the time of the offense.\\n\\n ....\\n\\n As previously discussed, there is no evidence\\n in [Spreitz]’s case to suggest that he suffered\\n any long-term effects from his alcohol or\\n drug abuse that precluded him, from\\n controlling his behavior. Petitioner did not\\n suffer from any cognitive or emotional\\n deficits that rendered him incapable of\\n controlling his conduct. Therefore, the trial\\n\\n26 SPREITZ V. RYAN\\n\\n court did not err in failing to find [Spreitz]’s\\n history of alcohol or substance abuse as a\\n separate, non-statutory mitigating factor.\\n\\n In other words, the PCR court concluded that Spreitz’s\\nlongstanding alcohol and substance abuse could only be\\nconsidered a nonstatutory mitigating factor if it “provide[d]\\na sufficient basis for explaining the defendant’s conduct,\\ncharacter, or ability to control his behavior at the time of the\\noffense.” This discussion both accurately described the\\ncausal nexus test and approved of it.\\n\\n Such reasoning is contrary to clearly established federal\\nlaw. As explained in McKinney, “the causal nexus test\\nclearly violates Eddings.” 813 F.3d at 810. Therefore,\\n“[b]ecause the state court used the wrong standard, we need\\nnot defer to that decision.” Hardy v. Chappell, 849 F.3d 803,\\n820 (9th Cir. 2016).\\n\\n C.\\n\\n Because the PCR court’s decision was “contrary to”\\nclearly established law, we review de novo the merits of\\nSpreitz’s Eddings claim. 28 U.S.C. § 2254(d)(1). We\\ntherefore consider whether the Arizona Supreme Court\\napplied its unconstitutional causal nexus test in reviewing\\nand affirming Spreitz’s death sentence. In so doing, “we\\nlook only to the decision of th[e Arizona Supreme Court],\\n. . . only [considering] . . . the decision of the sentencing\\njudge . . . to the degree it was adopted or substantially\\nincorporated by the Arizona Supreme Court.” McKinney,\\n813 F.3d at 819.\\n\\n SPREITZ V. RYAN 27\\n\\n 1.\\n\\n Our decision in McKinney frames our consideration of\\nthe Arizona Supreme Court’s decision and Spreitz’s claim.\\nMcKinney makes clear that the Arizona Supreme Court\\nconsistently articulated and applied its unconstitutional\\ncausal nexus test during the period in which Spreitz’s death\\nsentence was litigated in the trial court and reviewed by the\\nArizona Supreme Court. Id. at 824. Moreover, McKinney\\nmakes plain that although on habeas review, we generally\\napply a “presumption that state courts know and follow the\\nlaw,” “the Arizona Supreme Court’s consistent articulation\\nand application of its causal nexus test . . . make such a\\ncourse impossible.” Id. at 803 (internal quotation marks and\\ncitation omitted). In appeals heard from the late 1980s until\\nat least 2002, see State v. Canez, 42 P.3d 564, 594 (Ariz.\\n2002), it was unmistakably clear that “the Arizona Supreme\\nCourt did not know and follow federal law.” Id. (internal\\nquotation marks omitted). 13 Therefore, just as in McKinney\\nand subsequent cases applying McKinney, see, e.g., Poyson,\\n879 F.3d at 889, “the presumption” that the Arizona\\nSupreme Court knew and followed the law “is rebutted” here\\nas well. McKinney, 813 F.3d at 804.\\n\\n That said, McKinney does not dispose of Spreitz’s claim.\\nWe must still review the Arizona Supreme Court’s decision\\nto determine whether it did, in fact, apply its unconstitutional\\ncausal nexus test in Spreitz’s case. In doing so, we are\\nmindful of the fact that at the time the Arizona Supreme\\nCourt decided Spreitz’s appeal, “if there is to be a\\npresumption, it is that the Arizona Supreme Court violated\\n\\n\\n 13\\n In 2005, the Arizona Supreme Court repudiated its earlier nexus\\ntest. See State v. Anderson, 111 P.3d 369, 391–92 (Ariz. 2005).\\n\\n28 SPREITZ V. RYAN\\n\\nthe dictates of Lockett and Eddings.” Greenway v. Ryan,\\n866 F.3d 1094, 1095 (9th Cir. 2017).\\n\\n 2.\\n\\n Against this backdrop, we are convinced that the Arizona\\nSupreme Court applied its causal nexus test with respect to\\nSpreitz’s evidence of longstanding alcohol and substance\\nabuse in violation of Eddings.\\n\\n We quote again the Arizona Supreme Court’s discussion\\nof Spreitz’s history of alcohol and substance abuse:\\n\\n The record demonstrates defendant’s\\n longtime substance abuse problems. We\\n note, however, that defendant’s general\\n problems with substance abuse are not\\n essential to our decision here. We therefore\\n decline to conclude that defendant was\\n impaired by alcohol consumption to an extent\\n that it interfered with his “capacity to\\n appreciate the wrongfulness of his conduct or\\n to conform his conduct to the requirements of\\n the law.” A[riz]. R[ev]. S[tat]. [Ann.] § 13-\\n 703(G)(1); see also State v. Medrano,\\n 185 Ariz. 192, 194, 914 P.2d 225, 227 (1996)\\n (citing Stokley, 182 Ariz. at 520, 898 P.2d at\\n 469).\\n\\n ....\\n\\n The sentencing judge found that defendant’s\\n ability to appreciate the wrongfulness of his\\n conduct was not impaired on the night of the\\n murder to any significant extent by substance\\n abuse, emotional disorders, situational stress,\\n\\n SPREITZ V. RYAN 29\\n\\n or by a combination of these. Our review of\\n the record convinces us that the trial court’s\\n finding was proper.\\n\\nSpreitz I, 945 P.2d at 1280–81. This discussion\\ndemonstrates that the Arizona Supreme Court primarily\\nconcluded that Spreitz failed to show statutory mitigation\\nunder Ariz. Rev. Stat. Ann. § 13-703(G)(1) based on his\\nintoxication on the night of the murder. “When applied\\nsolely in the context of statutory mitigation under\\n§ 13-703(G)(1), the causal nexus test does not violate\\nEddings.” McKinney, 813 F.3d at 810.\\n\\n The plain text of the Arizona Supreme Court’s decision,\\nhowever, also clearly demonstrates that the court did not\\nsolely apply its causal nexus test to Spreitz’s evidence of\\nstatutory mitigation. The court initially recognized Spreitz’s\\n“general problems with substance abuse”—evidence that\\nshould have been relevant as nonstatutory mitigation—but\\nconcluded those problems were “not essential to [its]\\ndecision” because they did not show Spreitz was “impaired\\non the night of the murder.” Spreitz I, 945 P.2d at 1280–81.\\nIn other words, the court did not acknowledge the relevance\\nof longtime substance abuse as nonstatutory mitigation in the\\nabsence of a causal relationship to the crime. Instead, the\\ncourt held that because there was no causal relationship,\\nSpreitz’s long-term alcohol abuse was not of significance—\\ni.e. “not essential”—to the court’s overall determination of\\neither statutory or nonstatutory mitigation.\\n\\n The Supreme Court, however, has been clear that “‘full\\nconsideration of evidence that mitigates against the death\\npenalty is essential if the sentencer is to give a reasoned\\nmoral response to the defendant’s background, character,\\nand crime.’” McKinney, 813 F.3d at 812 (first emphasis\\n\\n30 SPREITZ V. RYAN\\n\\nadded) (quoting Penry I, 492 U.S. at 319, 328).\\nAccordingly, “Eddings requires that ‘[t]he sentencer must\\nalso be able to consider and give effect’” to any relevant\\nmitigation evidence proffered by the defendant, which\\nincludes evidence going toward nonstatutory mitigation. Id.\\n(quoting Penry I, 492 U.S. at 319, 328). The Arizona\\nSupreme Court’s refusal to consider Spreitz’s alcohol and\\nsubstance abuse beyond its connection, if any, to Reid’s\\nmurder constitutes application of its unconstitutional causal\\nnexus test to relevant nonstatutory mitigation evidence.\\n\\n Our understanding of the Arizona Supreme Court’s\\nopinion is bolstered by the fact that in its discussion of\\nSpreitz’s longstanding alcohol and substance abuse, the\\ncourt cited to a prior decision that likewise applied the\\nunconstitutional causal nexus test, State v. Medrano,\\n914 P.2d 255 (Ariz. 1996). In McKinney, we explained that\\nin Medrano, the Arizona Supreme Court upheld the\\nsentencing judge’s determination that the defendant’s\\ncocaine use failed as a nonstatutory mitigating circumstance\\nbecause he did not show that it contributed to his conduct on\\nthe night of the murder. McKinney, 813 F.3d at 825–26. In\\nso doing, “[t]he Arizona Supreme Court applied the causal\\nnexus test.” Id. at 825. Although the Arizona Supreme\\nCourt’s citation to Medrano is not dispositive, it certainly\\ncorroborates our understanding of the court’s opinion. See\\nalso Poyson, 879 F.3d at 890 (emphasizing, among other\\nconsiderations, that the Arizona Supreme Court cited a\\npassage from a case specifically identified in McKinney as\\napplying the unconstitutional causal nexus test in order to\\nconclude that the state court also applied that test to Poyson’s\\nevidence). Moreover, as we explained in McKinney, “the\\nArizona Supreme Court applied its unconstitutional nexus\\ntest consistently” for fifteen years because it had “a mistaken\\nunderstanding of Eddings.” 813 F.3d at 826. It is thus\\n\\n SPREITZ V. RYAN 31\\n\\nentirely logical that in citing the exact portion of Medrano\\nwhere it previously applied its causal nexus test, the Arizona\\nSupreme Court was applying the same test in Spreitz’s case.\\nSee id. at 821; see also Poyson, 879 F.3d at 890.\\n\\n We find further support for our understanding of the\\nArizona Supreme Court’s unconstitutional treatment of\\nSpreitz’s alcohol and substance abuse when we compare it\\nwith the manner in which the court discussed other\\nmitigating factors. As detailed earlier, in evaluating the\\nevidence of Spreitz’s general history of alcohol and\\nsubstance abuse, the court stated that this evidence was “not\\nessential” to its decision. Spreitz I, 945 P.2d at 1280. By\\ncontrast, when the court considered other mitigating factors,\\nit specifically discussed the weight it would give those\\nfactors, as opposed to simply dismissing them outright. For\\nexample, when discussing Spreitz’s subnormal upbringing,\\nthe court stated, “Although we recognize defendant’s\\nupbringing as a mitigating circumstance, we accord it little\\nweight.” Id. at 1280 (emphasis added). As another example,\\nafter “finding that since his arrest, [Spreitz] has\\ndemonstrated remorse,” the court stated that his “remorse for\\nhis actions does little to counterbalance especial cruelty as a\\nserious aggravating circumstance.” Id. at 1281 (emphasis\\nadded). And finally, when discussing the sentencing judge’s\\ndetermination regarding Spreitz’s emotional immaturity, the\\ncourt stated, “We also find that the sentencing judge . . .\\nproperly found that his emotional immaturity was not a\\nsignificant mitigating factor.” Id. (emphasis added). In\\nother words, when the Arizona Supreme Court wanted to\\nassign weight to a given factor, it said so. This pattern stands\\nin stark contrast to the court’s conclusion that anything\\nrelated to Spreitz’s long-term alcohol or substance abuse that\\nfell short of the statutory definition could not serve as\\nmitigation. Id. at 1280–81.\\n\\n32 SPREITZ V. RYAN\\n\\n Finally, as also discussed earlier, the PCR court\\ndistinctly interpreted the Arizona Supreme Court’s decision\\nas having applied a causal nexus test. The PCR court\\nexplained that “[w]ithout some basis for explaining or\\ndefining the individual’s behavior at the time of the offense,\\n[Spreitz]’s history of alcohol or substance abuse would be\\ninconsequential (which is exactly what the trial court and\\nSupreme Court concluded).” Moreover, in articulating how\\nthe Arizona Supreme Court had applied its causal nexus test\\nin Spreitz’s direct appeal, the PCR court, as had the Arizona\\nSupreme Court, cited to a decision of the Arizona Supreme\\nCourt indisputably applying an unconstitutional nexus test,\\nState v. Kayer, 984 P.2d 31 (Ariz. 1999). See McKinney,\\n813 F.3d at 816 (explaining that Kayer held “that the\\ndefendant’s mental impairment ‘was not established as a\\nnonstatutory mitigating factor’ in part because ‘defendant\\noffered no evidence to show the requisite causal nexus that\\nmental impairment affected his judgment or his actions at the\\ntime of the murder’” (quoting Kayer, 984 P.2d at 46)). The\\nfact that the PCR court interpreted the Arizona Supreme\\nCourt’s opinion as we do—and then itself applied the\\nconstitutionally erroneous nexus test—lends further support\\nto our reading.\\n\\n We recognize that the Arizona Supreme Court’s decision\\non the particular issue before us was framed by language that\\nmight, in a decisional vacuum, suggest the court knew,\\nunderstood, and applied the law with respect to Eddings.\\nThe court acknowledged that Spreitz argued that his history\\nof alcohol and drug abuse served as both a statutory and\\nnonstatutory mitigating factor. Spreitz I, 945 P.2d at 1279.\\nIt then stated that it “must consider any aspect of the\\ndefendant’s character or record . . . relevant to determining\\nwhether the death penalty should be imposed.” Id. And\\nfinally, in reaching the conclusion to affirm Spreitz’s death\\n\\n SPREITZ V. RYAN 33\\n\\nsentence, the court explained that it had “examin[ed] the\\nentire record and reweigh[ed] the applicable aggravating and\\nmitigating factors.” Id. at 1281–82.\\n\\n The Arizona Supreme Court’s acknowledgment of what\\nSpreitz argued does not indicate that it relied on or accepted\\nthat argument as correct. Further, we cannot, given\\nMcKinney, Poyson, and Greenway, proceed in a decisional\\nvacuum.\\n\\n Considered in light of the baseline that the Arizona\\nSupreme Court was consistently applying its\\nunconstitutional causal nexus test in death penalty cases at\\nthe time Spreitz’s appeal was decided, as well as its citations\\nto cases held in McKinney to be applying that standard, we\\nare confident that the manner in which the Arizona Supreme\\nCourt disposed of Spreitz’s evidence of his longstanding\\nalcohol and substance abuse confirms that it applied the\\ncausal nexus test it was using during that period.\\n\\n Styers v. Schriro, 547 F.3d 1026 (9th Cir. 2008) supports\\nthis conclusion. Styers stated that the mere fact that “the\\nArizona Supreme Court stated that it had ‘considered all of\\nthe proffered mitigation,’” is not sufficient to contradict the\\nfact that “its analysis prior to this statement indicated\\notherwise.” Id. at 1035 (internal citation omitted). In Styers,\\nnotwithstanding assertions to the contrary, the Arizona\\nSupreme Court refused to consider evidence of the\\ndefendant’s Post Traumatic Stress Disorder (“PTSD”)\\nbecause he had failed to connect it causally to his crimes. Id.\\nHere, too, the Arizona Supreme Court’s actual analysis of\\nSpreitz’s longstanding alcohol and substance abuse reveals\\nthat it refused to consider this particular aspect of Spreitz’s\\ncharacter free from any connection to the crime. Rather, the\\ncourt’s discussion demonstrates that in evaluating Spreitz’s\\nlong-term alcohol and substance abuse, it applied a causal\\n\\n34 SPREITZ V. RYAN\\n\\nnexus test in violation of Eddings and therefore refused, as a\\nmatter of law, to consider it in assessing Spreitz’s evidence\\nof nonstatutory mitigating circumstances.\\n\\n To the extent the state argues the Arizona Supreme\\nCourt’s opinion is ambiguous as to whether it applied an\\nunconstitutional causal nexus test or not—and we believe\\nthat it is not—McKinney requires resolving that ambiguity\\nin favor of Spreitz. McKinney held it clear that the Arizona\\nSupreme Court “consistently articulated and applied its\\ncausal nexus test” at the time the trial court imposed\\nSpreitz’s death sentence and the Arizona Supreme Court\\nreviewed and affirmed that sentence. McKinney, 813 F.3d\\nat 803 (emphasis in original). Again, “if there is to be a\\npresumption, it is that the Arizona Supreme Court violated\\nthe dictates of Lockett and Eddings.” Greenway, 866 at\\n1095.\\n\\n In sum, because (1) the Arizona Supreme Court clearly\\nstated that Spreitz’s longstanding alcohol and substance\\nabuse was “not essential” to its decision, which was in line\\nwith the court’s consistent practice of applying a causal\\nnexus test; (2) the Arizona Supreme Court cited to Medrano,\\nin which it had previously articulated a causal nexus test;\\n(3) the Arizona Supreme Court carefully laid out the weight\\nit would accord other mitigating factors as opposed to simply\\nignoring them; and (4) the PCR court likewise interpreted\\nthe Arizona Supreme Court’s opinion as applying a causal\\nnexus test, we conclude that the Arizona Supreme Court\\nheld, as a matter of law, that Spreitz’s long-term alcohol and\\nsubstance abuse was not a nonstatutory mitigating factor and\\ntherefore refused to consider it. This holding was in clear\\nviolation of Eddings.\\n\\n SPREITZ V. RYAN 35\\n\\n 3.\\n\\n In light of our conclusion that the Arizona Supreme\\nCourt failed to comply with Eddings, we must next decide\\nwhether Spreitz was prejudiced by this error. On federal\\nhabeas review, relief is warranted if an error “had substantial\\nand injurious effect or influence” on the challenged decision.\\nSee Brecht v. Abrahamson, 507 U.S. 619, 623 (1993). The\\npetitioner is “not entitled to habeas relief” unless he can\\nestablish that the error “resulted in actual prejudice.” Davis\\nv. Ayala, 135 S. Ct. 2187, 2197 (2015) (internal quotation\\nmarks omitted). Here, because we are “in grave doubt about\\nwhether [the Arizona Supreme Court’s] error of federal law\\nhad substantial and injurious effect or influence in\\ndetermining [whether to affirm Spreitz’s death sentence],\\nthat error is not harmless. And, [Spreitz] must win.”\\nMcKinney, 813 F.3d at 822 (quoting O’Neal v. McAninch,\\n513 U.S. 432, 435 (1995)).\\n\\n We previously have required resentencing based on\\nprejudicial Eddings error where the Arizona Supreme Court\\nrefused to consider, as a matter of law, mitigation evidence\\n“central to” the petitioner’s “plea for leniency.” See id. at\\n823; see also Coleman v. Calderon, 210 F.3d 1047, 1051\\n(9th Cir. 2000) (concluding that a penalty-phase jury\\ninstruction error was not harmless under Brecht because “it\\nundermined the very core of Coleman’s plea for life”). 14 For\\nthe reasons that follow, it is clear that Spreitz’s long-term\\n\\n 14\\n We do not suggest that Brecht prejudice may only be established\\nif inappropriately considered mitigating evidence was at “the very core\\nof [a defendant’s] plea for life.” Coleman, 210 F.3d at 1051. It is\\nconceivable that other forms of Eddings error could have “a substantial\\nand injurious effect” on a defendant’s sentence if they would be\\nsignificant to the sentencer, despite the defendant’s failure to rely on\\nthem as central to his or her argument. Brecht, 507 U.S. at 623.\\n\\n36 SPREITZ V. RYAN\\n\\nalcohol and substance abuse was important mitigation\\nevidence, and was so closely related as to become central to\\nhis plea for leniency. The Arizona Supreme Court’s refusal\\nto consider that evidence had a substantial and injurious\\neffect on its sentencing determination that requires\\nresentencing.\\n\\n While a defendant’s evidence of longstanding alcohol\\nand/or substance abuse can be probative mitigation evidence\\non its own, a “history of substance abuse [is] substantially\\nmitigating when . . . combined with other mitigating\\nevidence.” Henry v. Ryan, 720 F.3d 1073, 1090 (9th Cir.\\n2013) (emphasis added) (listing cases and holding that the\\ndefendant’s alcoholism “would not have had a substantial\\nand injurious effect” because it “st[ood] alone, was similar\\nto evidence already considered by the sentencing courts, and\\nwas of limited probative value”) (footnote omitted). 15 Here,\\nSpreitz’s evidence regarding his history of alcohol and\\nsubstance abuse—spanning nearly half his life by the time\\nhe committed the crime at the age twenty-two—was the\\n\\n 15\\n The dissent makes much of Henry, suggesting it dictates the\\noutcome of our prejudice analysis. Dissent at 29. Not so. In that case,\\nthe state court had found that Henry’s intoxication on the day of the\\nmurder did constitute a statutory mitigating factor but was not\\nsufficiently substantial to outweigh the two aggravating factors. Henry,\\n720 F.3d at 1090. Henry argued that the state court erred by failing to\\nconsider evidence of his “historical alcoholism” in addition to his\\nintoxication on the day of the crime. Id. We concluded that any error\\nwas harmless because the history of substance abuse “was similar to\\nevidence already considered by the sentencing courts and was of limited\\nprobative value.” Id. In other words, because the state court had found\\ninsufficient evidence of alcohol use that was tied to the murder, we could\\nnot conclude that general alcohol use that was not tied to the murder\\ncould have been determinative. Here, by contrast, the Arizona Supreme\\nCourt found that Spreitz’s evidence of intoxication did not rise to the\\nlevel of statutory mitigation. Accordingly, there is not the same\\ncontextual harmlessness problem as in Henry.\\n\\n SPREITZ V. RYAN 37\\n\\nfactor on which much of his other mitigation evidence\\nturned.\\n\\n In both his report and his testimony, Dr. Flynn discussed\\nat length the extent to which alcohol consumed Spreitz’s life\\nin the decade leading up to Reid’s murder. Spreitz began\\ndrinking when he was twelve or thirteen; by fifteen, he was\\ndrinking steadily on the weekends and drinking vodka before\\nschool. According to Dr. Flynn, “[d]rug and alcohol abuse\\ndominated his teenage years.” After he left home, Spreitz’s\\ndrinking intensified: Spreitz experienced blackouts multiple\\ntimes and, a week before the murder, Spreitz was\\n“intoxicated a majority of the time.” Dr. Flynn concluded\\nthat “[i]t appears completely clear from the available\\ninformation that . . . Spreitz had a long-standing problem\\nwith alcohol which probably reached the level of physical\\ndependence.”\\n\\n Critically, Spreitz’s decade-long struggle with alcohol\\nand other addictive substances reflects the true gravity of his\\n“subnormal upbringing,” a mitigating circumstance\\nrecognized by the Arizona Supreme Court. See Spreitz I,\\n945 P.2d at 1280. Both Spreitz’s sentencing memorandum\\nand Dr. Flynn detailed how his difficult childhood led to\\nSpreitz’s early dependence on alcohol. For instance, Dr.\\nFlynn explained that Spreitz turned to alcohol as a coping\\nmechanism to deal with the “pervasive subabusive\\nemotional battering and neglect that he received from his\\nmother,” whom he described as “punitive, controlling, [and]\\nemotionally cold.” Dr. Flynn found that Spreitz “slipped\\nquietly into alcoholic numbness” after his “failure to live up\\nto his mother’s standards or earn her love.” Reflecting\\nparental neglect, Spreitz’s parents “failed to arrange . . . for\\nan appropriate rehabilitation program” even though\\n“[v]irtually everyone . . . who spent much time with him\\n\\n38 SPREITZ V. RYAN\\n\\ndescribed him as a heavy drinker.” “Given the intensity and\\nduration” of Spreitz’s reliance on alcohol, “the mother (and\\nstep-father) were either so uninvolved with him to have\\nfailed to notice the problem, or they were aware of it and\\nfailed to arrange (or attempt to arrange) for an appropriate\\nrehabilitation program.”\\n\\n Because Spreitz did not receive treatment during his\\nteenage years, his dependence on alcohol only deepened. In\\nhis sentencing memorandum, Spreitz, too, argued that his\\nlong-term alcohol abuse reflects the impact that his\\nupbringing by an emotionally unavailable, manipulative,\\npunitive mother had on him: it was “[n]o wonder that by 12\\nor 13 [h]e began drinking alcohol and smoking marijuana.”\\nAlthough the Arizona Supreme Court recognized that\\nSpreitz’s “subnormal” upbringing was “a mitigating\\ncircumstance,” it “accord[ed] it little weight.” Spreitz I,\\n945 P.2d at 1280. Had the court properly considered\\nSpreitz’s longstanding history of alcohol abuse in\\nconjunction with and as a manifestation of his disturbing\\nchildhood, both mitigating circumstances—and most\\nimportantly, the two in combination—would have been\\nmore compelling.\\n\\n The dissent argues that Spreitz’s long-term history of\\nalcohol abuse cannot be central to his plea for leniency\\nbecause much of his mitigation evidence focused on his\\n“abusive and dysfunctional childhood, upbringing, and\\ninteractions with his mother,” Dissent at 66, not his history\\nof alcohol and substance abuse. This argument, however,\\nmisses the point, as it overlooks the interlinked relationship\\nbetween Spreitz’s alcoholism and his subnormal childhood:\\nSpreitz’s history of alcohol and substance abuse is central to\\nevaluating the actual significance of his childhood as a\\nmitigating factor, and vice versa.\\n\\n SPREITZ V. RYAN 39\\n\\n In addition to being a symptom of his dysfunctional\\nchildhood, Spreitz’s alcohol and substance abuse can be seen\\nas linked to his emotional immaturity, another nonstatutory\\nmitigating circumstance recognized by the Arizona Supreme\\nCourt but described as not “significant.” Spreitz I, 945 P.2d\\nat 1282. In his report, Dr. Flynn explained that Spreitz was\\n“very likely to have been socially and emotionally\\nimmature” due, at least in part, to “[y]ears of alcoholism\\nintoxication [that] wipe[d] out many of the . . . healthy\\ndevelopmental processes requisite to age-appropriate social\\nand emotional maturity.” Similarly, in his sentencing\\nmemorandum, Spreitz connected “[h]is developmental\\nfailure” with his “[n]umbing by use of alcohol.” Again, had\\nthe Arizona Supreme Court properly considered Spreitz’s\\nlongstanding alcohol and substance abuse, another\\nmitigating factor—Spreitz’s emotional immaturity—would\\nhave been brought into proper perspective.\\n\\n Thus, by failing to consider Spreitz’s longstanding\\nalcohol and substance abuse, the Arizona Supreme Court\\nwas left with a critical void in Spreitz’s narrative: Spreitz’s\\nsubnormal childhood was so emotionally disturbing that it\\nled him to drink by the age of twelve or thirteen, which, in\\nturn, disrupted his normal development and contributed to\\nhis emotional immaturity. As both a symptom and cause of\\nother mitigating factors, Spreitz’s longstanding alcohol\\nabuse was thus central to his plea for leniency. 16 Without\\n\\n 16\\n Contrary to the dissent’s suggestion, Spreitz did not focus on his\\nintoxication at the time of the crime to the exclusion of his history of\\nalcohol abuse. Dissent at 22–24. Rather, both he and Dr. Flynn\\ndistinguished between Spreitz’s history of alcohol abuse as nonstatutory\\nmitigation and his intoxication at the time of the crime as statutory\\nmitigation. It is Spreitz’s history of alcohol abuse that underlies and\\namplifies much of his other mitigation evidence, not his intoxication at\\nthe time of Reid’s murder.\\n\\n40 SPREITZ V. RYAN\\n\\nconsideration of the evidence of his long-term substance and\\nalcohol abuse, the Arizona Supreme Court’s conclusions\\nregarding all of Spretiz’s mitigation evidence were\\nfundamentally flawed.\\n\\n We come to this conclusion fully recognizing, as the\\ndissent emphasizes, that “the severity of an aggravating\\ncircumstance must be considered when determining whether\\nan Eddings error is harmless.” Dissent at 71. The Arizona\\nSupreme Court described what Spreitz’s “confession and\\nphysical evidence” revealed: that “Spreitz beat and raped\\n[Reid] in a brutal assault that lasted many minutes before he\\ncrushed her skull.” Spreitz I, 945 P.2d at 1279. For this\\nreason, the Arizona Supreme Court affirmed the sentencing\\ncourt’s finding that Reid’s murder was especially cruel. Id.\\nEven so, we found prejudicial nexus-error in McKinney\\nwhere the murder was done in a cruel manner for pecuniary\\ngain as part of a double murder. McKinney, 813 F.3d at 823.\\n“When the choice is between life and death,” “the risk that\\nthe death penalty will be imposed in spite of facts which may\\ncall for a less severe penalty. . . . is unacceptable.” Abdul-\\nKabir v. Quarterman, 550 U.S. 233, 264 (2007) (quoting\\nLockett v. Ohio, 438 U.S. 586, 605 (1978)). Accordingly,\\nthe sentencer “must be able to give meaningful consideration\\nand effect to all mitigating evidence that might provide a\\nbasis for refusing to impose the death penalty on a particular\\nindividual, notwithstanding the severity of his crime.” Id. at\\n247 (emphasis added).\\n\\n Here, the Arizona Supreme Court failed to give any\\nmeaning to Spreitz’s longstanding alcohol and substance\\nabuse, an oversight which, in turn, minimized the value of\\nother mitigating evidence as well. Because the court’s\\nanalysis was missing a core component, we “cannot say,\\nwith fair assurance” that the Arizona Supreme Court’s\\n\\n SPREITZ V. RYAN 41\\n\\nEddings error did not affect its decision. McKinney,\\n813 F.3d at 822. As a result, the Arizona Supreme Court’s\\nrefusal to consider the evidence was not harmless under\\nBrecht.\\n\\n IV. Conclusion\\n\\n We reverse in part the district court’s judgment denying\\na writ of habeas corpus. We remand with instructions to\\ngrant the writ with respect to Spreitz’s sentence unless the\\nstate, within a reasonable period, either corrects the\\nconstitutional error in his death sentence or vacates the\\nsentence and imposes a lesser sentence consistent with the\\nlaw.\\n\\n REVERSED in part and REMANDED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1273,"cells":{"id":{"kind":"number","value":4597698,"string":"4,597,698"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Vijayakumar Thuraissigiam v. Usdhs"},"decision_date":{"kind":"string","value":"2019-03-07"},"docket_number":{"kind":"string","value":"18-55313"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: A. Wallace Tashima, M. Margaret McKeown, and Richard A. Paez, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"TASHIMA, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nVIJAYAKUMAR THURAISSIGIAM, No. 18-55313\\n Petitioner-Appellant,\\n D.C. No.\\n v. CV 18-135 AJB\\n\\nU.S. DEPARTMENT OF HOMELAND\\nSECURITY; U.S. CUSTOMS AND\\nBORDER PROTECTION; U.S. OPINION\\nCITIZENSHIP AND IMMIGRATION\\nSERVICES; U.S. IMMIGRATION AND\\nCUSTOMS ENFORCEMENT; KIRSTJEN\\nNIELSEN, Secretary of DHS;\\nWILLIAM P. BARR, Attorney\\nGeneral; KEVIN K. MCALEENAN,\\nActing Commissioner of CBP;\\nTHOMAS HOMAN; L. FRANCIS\\nCISSNA, Director of USCIS; PETE\\nFLORES, San Diego Field Director,\\nCBP; GREGORY ARCHAMBEAULT,\\nSan Diego Field Office Director,\\nICE; FRED FIGUEROA, Warden, Otay\\nMesa Detention Center,\\n Respondents-Appellees.\\n\\n\\n Appeal from the United States District Court\\n for the Southern District of California\\n Anthony J. Battaglia, District Judge, Presiding\\n\\f2 THURAISSIGIAM v. USDHS\\n\\n Argued and Submitted May 17, 2018\\n Portland, Oregon\\n\\n Filed March 7, 2019\\n\\n Before: A. Wallace Tashima, M. Margaret McKeown,\\n and Richard A. Paez, Circuit Judges.\\n\\n Opinion by Judge Tashima\\n\\n\\n SUMMARY*\\n\\n\\n Immigration\\n\\n In an action in which Vijayakumar Thuraissigiam filed a\\nhabeas petition to challenge procedures leading to his\\nexpedited removal order, the panel reversed the district\\ncourt’s dismissal of the petition for lack of subject matter\\njurisdiction, held that 8 U.S.C. § 1252(e)(2) violates the\\nSuspension Clause as applied to Thuraissigiam, and\\nremanded.\\n\\n Under 8 U.S.C. § 1225(b)(1)(A)(i), when a U.S. Customs\\nand Border Protection (“CBP”) officer determines that a\\nnoncitizen arriving at a port of entry is inadmissible for\\nmisrepresenting a material fact or lacking necessary\\ndocumentation, the officer must place the noncitizen in so-\\ncalled “expedited removal” proceedings. Expedited removal\\nalso applies to inadmissible noncitizens arrested within 100\\n\\n *\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f THURAISSIGIAM v. USDHS 3\\n\\nmiles of the border and unable to prove that they have been\\nin the United States for more than the prior two weeks.\\n\\n The Department of Homeland Security (“DHS”) removes\\nnoncitizens eligible for expedited removal without further\\nhearing or review, subject to only one exception: If, in an\\ninterview with a CBP officer, the noncitizen indicates an\\nintent to apply for asylum or a fear of persecution, DHS must\\nrefer the noncitizen for an interview with an asylum officer.\\nIf the asylum officer finds no credible fear of persecution, the\\nnoncitizen will be removed. A supervisor reviews the asylum\\nofficer’s credible fear determination, and the noncitizen may\\nalso request de novo review by an immigration judge.\\n\\n Congress sharply circumscribed judicial review of the\\nexpedited removal process such that “no court shall have\\njurisdiction to review . . . any individual determination [or]\\n. . . the application of [§ 1225(b)(1)] to individual aliens”\\noutside of the review permitted by the habeas review\\nprovision, § 1252(e). 8 U.S.C. § 1252(a)(2)(A)(iii). Under\\n§ 1252(e)(2), a person in expedited removal proceedings may\\nfile a habeas petition in federal district court to contest three\\nDHS determinations: whether the person is a noncitizen,\\nwhether he “was ordered removed” via expedited removal,\\nand whether he is a lawful permanent resident or has another\\nstatus exempting him from expedited removal.\\n\\n Thuraissigiam is a native and citizen of Sri Lanka and a\\nTamil, an ethnic minority group in Sri Lanka. After crossing\\ninto the United States, he was arrested 25 yards north of the\\nMexican border, and placed in expedited removal\\nproceedings. He was referred for a credible fear interview\\nafter he indicated a fear of persecution in Sri Lanka, but an\\nasylum officer determined that Thuraissigiam had not\\n\\f4 THURAISSIGIAM v. USDHS\\n\\nestablished a credible fear of persecution. A supervisor\\napproved the decision, and an immigration judge affirmed the\\nnegative credible fear finding in a check-box decision and\\nreturned the case to DHS for Thuraissigiam’s removal.\\n\\n Thuraissigiam filed a habeas petition in federal district\\ncourt, arguing that his expedited removal order violated his\\nstatutory, regulatory, and constitutional rights. The district\\ncourt dismissed the petition for lack of subject matter\\njurisdiction, concluding that § 1252(e) did not authorize\\njurisdiction over Thuraissigiam’s claims and rejecting his\\nSuspension Clause arguments.\\n\\n The panel concluded, in line with this court’s precedents,\\nthat § 1252(e)(2) does not authorize jurisdiction over\\nThuraissigiam’s petition because § 1252(e)(2) limits a district\\ncourt to reviewing three basic factual determinations related\\nto an expedited removal order, and Thuraissigiam’s petition\\ndoes not challenge any of those determinations.\\n\\n The panel next considered whether the habeas review\\navailable to Thuraissigiam under § 1252(e)(2) satisfied the\\nrequirements of the Suspension Clause. In doing so, the\\npanel observed that Boumediene v. Bush, 553 U.S. 723\\n(2008), provides an analytical template for evaluating a\\nSuspension Clause challenge: at step one, the court examines\\nwhether the Suspension Clause applies to the petitioner; and,\\nif so, at step two, the court examines whether the substitute\\nprocedure provides review that satisfies the Clause.\\n\\n The panel also observed that INS v. St. Cyr, 533 U.S. 289,\\n302 (2001), illuminates how to approach both Boumediene\\nsteps. The panel explained that, like Boumedine, St. Cyr\\nlooked to the 1789-era historical application of the writ of\\n\\f THURAISSIGIAM v. USDHS 5\\n\\nhabeas corpus, but that St. Cyr also looked to habeas\\nprecedents from the co-called “finality era,” a period from\\n1891 to 1952 during which the statutory scheme precluded\\njudicial intervention in immigration enforcement, except as\\nrequired by the Constitution.\\n\\n Drawing from Boumedine and St. Cyr, the panel\\nconcluded that it must evaluate Thuraissigiam’s Suspension\\nClause challenge in two steps: First, to determine whether\\nThuraissigiam may invoke the Suspension Clause, the court\\nexamines 1789-era practice, the finality era cases, and other\\nrelevant precedents. Second, the court asks whether\\n§ 1252(e)(2) provides Thuraissigiam a “meaningful\\nopportunity to demonstrate that he is being held pursuant to\\n‘the erroneous application or interpretation’ of relevant law.”\\nBoumediene, 553 U.S. at 779 (quoting St. Cyr, 533 U.S. at\\n302).\\n\\n In applying Boumediene step one, which concerns the\\nreach of the Suspension Clause, the panel concluded that\\nThuraissigiam, who was arrested within the United States,\\nmay invoke the Suspension Clause. In so concluding, the\\npanel observed that, in the finality era, the Court permitted\\neven arriving noncitizens to invoke habeas review.\\n\\n The panel considered at Boumediene step two whether\\nhabeas review under § 1252(e) is so limited so as effectively\\nto suspend the writ as applied to Thuraissigiam. In doing so,\\nthe panel rejected the government’s contention that because,\\nin its view, Thuraissigiam lacks due process rights, there are\\nno rights for the Suspension Clause to protect, explaining that\\nBoumediene foreclosed that argument by holding that,\\nwhether or not due process was satisfied, the Suspension\\nClause might require more. Reviewing the relevant\\n\\f6 THURAISSIGIAM v. USDHS\\n\\nprecedent, the panel concluded that the Suspension Clause\\nrequires review of Thuraissigiam’s claims.\\n\\n The panel held that 8 U.S.C. § 1252(e)(2) violates the\\nSuspension Clause as applied to Thuraissigiam. Noting the\\nmeager procedural protections afforded by the administrative\\nscheme governing credible fear determinations in this\\ncontext, and the fact that § 1252(a)(2) prevents any judicial\\nreview of whether DHS complied with the procedures in an\\nindividual case or applied the correct standard, the panel\\nconcluded that it is obvious that the constitutional minimum\\n– whether Thuraissigiam was detained pursuant to the\\n“erroneous interpretation or application of relevant law” – is\\nnot satisfied by the scheme set out by § 1252(e)(2).\\n\\n The panel further declined to interpret § 1252(e)(2) to\\navoid the serious Suspension Clause problems engendered by\\nthe statute, explaining that the constitutional avoidance canon\\ndoes not apply because the statute cannot bear a reading that\\navoids the constitutional problems it creates.\\n\\n Finally, the panel observed that it did not profess to\\ndecide in this opinion what right or rights Thuraissigiam may\\nvindicate via use of the writ, and remanded the case for the\\ndistrict court to exercise jurisdiction to consider\\nThuraissigiam’s legal challenges to the procedures leading to\\nhis expedited removal order.\\n\\f THURAISSIGIAM v. USDHS 7\\n\\n COUNSEL\\n\\nLee P. Gelernt (argued) and David Hausman, American Civil\\nLiberties Union Foundation Immigrants’ Rights Project, New\\nYork, New York; Jennifer Chang Newell and Cody Wofsy,\\nAmerican Civil Liberties Union Foundation Immigrants’\\nRights Project, San Francisco, California; David Loy, ACLU\\nFoundation of San Diego & Imperial Counties, San Diego,\\nCalifornia; for Petitioner-Appellant.\\n\\nJoshua S. Press (argued), and Joseph A. Darrow, Trial\\nAttorneys; Erez Reuveni, Assistant Director; William C.\\nPeachey, Director; Office of Immigration Litigation, Civil\\nDivision, United States Department of Justice, Washington,\\nD.C.; for Respondent-Appellees.\\n\\nMatthew E. Price, Jenner & Block, New York, New York;\\nBlaine Bookey, Karen Musalo, and Eunice Lee, Center for\\nGender & Refugee Studies, University of California Hastings\\nCollege of the Law, San Francisco, Calfornia; for Amici\\nCuriae Refugee and Human Rights Organizations and\\nScholars.\\n\\nEthan D. Dettmer, Soolean Choy, Priyah Kaul, and Eli M.\\nLazarus, Gibson Dunn & Crutcher, San Francisco, California;\\nJoshua S. Lipshutz, Gibson Dunn & Crutcher, Washington,\\nD.C.; for Amici Curiae Scholars of Immigration Law.\\n\\nNoah A. Levine, Wilmer Cutler Pickering Hale & Dorr, New\\nYork, New York, for Amici Curiae Scholars of Habeas\\nCorpus Law.\\n\\f8 THURAISSIGIAM v. USDHS\\n\\nAnjali Srinivasan and Leo L. Lam, Keker Van Nest & Peters,\\nSan Francisco, California, for Amici Curiae Scholars of Sri\\nLankan Politics.\\n\\n\\n OPINION\\n\\nTASHIMA, Circuit Judge:\\n\\n Vijayakumar Thuraissigiam filed a habeas petition in\\ndistrict court pursuant to 8 U.S.C. § 1252(e)(2) to challenge\\nthe procedures leading to his expedited removal order. The\\ncourt dismissed the petition for lack of subject matter\\njurisdiction. We reverse. Although § 1252(e)(2) does not\\nauthorize jurisdiction over the claims in Thuraissigiam’s\\npetition, the Suspension Clause, U.S. Const. art. I, § 9, cl. 2,\\nrequires that Thuraissigiam have a “meaningful opportunity\\nto demonstrate that he is being held pursuant to ‘the\\nerroneous application or interpretation’ of relevant law.”\\nBoumediene v. Bush, 553 U.S. 723, 779 (2008) (quoting INS\\nv. St. Cyr, 533 U.S. 289, 302 (2001)). Because § 1252(e)(2)\\ndoes not provide that meaningful opportunity, the provision\\nviolates the Suspension Clause as applied to Thuraissigiam.\\n\\n BACKGROUND\\n\\nI. Statutory Background\\n\\n When a U.S. Customs and Border Protection (“CBP”)\\nofficer determines that a noncitizen arriving at a port of entry\\nis inadmissible for misrepresenting a material fact or lacking\\n\\f THURAISSIGIAM v. USDHS 9\\n\\nnecessary documentation,1 the officer must place the\\nnoncitizen in so-called “expedited removal” proceedings.\\n8 U.S.C. § 1225(b)(1)(A)(i). By regulation, the Department\\nof Homeland Security (“DHS”), of which CBP is a\\nconstituent agency, also applies expedited removal to\\ninadmissible noncitizens arrested within 100 miles of the\\nborder and unable to prove that they have been in the United\\nStates for more than the prior two weeks. Designating Aliens\\nfor Expedited Removal, 69 Fed. Reg. 48877-01, 48879-80\\n(Aug. 11, 2004);2 see also 8 U.S.C. § 1225(b)(1)(A)(iii)(II).\\n\\n DHS removes noncitizens eligible for expedited removal\\n“without further hearing or review,” subject to only one\\nexception. 8 U.S.C. § 1225(b)(1)(A)(i). If, in an interview\\nwith a CBP officer, the noncitizen indicates an intent to apply\\nfor asylum or a fear of persecution, DHS must refer the\\nnoncitizen for an interview with an asylum officer. Id.\\n§ 1225(b)(1)(A)(ii); 8 C.F.R. § 208.30. If that asylum officer\\ndetermines that the noncitizen’s fear of persecution is\\n\\n\\n 1\\n See 8 U.S.C. § 1182(a)(6)(C) (misrepresentation bar); id.\\n§ 1182(a)(7) (documentation bar).\\n 2\\n Congress gave the Attorney General the authority to extend\\nexpedited removal to some or all inadmissible noncitizens who cannot\\nprove that they have been in the United States for more than two years\\nprior; thus, the current regime does not represent the full exercise of\\nexecutive authority permitted by statute. 8 U.S.C. § 1225(b)(1)(A)(iii).\\nDHS also applies expedited removal to noncitizens who entered the\\nUnited States by sea and who have not been in the United States for two\\nyears. See Notice Designating Aliens Subject to Expedited Removal\\nUnder Section 235(b)(1)(A)(iii) of the Immigration and Nationality Act,\\n67 Fed. Reg. 68,924, 68,924–25 (Nov. 13, 2002). The current regime\\nmay, however, expand; a January 2017 executive order instructs the\\nSecretary of DHS to apply expedited removal to the fullest extent of the\\nlaw. See Exec. Order No. 13,767, 82 Fed. Reg. 8793 (Jan. 25, 2017).\\n\\f10 THURAISSIGIAM v. USDHS\\n\\ncredible, the noncitizen is referred to non-expedited removal\\nproceedings, in which the noncitizen may apply for asylum or\\nother forms of relief from removal. See 8 U.S.C.\\n§ 1225(b)(1)(B)(ii); 8 C.F.R. § 208.30(f); 8 C.F.R.\\n§ 1003.42(f). If the asylum officer finds no credible fear of\\npersecution, the noncitizen will be removed. 8 U.S.C.\\n§ 1225(b)(1)(B)(iii). A supervisor reviews the asylum\\nofficer’s credible fear determination, 8 C.F.R.\\n§§ 208.30(e)(7), 235.3(b)(2), (b)(7), and a noncitizen may\\nalso request de novo review by an immigration judge (“IJ”).\\n8 U.S.C. § 1225(b)(1)(B)(iii)(III); 8 C.F.R. § 1003.42. In\\n2016, DHS conducted over 141,000 expedited removals. See\\nRefugee and Human Rights Amicus Br. 10. All individuals\\nplaced in expedited removal proceedings are subject to\\nmandatory detention pending a final determination of credible\\nfear of persecution or until they are removed. 8 U.S.C.\\n§ 1225(b)(1)(B)(iii)(IV).\\n\\n Congress sharply circumscribed judicial review of the\\nexpedited removal process. “[N]o court shall have\\njurisdiction to review . . . any individual determination [or]\\n. . . the application of [§ 1225(b)(1)] to individual aliens”\\noutside of the review permitted by the habeas review\\nprovision, § 1252(e). 8 U.S.C. § 1252(a)(2)(A)(iii). Under\\n§ 1252(e)(2), a person in expedited removal proceedings may\\nfile a habeas petition in federal district court to contest three\\nDHS determinations: whether the person is a noncitizen,\\nwhether he “was ordered removed” via expedited removal,\\nand whether he is a lawful permanent resident or has another\\nstatus exempting him from expedited removal. Id.\\n§ 1252(e)(2)(A)–(C). Review of whether a petitioner “was\\nordered removed” is “limited to whether such an order in fact\\nwas issued and whether it relates to the petitioner. Id.\\n§ 1252(e)(5). “There shall be no review of whether the alien\\n\\f THURAISSIGIAM v. USDHS 11\\n\\nis actually inadmissible or entitled to any relief from\\nremoval.” Id.; see also 8 C.F.R. § 1003.42(f) (“No appeal\\nshall lie from a review of an adverse credible fear\\ndetermination made by an immigration judge.”).3\\n\\nII. Factual Background\\n\\n Thuraissigiam is a native and citizen of Sri Lanka and a\\nTamil, an ethnic minority group in Sri Lanka. See Scholars\\nof Sri Lankan Politics Amicus Br. 3. Thuraissigiam fled his\\nhome country in June 2016 and made his way to Mexico. On\\nFebruary 17, 2017, Thuraissigiam crossed the border into the\\nUnited States. Late that night, he was arrested by a CBP\\nofficer four miles west of the San Ysidro, California, port of\\nentry, 25 yards north of the border.\\n\\n DHS placed Thuraissigiam in expedited removal\\nproceedings. Pursuant to 8 U.S.C. § 1225(b)(1)(A)(ii), CBP\\nreferred Thuraissigiam for an interview with an asylum\\nofficer after he indicated a fear of persecution in Sri Lanka.\\nOn March 9, an asylum officer from the United States\\nCitizenship and Immigration Services (“USCIS”) interviewed\\nThuraissigiam and determined that he had not established a\\ncredible fear of persecution. A supervisor approved the\\n\\n\\n 3\\n Under 8 U.S.C. § 1252(e)(3), a person may challenge the\\nconstitutionality and legality of the expedited removal provisions,\\nregulations implementing those provisions, or written policies to\\nimplement the provisions. Such challenges, however, must be brought\\nwithin 60 days after implementation and only in the District of Columbia.\\nId. § 1252(e)(3)(A)–(B). Various expedited removal provisions and\\nimplementing regulations survived a § 1252(e)(3) challenge in American\\nImmigration Lawyers Ass’n v. Reno, 18 F. Supp. 2d 38 (D.D.C. 1998),\\nalthough the plaintiffs did not raise a Suspension Clause argument about\\nthe extent of habeas review. See id. at 41.\\n\\f12 THURAISSIGIAM v. USDHS\\n\\ndecision. Thuraissigiam then requested review by an IJ, who\\naffirmed the negative credible fear finding in a check-box\\ndecision and returned the case to DHS for Thuraissigiam’s\\nremoval.\\n\\nIII. District Court Proceedings\\n\\n In January 2018, Thuraissigiam filed a habeas petition in\\nfederal district court, naming as respondents DHS, several of\\nits constituent agencies, and individual agency officials.\\nThuraissigiam argued that his “expedited removal order\\nviolated his statutory, regulatory, and constitutional rights,”\\nsought to vacate the order, and requested relief in the form of\\na “new, meaningful opportunity to apply for asylum and other\\nrelief from removal.” Thuraissigiam alleged that in Sri Lanka\\nhe had been harassed for supporting a Tamil political\\ncandidate. In 2007, he was “detained and beaten” by Sri\\nLankan army officers, and told not to support the candidate.\\nIn 2014, after Thuraissigiam continued to support the\\ncandidate, government intelligence officers kidnapped,\\nbound, and beat him during an interrogation about his\\npolitical activities. Thuraissigiam alleged that he “was\\nlowered into a well, simulating drowning, threatened with\\ndeath, and then suffocated, causing him to lose\\nconsciousness.”\\n\\n Thuraissigiam also made various factual allegations about\\nthe expedited removal procedures to which he was subject\\nafter being apprehended. For one, he alleged that the asylum\\nofficer failed to “elicit all relevant and useful information\\nbearing on whether the applicant has a credible fear of\\npersecution or torture” in violation of 8 C.F.R. § 208.30(d)\\nand “failed to consider relevant country conditions evidence”\\nin violation of 8 U.S.C. § 1225(b)(1)(B)(v) and 8 C.F.R.\\n\\f THURAISSIGIAM v. USDHS 13\\n\\n§ 208.30(e)(2). Thuraissigiam also alleged that there were\\n“communication problems” between the asylum officer,\\nThuraissigiam, and the translator, in violation of 8 C.F.R.\\n§ 208.30(d)(1)–(2). Thuraissigiam alleged that the IJ hearing\\nincluded the same procedural and substantive flaws, and that\\nat both hearings, he was unaware whether “information he\\noffered would be shared with the Sri Lankan government.”\\nThuraissigiam’s petition asserted two causes of action:\\n\\n First, DHS’ credible fear screening deprived\\nThuraissigiam “of a meaningful right to apply for asylum”\\nand other forms of relief, in violation of 8 U.S.C.\\n§ 1225(b)(1), its implementing regulations, and the United\\nStates Convention Against Torture, implemented in the\\nForeign Affairs Reform and Restructuring Act of 1998\\n(“FARRA”), Pub. L. No. 105-277, div. G., Title XXII,\\n§ 2242, 112 Stat. 2681 (1998). The asylum officer and IJ also\\nviolated those statutes “by applying an incorrect legal\\nstandard” to Thuraissigiam’s credible fear application.\\n\\n Second, the asylum officer and IJ violated\\nThuraissigiam’s rights under the Due Process Clause of the\\nFifth Amendment by “not providing him with a meaningful\\nopportunity to establish his claims, failing to comply with the\\napplicable statutory and regulatory requirements, and in not\\nproviding him with a reasoned explanation for their\\ndecisions.”\\n\\n The district court dismissed the habeas petition for lack of\\nsubject matter jurisdiction. Thuraissigiam v. U.S. Dep’t of\\nHomeland Sec., 287 F. Supp. 3d 1077 (S.D. Cal. 2018).\\nRelying on our precedents, the district court concluded that\\n8 U.S.C. § 1252(e) did not authorize jurisdiction over the\\nclaims in Thuraissigiam’s petition. Id. at 1082. Next, the\\n\\f14 THURAISSIGIAM v. USDHS\\n\\ncourt rejected Thuraissigiam’s Suspension Clause arguments.\\nAlthough the court concluded that Thuraissigiam could\\ninvoke the Suspension Clause, it held that the statute’s “strict\\nrestraints” on habeas review of expedited removal orders did\\nnot effectively suspend the writ of habeas corpus and were\\ntherefore constitutionally sound. Id. at 1082–83.4\\n\\n Thuraissigiam timely appealed the district court’s\\ndismissal and moved for a stay of removal pending appeal.\\nA motions panel of our court initially denied Thuraissigiam’s\\nstay motion, but later vacated that order and stayed\\nThuraissigiam’s removal pending appeal.\\n\\n STANDARD OF REVIEW\\n\\n We have jurisdiction under 28 U.S.C. § 1291. We review\\nde novo the district court’s dismissal of Thuraissigiam’s\\nhabeas petition for lack of subject matter jurisdiction. Bishop\\nPaiute Tribe v. Inyo County, 863 F.3d 1144, 1151 (9th Cir.\\n2017); see also Garcia de Rincon v. Dep’t of Homeland Sec.,\\n539 F.3d 1133, 1136 (9th Cir. 2008).\\n\\n DISCUSSION\\n\\n We must decide whether a federal district court has\\njurisdiction to review the claims in Thuraissigiam’s petition.\\nWe first inquire whether 8 U.S.C. § 1252(e)(2) authorizes\\njurisdiction over Thuraissigiam’s petition. Concluding that\\n§ 1252(e)(2) does not authorize jurisdiction, we then address\\n\\n\\n\\n 4\\n The district court also denied various stay motions that\\nThuraissigiam had filed, concluding that they were moot due to the\\npetition’s dismissal. 287 F. Supp. 3d at 1078.\\n\\f THURAISSIGIAM v. USDHS 15\\n\\nwhether the provision restricting habeas review violates the\\nSuspension Clause.\\n\\nI. Jurisdiction Under 8 U.S.C. § 1252(e)(2)\\n\\n Thuraissigiam contends that 8 U.S.C. § 1252(e)(2)(B)\\nauthorizes review of the statutory, regulatory, and\\nconstitutional claims raised in his habeas petition. We\\ndisagree. Section 1252(e)(2), including Subsection (B), limits\\na district court to reviewing three basic factual determinations\\nrelated to an expedited removal order. Because\\nThuraissigiam’s petition does not challenge any of those\\ndeterminations, § 1252(e)(2) does not authorize jurisdiction\\nover the petition.\\n\\n A court applying habeas review under § 1252(e)(2) is\\nlimited to determining:\\n\\n (A) whether the petitioner is an alien,\\n\\n (B) whether the petitioner was ordered\\n removed under such section, and\\n\\n (C) whether the petitioner can prove by a\\n preponderance of the evidence that the\\n petitioner is an alien lawfully admitted for\\n permanent residence, has been admitted as a\\n refugee under section 1157 of this title, or has\\n been granted asylum under section 1158 of\\n this title . . . .\\n\\nCongress also provided express limitations on review under\\nSubsection (B):\\n\\f16 THURAISSIGIAM v. USDHS\\n\\n In determining whether an alien has been\\n ordered removed under section 1225(b)(1) of\\n this title, the court’s inquiry shall be limited to\\n whether such an order in fact was issued and\\n whether it relates to the petitioner. There\\n shall be no review of whether the alien is\\n actually inadmissible or entitled to any relief\\n from removal.\\n\\nId. § 1252(e)(5). Nonetheless, Thuraissigiam stakes his claim\\non Subsection (B).\\n\\n We considered and rejected a nearly identical argument\\nin Garcia de Rincon. The petitioner in Garcia de Rincon\\nraised a due process challenge to an expedited removal order.\\n539 F.3d at 1136. Characterizing § 1252(e) as among the\\n“most stringent” jurisdiction-limiting provisions in the\\nimmigration statutes, we held that § 1252(e)(2) permits\\nreview only of “habeas petitions alleging that the petitioner\\nis not an alien or was never subject to an expedited removal\\norder.” Id. at 1135, 1139. We therefore lacked jurisdiction\\nbecause the petitioner’s due process claims were not\\nencompassed by those enumerated grounds. Id. at 1139.\\nLikewise, Thuraissigiam’s claims of procedural violations are\\nplainly not claims about whether Thuraissigiam “was never\\nsubject to an expedited removal order.” 8 U.S.C. § 1252(e).\\n\\n Thuraissigiam contends that such a reading of Subsection\\n(B) renders superfluous the prohibition in § 1252(e)(5)\\nagainst “review of whether the alien is actually inadmissible\\nor entitled to any relief from removal.” However,\\n§ 1252(e)(5) functions not to repeat § 1252(e)(2)(B), but to\\nexplain it. Moreover, Thuraissigiam’s petition is barred by\\nthe first sentence in § 1252(e)(5), not the second sentence.\\n\\f THURAISSIGIAM v. USDHS 17\\n\\nBecause he asks the district court to review the government’s\\nprocedures, those claims are beyond the scope of “whether\\nsuch an [expedited removal] order in fact was issued and\\nwhether it relates to the petitioner.”5 8 U.S.C. § 1252(e); see\\nalso United States v. Barajas-Alvarado, 655 F.3d 1077, 1082\\n(9th Cir. 2011) (reaffirming that jurisdiction under\\n§ 1252(e)(2) “does not extend to review of the claim that an\\nalien was wrongfully deprived of the administrative review\\npermitted under the statute and applicable regulations”).\\n\\n Thuraissigiam relies on Smith v. U.S. Customs & Border\\nProtection, 741 F.3d 1016 (9th Cir. 2014), to contend that we\\nhave adopted a more expansive view of Subsection (B).\\nSpecifically, he contends that Smith reviewed “whether the\\npetitioner belonged in the expedited removal system,” and\\nthat a court may thus review his petition. Smith, however,\\ndoes not support Thuraissigiam’s argument. In Smith, CBP\\nplaced the petitioner, a Canadian citizen arriving at the\\nborder, in expedited removal proceedings for lacking certain\\ndocuments. Id. at 1019. The petitioner alleged that CBP\\nexceeded its authority under the expedited removal statute\\nbecause certain document requirements are waived for\\nCanadians, and argued that Subsection (B) authorized review.\\nId. at 1019, 1021. “Accepting [petitioner’s] theory at face\\nvalue,” we reviewed whether CBP in fact classified him as an\\n“intending immigrant.” Id. at 1021. Concluding that CBP had\\ndone so, we held that § 1252(e)(2) “permit[ted] us to go no\\n\\n\\n 5\\n Thuraissigiam also makes a structural argument, contending that\\nbecause Congress provided for some review of asylum claims even in\\nexpedited removal cases, Congress must not have intended to strip judicial\\nreview to “police the boundaries of those limits.” This argument ignores\\nthe plain language of the statute, which evidences Congress’ intent to do\\njust that.\\n\\f18 THURAISSIGIAM v. USDHS\\n\\nfurther” and did not discuss the merits of CBP’s\\nclassification. Id. at 1021–22 & n.4.6 Therefore, Smith\\nreviewed only how CBP classified the petitioner, which is\\nfairly encompassed by whether “[the petitioner] was ordered\\nremoved” under the expedited removal provision. Id. at 1022\\n(internal quotation marks omitted). By contrast,\\nThuraissigiam asks the district court to pass judgment on the\\nprocedures leading to his removal order. The limited review\\nprovided under § 1252(e)(2) does not encompass such\\nclaims.7\\n\\n Therefore, in line with our precedents, we conclude that\\n§ 1252(e) does not authorize habeas review of\\nThuraissigiam’s petition. We do not here address\\nThuraissigiam’s request that we apply the canon of\\nconstitutional avoidance to interpret § 1252(e) to provide\\njurisdiction over his legal claims. That canon only comes\\ninto play if we conclude that § 1252(e) raises serious\\nconstitutional questions; thus, we first address\\nThuraissigiam’s Suspension Clause argument before\\ncontemplating the application of that canon. See St. Cyr,\\n533 U.S. at 299–300, 314 (explaining constitutional\\navoidance canon and applying it upon concluding that the\\nstatute in question raised serious constitutional questions).\\n\\n\\n 6\\n The Smith petitioner also contended that § 1252(e)(2) violated the\\nSuspension Clause, but we did not reach the argument because we held\\nthat the statute permitted limited review of his petition. 741 F.3d at 1022\\nn.6.\\n 7\\n We have held that in appeals from convictions for criminal reentry,\\na defendant may collaterally attack a removal order that forms the basis\\nfor his conviction. See United States v. Ochoa-Oregel, 904 F.3d 682, 686\\n(9th Cir. 2018) (considering collateral attack on expedited removal order).\\nBut that rule does not apply to this case.\\n\\f THURAISSIGIAM v. USDHS 19\\n\\nII. Suspension Clause\\n\\n The Suspension Clause mandates, “The Privilege of the\\nWrit of Habeas Corpus shall not be suspended, unless when\\nin Cases of Rebellion or Invasion the public Safety may\\nrequire it.” U.S. Const. art. I, § 9, cl. 2. Our nation’s\\nfounders viewed the writ as a “vital instrument” to secure\\nindividual liberty. Boumediene, 553 U.S. at 743. “The\\nClause protects the rights of the detained by a means\\nconsistent with the essential design of the Constitution. It\\nensures that, except during periods of formal suspension, the\\nJudiciary will have a time-tested device, the writ, to maintain\\nthe ‘delicate balance of governance’ that is itself the surest\\nsafeguard of liberty.” Id. at 745 (citing Hamdi v. Rumsfeld,\\n542 U.S. 507, 536 (2004) (opinion of O’Connor, J.)). The\\nSuspension Clause prevents Congress from passing a statute\\nthat effectively suspends the writ absent rebellion or invasion.\\nSee Felker v. Turpin, 518 U.S. 651, 663–64 (1996). Thus, the\\nquestion in this case is whether 8 U.S.C. § 1252(e)(2)\\neffectively suspends the writ. Put another way, the question\\nis whether the habeas review available to Thuraissigiam\\nunder § 1252(e)(2) satisfies the requirements of the\\nSuspension Clause.\\n\\n The Supreme Court has not yet answered that question.\\nIn fact, the Court has rarely addressed who may invoke the\\nSuspension Clause and the extent of review the Clause\\nrequires. For example, only in Boumediene has the Court\\nconcluded that a statute violated the Suspension Clause.\\nGerald L. Neuman, The Habeas Corpus Suspension Clause\\nAfter Boumediene v. Bush, 110 Colum. L. Rev. 537, 538\\n(2010). In the Court’s other most recent Suspension Clause\\ncase, St. Cyr, the Court, after extensive analysis of the\\nSuspension Clause issues at play, interpreted the statute to\\n\\f20 THURAISSIGIAM v. USDHS\\n\\navoid those issues. 533 U.S. at 336–37. Of the federal courts\\nof appeals, only the Third Circuit has addressed the precise\\nquestion before us, whether § 1252(e)(2) as applied to\\nnoncitizen petitioners in expedited removal violates the\\nSuspension Clause. See Castro v. U.S. Dep’t of Homeland\\nSec., 835 F.3d 422 (3d Cir. 2016) (concluding that, due to\\ntheir status, such petitioners could not invoke the Suspension\\nClause), cert. denied, 137 S. Ct. 1581 (2017).\\n\\n Boumediene traced the writ of habeas corpus to its origins\\nas a tool of the English crown, citing the detailed historical\\naccount in Paul D. Halliday and G. Edward White, The\\nSuspension Clause: English Text, Imperial Contexts, and\\nAmerican Implications, 94 Va. L. Rev. 575 (2008). 553 U.S.\\nat 740. As Halliday and White explain, the writ in England\\nwas the vehicle “to determine the rightness of constraints\\nimposed on the bodies of the king’s subjects of all kinds.” 94\\nVa. L. Rev. at 607. The writ was on occasion suspended in\\nEngland. Id. at 619; see Boumediene, 553 U.S. at 741.\\nAccording to Boumediene, that history “no doubt confirmed\\n[the Framers’] view that pendular swings to and away from\\nindividual liberty were endemic to undivided, uncontrolled\\npower.” Id. at 742; see also Amanda L. Tyler, A “Second\\nMagna Carta”: The English Habeas Corpus Act and the\\nStatutory Origins of the Habeas Privilege, 91 Notre Dame L.\\nRev. 1949, 1985–86 (2016) (describing how suspensions of\\nthe writ in colonial America motivated the States’ desire to\\nimport similar habeas protections after gaining\\nindependence); Halliday & White, 94 Va. L. Rev. at 671\\n(highlighting the Framers’ desire to restore “the traditional\\norder of writs and suspensions”).\\n\\n As Boumediene summed it up, the Suspension Clause is\\nrooted in the Framers’ first-hand experience “that the\\n\\f THURAISSIGIAM v. USDHS 21\\n\\ncommon-law writ all too often had been insufficient to guard\\nagainst the abuse of monarchial power.” 553 U.S. at 739–40.\\nThe Clause, therefore, is “not merely about suspending the\\nprivilege of the writ of habeas corpus, but about the meaning\\nof the ‘privilege of the writ’ itself.” Halliday & White,\\n94 Va. L. Rev. at 699. “Indeed, common law habeas corpus\\nwas, above all, an adaptable remedy . . . [whose] precise\\napplication and scope changed depending upon the\\ncircumstances.” Boumediene, 553 U.S. at 779–80 (citing,\\ninter alia, Richard H. Fallon, Jr. & Daniel J. Meltzer, Habeas\\nCorpus Jurisdiction, Substantive Rights, and the War on\\nTerror, 120 Harv. L. Rev. 2029, 2102 (2007)).\\n\\n In examining how the Supreme Court has defined the\\nSuspension Clause’s requirements, Boumediene is our\\nstarting point, even if it does not provide a direct answer to\\nThuraissigiam’s challenge. Boumediene and its predecessors,\\nlike St. Cyr, do provide an analytical blueprint. We therefore\\nreview those precedents before deciding how best to apply\\ntheir principles to this appeal.\\n\\n A. Boumediene v. Bush\\n\\n In Boumediene, the Supreme Court struck down a War on\\nTerror-era law after detainees at the Guantanamo Bay prison\\nin Cuba brought a Suspension Clause challenge. 553 U.S. at\\n732–33. In the wake of the September 11, 2001, attacks, the\\nU.S. Department of Defense created Combatant Status\\nReview Tribunals (“CSRTs”) to decide if detainees were\\n“enemy combatants.” Id. at 733. The Boumediene\\npetitioners, who had all appeared before CSRTs and been\\ndeemed enemy combatants, sought a writ of habeas corpus\\nunder the general habeas statute, 28 U.S.C. § 2241. Id. at\\n734. After protracted litigation, Congress passed the\\n\\f22 THURAISSIGIAM v. USDHS\\n\\nDetainee Treatment Act of 2005 (“DTA”), which amended\\n§ 2241 to bar judicial review of habeas petitions filed by\\nGuantanamo detainees and to vest review of CSRT decisions\\nexclusively in the D.C. Circuit. Id. at 735 (citing DTA\\n§ 1005(e), 119 Stat. 2742). Section 7 of the Military\\nCommissions Act of 2006 (“MCA”) made those provisions\\nretroactive. Id. at 736. See generally Hamad v. Gates,\\n732 F.3d 990, 996–99 (9th Cir. 2013) (describing\\nBoumediene’s place in the line of Guantanamo detainee\\ncases). The Court took a two-step approach to evaluating the\\ndetainees’ challenge to the MCA.\\n\\n At step one, the Court evaluated whether the Guantanamo\\ndetainees – as enemy combatants detained on foreign soil –\\ncould even invoke the Suspension Clause. See Boumediene,\\n553 U.S. at 739. In so doing, the Court affirmed that\\nalthough the writ’s protections may have expanded since the\\nConstitution’s drafting, “at the absolute minimum,” the\\nClause protects the writ as it existed in 1789. Id. at 746\\n(citing St. Cyr, 533 U.S. at 301). The Court therefore\\nexamined historical authorities to determine the scope of the\\nwrit in 1789, and whether it ran to “an enemy alien detained\\nabroad.” Id. at 752. Although noting that “at common law a\\npetitioner’s status as an alien was not a categorical bar to\\nhabeas corpus,” the Court concluded that the historical record\\ndid not provide a definitive answer. Id. at 747, 752. Instead,\\nthe Court turned to its extraterritoriality precedents and from\\nthem concluded that “questions of extraterritoriality turn on\\nobjective factors and practical concerns, not formalism.” Id.\\nat 764. Boumediene drew from Johnson v. Eisentrager,\\n339 U.S. 763 (1950), another case about the extraterritorial\\napplication of the Suspension Clause, three non-exclusive\\nfactors relevant to the Clause’s extraterritorial scope:\\n\\f THURAISSIGIAM v. USDHS 23\\n\\n (1) the citizenship and status of the detainee\\n and the adequacy of the process through\\n which that status determination was made;\\n (2) the nature of the sites where apprehension\\n and then detention took place; and (3) the\\n practical obstacles inherent in resolving the\\n prisoner’s entitlement to the writ.\\n\\n553 U.S. at 766.8 Applying those factors, the Court\\nconcluded that the detainees could invoke the Suspension\\nClause. Id. at 771.\\n\\n At step two, the Court considered whether Congress had\\nsuspended the writ without an adequate substitute. The Court\\nacknowledged that there are “few precedents addressing what\\nfeatures an adequate substitute for habeas corpus must\\ncontain.” Id. at 772. For example, the Court had previously\\nupheld provisions of the Antiterrorism and Effective Death\\nPenalty Act of 1996 (“AEDPA”) against a Suspension Clause\\nchallenge because the provisions “did not constitute a\\nsubstantial departure from common-law habeas procedures.”\\nId. at 774 (citing Felker, 518 U.S. at 664); see also Neuman,\\n110 Colum. L. Rev. at 542 (stating that “what matters is the\\nsubstance, not the form, of the Great Writ,” and that\\n“Congress can rename or reconfigure the procedure by which\\ncourts examine the lawfulness of detention,” as long as the\\nsubstitute is adequate).\\n\\n 8\\n In Rasul v. Bush, 542 U.S. 466 (2004), the Court had first discussed\\nEisentrager’s applicability to the question of who may invoke the\\nSuspension Clause. Id. at 487 (Kennedy, J., concurring) (“A faithful\\napplication of Eisentrager, then, requires an initial inquiry into the general\\ncircumstances of the detention to determine whether the Court has the\\nauthority to entertain the petition and to grant relief after considering all\\nof the facts presented.”).\\n\\f24 THURAISSIGIAM v. USDHS\\n\\n In Boumediene, the Court gleaned from its precedents two\\n“easily identified attributes of any constitutionally adequate\\nhabeas corpus proceeding.” 553 U.S. at 779. First, the\\n“privilege of habeas corpus entitles the prisoner to a\\nmeaningful opportunity to demonstrate that he is being held\\npursuant to ‘the erroneous application or interpretation’ of\\nrelevant law.” Id. (quoting St. Cyr, 533 U.S. at 302). Second,\\n“the habeas court must have the power to order the\\nconditional release of an individual unlawfully detained.” Id.\\nBeyond those minimum requirements, “depending on the\\ncircumstances, more may be required.” Id.\\n\\n The Court further emphasized that “the necessary scope\\nof habeas review in part depends upon the rigor of any earlier\\nproceedings.” Id. at 781; see also id. at 786 (noting that\\n“habeas corpus review may be more circumscribed if the\\nunderlying detention proceedings are more thorough”). For\\nthat reason, courts sitting in habeas afford deference when\\nreviewing another court’s decision, but when a petitioner is\\n“detained by executive order . . . the need for collateral\\nreview is most pressing.” Id. at 783. To be effective, the\\n“habeas court must have sufficient authority to conduct a\\nmeaningful review of both the cause for detention and the\\nExecutive’s power to detain.” Id. Applying those principles\\nto the CSRTs and D.C. Circuit review, the Court concluded\\nthat the MCA did not provide an adequate substitute because\\nthe D.C. Circuit could not “consider newly discovered\\nevidence that could not have been made part of the CSRT\\nrecord.” Id. at 790. The Court then concluded that it was not\\npossible to read into the statute provisions for the procedures\\nnecessary to satisfy the Suspension Clause, and therefore held\\nit unconstitutional. Id. at 792.\\n\\f THURAISSIGIAM v. USDHS 25\\n\\n Boumediene provides an analytical template for\\nevaluating a Suspension Clause challenge: at step one, we\\nexamine whether the Suspension Clause applies to the\\npetitioner; and, if so, at step two, we examine whether the\\nsubstitute procedure provides review that satisfies the Clause.\\nHow more specifically to apply that template is less clear,\\ngiven that the Court generated its three-factor test at step one\\nin light of the extraterritoriality question in Boumediene. See\\nid. at 764, 766. Those factors, as both parties acknowledge,\\ndo not map precisely onto this case because Thuraissigiam\\nwas apprehended and is detained on U.S. soil.9 Yet, the\\nmanner in which the Court divined those factors informs our\\napproach here. Boumediene relied on Eisentrager and related\\ncases, but also looked to 1789-era application of the writ to\\ndetermine whether petitioners similarly situated to\\nGuantanamo detainees had been able to invoke the Clause.\\nAlthough the Court emphasized that the history was not\\ndispositive, it made clear that “settled precedents or legal\\ncommentaries in 1789 . . . can be instructive.” Id. at 739.10\\n\\n\\n\\n\\n 9\\n We too have applied the three Boumediene factors more readily\\nwhen asking whether a noncitizen outside the United States – again,\\nunlike Thuraissigiam – can claim the Constitution’s protections. See\\nIbrahim v. Dep’t of Homeland Sec., 669 F.3d 983, 995 (9th Cir. 2012)\\n(evaluating extraterritoriality in context of First and Fifth Amendment\\nclaims).\\n 10\\n Indeed, the Supreme Court’s proposition that the Suspension Clause\\nat least protects the writ as it existed in 1789 “necessarily invites reference\\nto history when interpreting and applying the Suspension Clause.”\\nAmanda L. Tyler, Habeas Corpus in Wartime 9 (2017); see also Omar v.\\nMcHugh, 646 F.3d 13, 19 (D.C. Cir. 2011) (Kavanaugh, J.) (“[H]istory\\nmatters: In habeas cases, we seek guidance from history ‘addressing the\\nspecific question before us.’” (quoting Boumediene, 553 U.S. at 746)).\\n\\f26 THURAISSIGIAM v. USDHS\\n\\n At step two, Boumediene held that, at a minimum, the\\nSuspension Clause entitles a petitioner “to a meaningful\\nopportunity to demonstrate he is being held to ‘the erroneous\\napplication or interpretation’ of relevant law.” Id. at 779\\n(quoting St. Cyr, 533 U.S. at 302). In considering whether\\nthe Clause required more in the circumstances of\\nBoumediene, the Court impliedly considered the rigor and\\ncharacter of the proceedings preceding habeas review. Also\\nrelevant to Thuraissigiam’s case, the Court affirmed that the\\nSuspension Clause protects “a right of first importance,” even\\nin circumstances – such as national security, in Boumediene\\n– where the executive’s power is at its zenith. Id. at 797–98.\\n\\n B. INS v. St. Cyr\\n\\n St. Cyr, which predated Boumediene by several years,\\nsheds additional light on the Court’s approach to Suspension\\nClause questions. The petitioner, St. Cyr, was a lawful\\npermanent resident admitted to the United States in 1986.\\n533 U.S. at 293. In 1996, St. Cyr pleaded guilty to a criminal\\ncharge that made him removable, although under pre-AEDPA\\nlaw (applicable at the time of his conviction), he was eligible\\nfor a discretionary waiver from the Attorney General. Id.\\nAfter AEDPA was passed, the government began removal\\nproceedings, with the Attorney General interpreting AEDPA\\nto have removed his discretion to grant St. Cyr a waiver. Id.\\nSt. Cyr filed a habeas petition alleging that the Attorney\\nGeneral’s interpretation was erroneous because St. Cyr’s\\nconviction predated AEDPA. Id. After the district court and\\nSecond Circuit agreed with St. Cyr, the government argued to\\n\\f THURAISSIGIAM v. USDHS 27\\n\\nthe Supreme Court that the courts lacked jurisdiction to\\nreview the Attorney General’s interpretation. Id. at 297–98.\\n\\n The Court stated that the government’s position had to\\novercome several presumptions, chief among them the\\n“strong presumption in favor of judicial review of\\nadministrative action and the longstanding rule requiring a\\nclear statement of congressional intent to repeal habeas\\njurisdiction.” Id. at 298 (citing Ex parte Yerger, 75 U.S.\\n(8 Wall.) 85, 102 (1869)). To address whether the statute\\nraised serious Suspension Clause questions, the Court started\\nfrom the principle that “[b]ecause of [the] Clause, some\\n‘judicial intervention in deportation cases’ is unquestionably\\n‘required by the Constitution.’” Id. at 300 (citing Heikkila v.\\nBarber, 345 U.S. 229, 235 (1953)). Because “at the absolute\\nminimum, the Suspension Clause protects the writ ‘as it\\nexisted in 1789,’” the Court looked at the writ’s application\\nbefore and after the drafting of the Constitution. Id. at 301\\n(quoting Felker, 518 U.S. at 663–64). Legal and historical\\nauthorities indicated that in both England and the United\\nStates “the writ of habeas corpus has served as a means of\\nreviewing the legality of Executive detention, and it is in that\\ncontext that its protections have been strongest.” Id.\\nMoreover, the writ was available both to “nonenemy aliens as\\nwell as citizens” and “encompassed detentions based on\\nerrors of law, including the erroneous application or\\ninterpretation of statutes.” Id. at 301–02.\\n\\f28 THURAISSIGIAM v. USDHS\\n\\n St. Cyr also looked to the so-called “finality era,”11 during\\nwhich the statutory scheme precluded judicial intervention in\\nimmigration enforcement, except as required by the\\nConstitution. Id. at 304–06. Despite that statutory bar, the\\nCourt in the finality era “allow[ed] for review on habeas of\\nquestions of law.” Id. at 304. Accordingly, the government’s\\nreading of the statute – to prohibit any judicial review of the\\nAttorney General’s interpretation – raised “Suspension\\nClause questions that . . . are difficult and significant.” Id.\\nMore directly, “to conclude that the writ is no longer\\navailable in this context would represent a departure from\\nhistorical practice in immigration law.” Id. at 305. After\\ncanvassing that historical practice, and noting that it was\\nconsistent with the writ’s “common-law antecedents,” the\\nCourt concluded that St. Cyr could have brought his habeas\\nclaims under that regime. Id. at 308. Thus, due to the serious\\nconstitutional questions raised, and because Congress had not\\nprovided a “clear, unambiguous, and express” intent to\\npreclude habeas jurisdiction over questions of law, the Court\\nconcluded that the statutes at issue did not repeal habeas\\njurisdiction. Id. at 314.\\n\\n St. Cyr further illuminates how to approach both\\nBoumediene steps. Like Boumediene, St. Cyr looked to the\\n1789-era historical application of the writ. St. Cyr also\\nlooked to the finality era because it provides evidence of what\\n\\n 11\\n The “finality era” refers to “an approximately sixty-year period\\nwhen federal immigration law rendered final (hence, the ‘finality’ era) the\\nExecutive’s decisions to admit, exclude, or deport noncitizens. This\\nperiod began with the passage of the Immigration Act of 1891, ch. 551,\\n26 Stat. 1084, and concluded when Congress enacted the Immigration and\\nNationality Act of 1952, Pub. L. No. 82-414, 66 Stat. 163, which\\npermitted judicial review of deportation orders through declaratory\\njudgment actions in federal district courts.” Castro, 835 F.3d at 436.\\n\\f THURAISSIGIAM v. USDHS 29\\n\\ndegree of habeas review is required under the Suspension\\nClause and to whom such review is guaranteed in the\\nimmigration enforcement context. St. Cyr’s resort to prior\\nhabeas cases aligns with Boumediene’s similar reliance on\\nEisentrager to resolve ambiguities in the 1789-era application\\nof the writ. That St. Cyr ultimately avoided the Suspension\\nClause question does not diminish its wisdom or relevance as\\nan example of the Court’s analytical approach to Suspension\\nClause questions. Consistent with Boumediene and St. Cyr,\\nwe conclude that both the common-law history of the writ\\nand the Court’s finality era cases are relevant to what and\\nwhom the Suspension Clause protects. See also Flores-\\nMiramontes v. INS, 212 F.3d 1133, 1141–43 (9th Cir. 2000)\\n(relying on common-law history and finality era cases in\\naddressing Suspension Clause challenge); see also Trinidad\\ny Garcia v. Thomas, 683 F.3d 952, 960 (9th Cir. 2012) (en\\nbanc) (Thomas, J., concurring) (discussing finality era cases\\nas evidence of rights protected by the Suspension Clause).\\n\\n C. The Third Circuit’s Decision in Castro\\n\\n Before addressing Thuraissigiam’s Suspension Clause\\nchallenge, we discuss the Third Circuit’s decision in Castro,\\nwhich involved an analogous challenge to § 1252(e).12 The\\n\\n\\n 12\\n The government’s contention that the Second and Seventh Circuits\\nhave addressed the question before us is incorrect. Neither case addresses\\nthe Suspension Clause. See Shunaula v. Holder, 732 F.3d 143, 147 (2d\\nCir. 2013) (addressing due process challenge to § 1252(e)(2) and\\n§ 1252(a)(2)(A)); Khan v. Holder, 608 F.3d 325, 328 (7th Cir. 2010)\\n(addressing § 1252(e)(2) in light of that circuit’s “safety valve” doctrine\\nfor “judicial correction of bizarre miscarriages of justice”). Likewise, the\\ncase cited in the government’s Rule 28(j) letter, Hamama v. Adducci,\\n912 F.3d 869 (6th Cir. 2018), does not address the Suspension Clause in\\nthe context of the procedures leading up to an expedited removal order.\\n\\f30 THURAISSIGIAM v. USDHS\\n\\nThird Circuit concluded that § 1252(e) does not violate the\\nSuspension Clause as applied to 28 asylum-seeking families\\nwho, like Thuraissigiam, raised constitutional, statutory, and\\nregulatory claims relating to their negative credible fear\\ndeterminations. 835 F.3d at 425, 428. The families were all\\napprehended shortly after entering the country, placed in\\nexpedited removal, and found not to have credible fear. Id.\\nat 427–28. As we do, the Third Circuit rejected the argument\\nthat § 1252(e)(2) provides jurisdiction over claims of legal\\nerror. Id. at 434.\\n\\n Turning to the petitioners’ Suspension Clause challenge,\\nthe court opined that the Supreme Court’s habeas cases are\\n“perhaps even competing” with the plenary power doctrine.\\nId. After reviewing Boumediene and St. Cyr, Castro\\ndiscussed the Court’s “commitment to the full breadth of\\n[that] doctrine, at least as to aliens at the border seeking\\ninitial admission to the country.” Id. at 443. Castro\\napproached step one of Boumediene by reference to the\\npetitioners’ status in light of Landon v. Plascencia, 459 U.S.\\n21 (1982), a case addressing due process, not habeas, rights.\\nCastro concluded that petitioners, as “recent surreptitious\\nentrants,” should be treated for constitutional purposes as\\n“alien[s] seeking initial admission to the United States.”\\n835 F.3d at 448. In Landon, the Court stated that such a\\nnoncitizen “has no constitutional rights regarding his\\napplication” for entry into the country. 459 U.S. at 32.\\nAccordingly, the Third Circuit concluded that the petitioners’\\nchallenge failed at step one, and did not address whether\\n§ 1252(e) was an adequate habeas substitute. 835 F.3d at\\n446. The court acknowledged that its discussion of the\\npetitioners’ status “appear[ed] to ignore” Supreme Court\\nprecedent relating to the due process rights of noncitizens\\nphysically present in the country, but concluded that no case\\n\\f THURAISSIGIAM v. USDHS 31\\n\\nhad clearly held that “arriving aliens” were entitled to due\\nprocess protections. Id. at 447–48.13\\n\\n We disagree with Castro’s resolution of how Boumediene\\nand St. Cyr require us to approach a Suspension Clause\\nchallenge. As explained at length above, the Court’s mode of\\nanalysis in both of those cases addressed the scope of the\\nSuspension Clause by reference to the writ as it stood in 1789\\nand relevant habeas corpus precedents. Castro explained that\\nit did not rely on St. Cyr’s description of the Court’s habeas\\napproach in immigration cases in the finality era by\\nemphasizing that, unlike the Castro petitioners, St. Cyr was\\na lawful permanent resident, and that St. Cyr discussed only\\nwhat the Suspension Clause might protect. Id. at 446.\\n\\n That St. Cyr did not affirmatively hold that the\\nSuspension Clause was violated does not render its\\ndescription of the finality era cases incorrect or its approach\\nirrelevant. Moreover, Castro’s decision to rely instead on\\nLandon is misplaced. Landon held that a permanent resident\\nwho traveled abroad and was detained when attempting to\\nreenter the United States should be placed in exclusion\\n\\n\\n\\n\\n 13\\n After argument, the Third Circuit decided Osorio-Martinez v.\\nAttorney General, 893 F.3d 153 (3d Cir. 2018), involving four juvenile\\npetitioners from Castro. After their original habeas petitions were\\ndismissed, the juveniles had been granted Special Immigrant Juvenile\\n(“SIJ”) status under 8 U.S.C. § 1108(a)(27)(J). Id. at 160. Applying\\nCastro, the Third Circuit held that § 1252(e) was an unconstitutional\\nsuspension of the writ as applied to the petitioners, by virtue of their\\n“significant ties to this country” and the constitutional and statutory rights\\nflowing to SIJ designees under 8 U.S.C. § 1255(a) & (h)(1). Id. at 167.\\n\\f32 THURAISSIGIAM v. USDHS\\n\\nproceedings rather than deportation. 459 U.S. at 22.14\\nAddressing the petitioner’s due process challenge to her\\nexclusion proceedings, the Court noted it had “long held that\\nan alien seeking initial admission to the United States\\nrequests a privilege and has no constitutional rights regarding\\nhis application.” Id. As explained by Judge Hardiman, the\\nCourt in Landon did not “purport to resolve a jurisdictional\\nquestion raising the possibility of an unconstitutional\\nsuspension of the writ of habeas corpus”; rather it addressed\\nonly the due process rights of a permanent resident. Castro,\\n835 F.3d at 450 (Hardiman, J., concurring dubitante); see\\nLandon, 459 U.S. at 32–35. Landon could not and did not\\naddress the much different question of whether a petitioner\\nlike Thuraissigiam may invoke the Suspension Clause.15\\n\\n Although often conflated, the rights protected by the\\nSuspension Clause are not identical to those under the Fifth\\nAmendment’s guarantee of due process. See Lee Kovarsky,\\n\\n\\n 14\\n At the time, Congress provided that removable noncitizens in the\\nUnited States were subject to deportation and those seeking initial entry\\nwere subject to exclusion. Id. at 25. Now all noncitizens are subject to\\nremoval, whether via 8 U.S.C. § 1225(b)(1) expedited removal or the\\nremoval procedures under 8 U.S.C. § 1229a.\\n 15\\n Regardless, we disagree with the government’s contention and\\nCastro’s conclusion that a person like Thuraissigiam lacks all procedural\\ndue process rights. See 835 F.3d at 447–48. The Supreme Court has been\\nclear that presence matters to due process. See, e.g., Mathews v. Diaz,\\n426 U.S. 67, 77 (1976); Zadvydas v. Davis, 533 U.S. 678, 693 (2001).\\nAnd we have held that a noncitizen situated almost exactly like\\nThuraissigiam had a constitutional right “to expedited removal\\nproceedings that conformed to the dictates of due process.” United States\\nv. Raya-Vaca, 771 F.3d 1195, 1203 (9th Cir. 2014); see also Immigration\\nScholars Amicus Br. (explaining why Thuraissigiam has procedural due\\nprocess rights).\\n\\f THURAISSIGIAM v. USDHS 33\\n\\nCustodial and Collateral Process: A Response to Professor\\nGarrett, 98 Cornell L. Rev. Online 1, 1 (2013) (“Due process\\nand the habeas privilege are distinct constitutional\\nphenomena, [but] federal courts almost pathologically\\nconfuse them.”). It is true that, historically, the Fifth\\nAmendment’s due process guarantee and the Suspension\\nClause have been applied in tandem, as their applicability was\\nrarely disputed. See Mary Van Houten, The Post-\\nBoumediene Paradox: Habeas Corpus or Due Process?,\\n67 Stan. L. Rev. Online 9, 10 (2014) (observing that these\\nprovisions “were almost always jointly applied before\\nBoumediene”). But this fact does not mean these rights\\nshould be elided, as made clear by the fact that the\\nConstitution, ratified two-and-a-half years before the Fifth\\nAmendment, see Bute v. People of State of Ill., 333 U.S. 640,\\n650 (1948), presupposed the existence of the writ of habeas\\ncorpus, see Boumediene, 553 U.S. at 739 (“Protection for the\\nprivilege of habeas corpus was one of the few safeguards of\\nliberty specified in a Constitution that, at the outset, had no\\nBill of Rights.”). Indeed, the writ “is almost the only remedy\\nmentioned in the Constitution” as originally ratified. Fallon\\n& Meltzer, 120 Harv. L. Rev. at 2037.\\n\\n Boumediene itself clearly recognized the distinction\\nbetween the Fifth Amendment’s due process rights and the\\nSuspension Clause – providing further reason not to treat\\nLandon’s discussion of due process rights as having any\\nbearing on the application of the Suspension Clause. In\\nBoumediene, the Court decided that the Guantanamo\\ndetainees could invoke the Suspension Clause without\\naddressing whether they had due process rights or whether\\nthe CSRTs satisfied due process. 553 U.S. at 785; see also id.\\nat 739 (starting from the proposition that “protection for the\\nprivilege of habeas corpus was one of the few safeguards of\\n\\f34 THURAISSIGIAM v. USDHS\\n\\nliberty specified in a Constitution that, at the outset, had no\\nBill of Rights[]”); Flores-Miramontes, 212 F.3d at 1142\\n(noting that habeas was available at common law prior to the\\ndrafting of the Constitution). The Court in Boumediene\\ntherefore explicitly declined to link due process rights and\\nSuspension Clause rights. See Hamad, 732 F.3d at 999\\n(noting that Boumediene did not address whether the due\\nprocess clause applied to the Guantanamo detainees); see also\\nKiyemba v. Obama, 555 F.3d 1022, 1027 (D.C. Cir. 2009)\\n(concluding on habeas review that Guantanamo detainees\\nlacked due process rights), vacated by 559 U.S. 131 (2010),\\nreinstated by 605 F.3d 1046, 1047 (D.C. Cir. 2010). Landon,\\na due process case, is not relevant to whether Thuraissigiam\\ncan invoke the Suspension Clause. For that reason, we\\ndecline to follow Castro’s approach and reject the\\ngovernment’s argument that Thuraissigiam’s purported lack\\nof due process rights is determinative of whether he can\\ninvoke the Suspension Clause.\\n\\n Instead, in accordance with Boumediene, we evaluate\\nThuraissigiam’s Suspension Clause challenge in two steps:\\nFirst, to determine whether Thuraissigiam may invoke the\\nSuspension Clause, we examine 1789-era practice, the\\nfinality era cases, and other relevant precedents. Second, we\\nask whether § 1252(e)(2) provides Thuraissigiam a\\n“meaningful opportunity to demonstrate that he is being held\\npursuant to ‘the erroneous application or interpretation’ of\\nrelevant law.” Boumediene, 553 U.S. at 779. At step two, we\\nkeep in mind that the character of the earlier proceedings\\nbears on the level of habeas review required. Id. at 781.\\n\\f THURAISSIGIAM v. USDHS 35\\n\\nIII. Application\\n\\n A. Garcia de Rincon and Pena\\n\\n At the outset, the government contends that our decisions\\nin Garcia de Rincon, 539 F.3d 1133, and Pena v. Lynch,\\n815 F.3d 452 (9th Cir. 2016), require us to affirm. Although\\nin both cases we rejected arguments that § 1252(e)(2)\\nauthorized jurisdiction, neither case answered the\\nconstitutional question before us today.\\n\\n In Garcia de Rincon, the petitioner was a noncitizen\\nliving in the United States who was stopped at the border\\nattempting to reenter after a visit to Mexico, and placed in\\nexpedited removal. 539 F.3d at 1135. After rejecting the\\npetitioner’s statutory challenge, we dismissed her argument\\n– “although . . . not articulated” as such – that the Suspension\\nClause required review of her petition. Id. at 1141. The\\nprecise question considered was whether “the INA provides\\nno adequate substitute for habeas review and therefore\\nsuspends the writ” – a Boumediene step two question,\\nalthough Garcia de Rincon never addressed Boumediene,\\nwhich had been decided months earlier. Id. We concluded\\nthat Li v. Eddy, 259 F.3d 1132 (9th Cir. 2001), vacated on\\nreh’g as moot, 324 F.3d 1109 (9th Cir. 2003), discredited the\\npetitioner’s “generalized due process argument,” the only\\nright she sought to vindicate via her petition. Id. Garcia de\\nRincon says nothing about whether Thuraissigiam can invoke\\nthe Suspension Clause, whether the Clause requires habeas\\nreview of statutory or legal claims, or what the Clause\\nrequires for a petitioner like Thuraissigiam who is within the\\nUnited States. Instead, the case addressed only whether\\n§ 1252(e)(2) suspends the writ when a petitioner lacks due\\nprocess rights. Put in Boumediene step-two terms, the due\\n\\f36 THURAISSIGIAM v. USDHS\\n\\nprocess clause was not “relevant law” for the Garcia de\\nRincon petitioner.16\\n\\n Pena also did not settle the question before us. In Pena,\\na noncitizen placed in expedited removal filed a petition for\\nreview of the Board of Immigration Appeals’ dismissal of his\\nappeal from an IJ’s decision affirming a negative credible\\nfear determination. 815 F.3d at 454. Because the petition\\nwas not brought under § 1252(e)(2), we concluded that we\\nlacked jurisdiction. Id. at 457. We went on to note that in\\nWebster v. Doe, 486 U.S. 592, 603 (1988), the Court had\\n“suggested that a litigant may be unconstitutionally denied a\\nforum when there is absolutely no avenue for judicial review\\nof a colorable claim of constitutional deprivation.” 815 F.3d\\nat 456 (emphasis in original). We concluded that Pena’s\\npetition did not raise Webster concerns because he lacked a\\ncolorable constitutional claim,17 and further noted that\\n§ 1252(e)(2) provides “some avenues of judicial review.” Id.\\nat 456–57. All that Pena says, therefore, is that § 1252(e)\\ndoes not implicate the Webster doctrine when a petitioner\\nfails to raise colorable constitutional claims. Pena never\\naddressed the Suspension Clause.\\n\\n Because neither Garcia de Rincon nor Pena addressed\\nwhether § 1252(e)(2) unlawfully suspends the writ as applied\\n\\n\\n 16\\n In case there were any doubt, Smith subsequently reserved the\\nquestion of whether, as applied to a noncitizen in expedited removal, the\\nSuspension Clause requires review beyond that provided for in\\n§ 1252(e)(2). See 741 F.3d at 1022 n.6. That reservation necessarily\\ndetermined that Garcia de Rincon had not settled the question.\\n 17\\n Pena claimed that the IJ violated due process by failing to elicit a\\nvoluntary waiver of his right to counsel, but we noted that that claim was\\ncontradicted by the record. Id. at 455–56.\\n\\f THURAISSIGIAM v. USDHS 37\\n\\nto a petitioner like Thuraissigiam, we reject the government’s\\nargument that those cases alone require us to affirm.\\n\\n B. Reach of the Suspension Clause\\n\\n At Boumediene step one, we must consider the reach of\\nthe Suspension Clause, or, in other words, whether\\nThuraissigiam is “barred from seeking the writ or invoking\\nthe protections of the Suspension Clause . . . because of [his]\\nstatus . . . .” Boumediene, 553 U.S. at 739. In Boumediene,\\nthe Court answered this question by reference to its\\nprecedents and the common law history of the writ. We\\ntherefore do the same.18\\n\\n\\n 18\\n As described above, the Court in Boumediene generated a three-\\nfactor test at step one in light of the extraterritoriality question presented.\\nThis test does not clearly fit in the present case, given that Thuraissigiam\\nwas apprehended and detained in the United States. See 553 U.S. at 764,\\n766. However, even were we to apply Boumediene’s three-factor test\\nhere, it would, as in Boumediene, support application of the Suspension\\nClause.\\n\\n The first factor, Thuraissigiam’s “citizenship and status” and “the\\nadequacy of the process through which that status determination was\\nmade,” Id. at 766, weighs in favor of applying the Suspension Clause.\\nThuraissigiam is a foreign national who contests his status – he contends\\nthat he has a credible fear of persecution and therefore qualifies as a\\nrefugee entitled to asylum. Like the CRST process at issue in\\nBoumediene, the determination as to whether a noncitizen has a credible\\nfear is not made via a “rigorous adversarial process to test the legality of\\n[his] detention.” Id. at 767. The determination is made by an asylum\\nofficer, 8 C.F.R. § 208.30(d), and although the noncitizen may consult\\nothers and even have them present a statement at the end of the interview,\\n8 C.F.R. § 208.30(d)(4), other hallmarks of the adversarial process are\\nlacking. If the noncitizen then chooses to contest an asylum officer’s\\nnegative credible fear determination, the noncitizen is entitled only to\\ncursory review by an IJ. 8 C.F.R. § 208.30(g); 8 C.F.R. § 1208.39(g)(2).\\n\\f38 THURAISSIGIAM v. USDHS\\n\\n As explained, in St. Cyr, the Court canvassed cases from\\nEngland and historical accounts to conclude that the writ was\\navailable before 1789 to “nonenemy aliens as well as to\\ncitizens.” 533 U.S. at 301; accord Boumediene, 553 U.S. at\\n747 (“We know that at common law a petitioner’s status as an\\n\\n\\nCritically, unlike in Boumediene, a noncitizen cannot seek review of the\\ncredible fear determination in an Article III court. See 8 U.S.C.\\n§ 1252(e)(2); cf. Boumediene, 553 U.S. at 767 (“And although the\\ndetainee can seek review of his status determination in the Court of\\nAppeals, that review process cannot cure all defects in the earlier\\nproceedings.”). Accordingly, the procedural protections available to\\nThuraissigiam and other noncitizens in expedited removal “fall well short\\nof the procedures and adversarial mechanisms that would eliminate the\\nneed for habeas corpus review.” Boumediene, 553 U.S. at 767 (internal\\nquotation marks omitted).\\n\\n As to the second factor, there is no question that Thuraissigiam was\\napprehended and detained within the sovereign territory of the United\\nStates. This factor weighs strongly in favor of finding Thuraissigiam has\\nrights under the Suspension Clause. See id. at 768–69. The government\\ninsists that the nature and length of a noncitizen’s detention is relevant to\\nthis factor. Not so. Boumediene only invokes these considerations under\\nstep two. The second factor (under step one) is wholly focused on the\\nlevel and duration of control exerted by the United States over the territory\\n– which is not at issue here, where the territory is the United States.\\nBoumediene, 553 U.S. at 768–69.\\n\\n As in Boumediene, the third factor is somewhat equivocal: “there are\\ncosts to holding the Suspension Clause applicable in a case of [asylum-\\nseekers in expedited removal proceedings.]” Id. at 769. But\\n“[c]ompliance with any judicial process requires some incremental\\nexpenditure of resources,” and direct review by the courts already exists\\nand functions in non-expedited removal proceedings. Id. Thus, here, as\\nin Boumediene, “[w]hile we are sensitive to [the government’s] concerns,\\nwe do not find them dispositive.” Id.\\n\\n Consequently, Boumediene’s extraterritorial step one factors, if they\\nwere relevant here, would support application of the Suspension Clause.\\n\\f THURAISSIGIAM v. USDHS 39\\n\\nalien was not a categorical bar to habeas corpus relief.”);\\nRasul, 542 U.S. at 481 (“At common law, courts exercised\\nhabeas jurisdiction over the claims of aliens detained within\\nsovereign territory of the realm. . . .”); see also Gerald L.\\nNeuman, Habeas Corpus, Executive Detention, and the\\nRemoval of Aliens, 98 Colum. L. Rev. 961, 989–90 (1998)\\n(collecting cases).\\n\\n After the adoption of the Constitution and its Suspension\\nClause, courts in the United States applied the same\\napproach. For example, in Ex parte D’Olivera, 7 F. Cas. 853\\n(C.C.D. Mass. 1813) (No. 3,967), a federal court in\\nMassachusetts permitted an arrested noncitizen seaman to\\ninvoke habeas. In later years, the Supreme Court continued\\nto hold that habeas was available to noncitizens – even\\nexcluded noncitizens stopped at the border. United States v.\\nJung Ah Lung, 124 U.S. 621, 628–32 (1888); see also\\nNeuman, 98 Colum. L. Rev. at 1006. Cases throughout the\\nfinality era, from the 1890s to the 1950s, which carry\\nsignificant weight here, held firm to this constitutional\\npremise. In Nishimura Ekiu v. United States, 142 U.S. 651\\n(1892), the Court affirmed that despite the finality law, “[a]n\\nalien immigrant, prevented from landing by any such officer\\nclaiming authority to do so under an act of congress, and\\nthereby restrained of his liberty, is doubtless entitled to a writ\\nof habeas corpus to ascertain whether the restraint is lawful.”\\nId. at 660.\\n\\n The Court continued that approach later in the finality era.\\nGegiow v. Uhl, 239 U.S. 3, 9 (1915) (“The courts are not\\nforbidden by the [finality] statute to consider whether the\\nreasons, when they are given, agree with the requirements of\\nthe act.”). In Gegiow, for example, the Court reversed the\\ngovernment’s legal conclusion that the petitioner was subject\\n\\f40 THURAISSIGIAM v. USDHS\\n\\nto exclusion as a public charge based on a lack of labor\\nopportunities in his immediate destination. Id. at 9–10; see\\nalso Shaughnessy v. United States ex rel. Mezei, 345 U.S.\\n206, 212–13 (1953) (stating that even though a noncitizen\\nwho had not entered the country lacks due process, he “may\\nby habeas corpus test the validity of his exclusion”); United\\nStates ex rel. Knauff v. Shaughnessy, 338 U.S. 537, 544\\n(1950) (addressing, but rejecting, noncitizen’s “contention\\nthat the regulations were not ‘reasonable’ as they were\\nrequired to be [under a federal statute]”). In United States ex\\nrel. Accardi v. Shaughnessy, the Court granted habeas on\\nlegal grounds to a noncitizen who entered from Canada\\n“without immigration inspection and without an immigration\\nvisa.” 347 U.S. 260, 262, 268 (1954). In Heikkila, the Court\\nexplained that the Constitution was the source for habeas\\nreview during the finality era, 345 U.S. at 234–35, and in St.\\nCyr, the Court clarified that the Suspension Clause was the\\nspecific source of such review. See 533 U.S. at 304. Indeed,\\nthe government points to no alternative reading.\\n\\n More broadly, the government offers no convincing\\nreason to discount the finality era, nor does it offer a\\ncompeting account of the common-law scope of the writ or of\\nthe finality era. The government, citing Mezei, 345 U.S. at\\n214, answers Boumediene step one by contending that\\nThuraissigiam, “as an alien apprehended immediately after\\ncrossing the border illegally, is no different from other aliens\\nat the border, and is therefore ‘assimilated to [that] status’ for\\nconstitutional purposes.” However, Mezei spoke only of such\\nassimilation for the purposes of due process, and it otherwise\\naffirmed the principle that habeas is available even when a\\npetitioner lacks due process rights. Id. at 213. And, crucially,\\nBoumediene never linked Suspension Clause rights to due\\nprocess rights. The government provides no authority from\\n\\f THURAISSIGIAM v. USDHS 41\\n\\nSuspension Clause cases to support its contention that\\nThuraissigiam lacks Suspension Clause rights. Because in\\nthe finality era the Court permitted even arriving noncitizens\\nto invoke habeas review, we conclude that Thuraissigiam,\\nwho was arrested within the United States, may invoke the\\nSuspension Clause.19\\n\\n C. Compliance with the Suspension Clause\\n\\n Having concluded that Thuraissigiam may invoke the\\nSuspension Clause, we must consider at Boumediene step two\\nwhether habeas review under § 1252(e) is so limited so as\\neffectively to suspend the writ as applied to Thuraissigiam.\\nAt a minimum, the Suspension Clause “entitles the\\n[petitioner] to a meaningful opportunity to demonstrate that\\nhe is being held pursuant to ‘the erroneous application or\\ninterpretation of relevant law.’” Boumediene, 553 U.S. at 779\\n(quoting St. Cyr, 553 U.S. at 302).\\n\\n Congress may modify the scope of habeas review so long\\nas the review “is neither inadequate nor ineffective to test the\\nlegality of a person’s detention.” Swain v. Pressley, 430 U.S.\\n372, 381 (1977); see Crater v. Galarza, 491 F.3d 1119,\\n1124–25 (9th Cir. 2007) (noting, in the context of a challenge\\nto AEDPA, that not all restrictions on habeas review\\neffectively suspend the writ). We bear in mind that “[a]t its\\nhistorical core, the writ of habeas corpus has served as a\\n\\n\\n 19\\n In so doing, we reject any argument that only noncitizens who have\\n“been lawfully admitted” may invoke the Suspension Clause. Because the\\nwrit is an indispensable separation of powers mechanism, “[t]he test for\\ndetermining the scope of this provision must not be subject to\\nmanipulation by those whose power it is designed to restrain.”\\nBoumediene, 553 U.S. at 765–66.\\n\\f42 THURAISSIGIAM v. USDHS\\n\\nmeans of reviewing the legality of Executive detention, and\\nit is in that context that its protections have been strongest.”\\nSt. Cyr, 533 U.S. at 300–01; see also Boumediene, 553 U.S.\\nat 783 (noting that in cases of executive detention, “the need\\nfor habeas corpus is more urgent”). Therefore, when\\nevaluating whether a substitute is adequate, we consider “the\\nrigor of any earlier proceedings” and “the intended duration\\nof the detention and the reasons for it.” Id. at 781, 783.\\n\\n The government urges a different approach to step two.\\nThe government contends that Thuraissigiam’s status matters\\nto the extent of review the Suspension Clause requires. The\\ngovernment even suggests we should apply the Boumediene\\nstep one extraterritorial factors to determine whether\\n§ 1252(e)(2) provides sufficient review. However, those\\nfactors have no bearing on step two; only step one considers\\nthe petitioner’s status. See Boumediene, 533 U.S. at 773–93\\n(considering, without any reference to Guantanamo\\ndetainees’ status, whether the DTA was an adequate habeas\\nstatute by assessing “the sum total of procedural protections\\nafforded to the detainee at all stages, direct and collateral”).\\nBoth logically and as applied in Boumediene, the\\ncircumstances relevant to step two – the extent of review the\\nSuspension Clause requires – are those relating to the\\ndetainer, not the detainee. We also reject the government’s\\ncontention that because, in its view, Thuraissigiam lacks due\\nprocess rights, there are no rights for the Suspension Clause\\nto protect. Boumediene foreclosed that argument by holding\\nthat, whether or not due process was satisfied, the Suspension\\nClause might require more. 553 U.S. at 785.\\n\\n As a reminder, Thuraissigiam’s petition contends that the\\ngovernment denied him a “fair procedure,” “appl[ied] an\\nincorrect legal standard” to his credible fear contentions, and\\n\\f THURAISSIGIAM v. USDHS 43\\n\\n“fail[ed] to comply with the applicable statutory and\\nregulatory requirements.” The core of his claim is that the\\ngovernment failed to follow the required procedures and\\napply the correct legal standards when evaluating his credible\\nfear claim. As Thuraissigiam’s brief states: “Petitioner’s\\nclaims merely assert his right to the meaningful credible fear\\nprocedure to which he is entitled under the immigration\\nstatute, regulations, and Constitution.” We therefore consider\\nwhether the Suspension Clause requires review of those\\nclaims.20 We conclude that the Clause requires such review\\nin Thuraissigiam’s case and that because § 1252(e)(2) fails to\\nprovide a meaningful opportunity for such review, it raises\\nserious Suspension Clause questions.\\n\\n At step two, the finality era again informs our analysis of\\nwhat the Suspension Clause requires when a removal order is\\nchallenged. Finality era precedent establishes that the Court\\nregularly reviewed on habeas “claims for statutory as well as\\nconstitutional error in deportation proceedings” and “claims\\n\\n\\n 20\\n Thuraissigiam’s petition indicates that he might be asking a federal\\ncourt to review the agency’s credible fear determination, as he contends\\nthat he “can show a significant possibility of prevailing on his claims for\\nasylum and other forms of relief.” The government accordingly contends\\nthat Thuraissigiam’s petition instead requests “ultimate application of a\\nlegal standard to factual determinations and weighing of evidence\\nunderlying the Executive’s negative credible-fear findings.” However, we\\nread Thuraissigiam’s petition to be explaining why, in his view, DHS’\\nprocedural errors matter, particularly given his express assertion that he\\nonly wants review of the procedural errors. We therefore do not consider\\nhere whether the Suspension Clause requires judicial review of DHS’\\ncredible fear determination on the merits. Cf. Ortiz-Alfaro v. Holder,\\n694 F.3d 955, 958 (9th Cir. 2012) (noting, based on St. Cyr, that\\n“depriving [petitioner] the opportunity for judicial review of a\\ndetermination that he lacks a reasonable fear of persecution could raise\\nserious constitutional concerns”).\\n\\f44 THURAISSIGIAM v. USDHS\\n\\nthat deportation hearings were conducted unfairly.” Flores-\\nMiramontes, 212 F.3d at 1143 (citing, inter alia, Fong Haw\\nTan v. Phelan, 333 U.S. 6, 8–10 (1948) (interpreting statute\\non habeas)); Kessler v. Strecker, 307 U.S. 22, 33–34 (1939)\\n(same)). In Gegiow, the Court also reviewed the executive’s\\napplication of a legal standard to undisputed facts, concluding\\nthat the government had incorrectly determined that the\\npetitioner was likely to become a public charge. 239 U.S. at\\n9–10. Similarly, we have interpreted the nature of habeas\\nreview, encompassing inquiry into “the erroneous application\\nor interpretation of statutes,” St. Cyr, 533 U.S. at 302, to\\nrequire that “mixed questions of fact and law – those\\ninvolving an application of law to undisputed fact . . . be\\nprovided meaningful judicial review.” Ramadan v. Gonzales,\\n479 F.3d 646, 652 (9th Cir. 2007) (per curiam).\\n\\n Thuraissigiam and amici point us to other examples of the\\nCourt reviewing not just pure legal questions like the one at\\nissue in St. Cyr, but also the application of a legal standard to\\nundisputed facts. See, e.g., Hansen v. Haff, 291 U.S. 559,\\n562–63 (1934) (rejecting government determination that\\n“petitioner’s entry was for the purpose” of immoral\\nrelations); Mahler v. Eby, 264 U.S. 32, 44 (1924) (holding\\nthat the government failed to comply “with all the statutory\\nrequirements”). Those cases suggest that the Suspension\\nClause requires review of legal and mixed questions of law\\nand fact related to removal orders, including expedited\\nremoval orders.\\n\\n As in Boumediene, the decision to place a noncitizen in\\nexpedited removal and the finding of whether that noncitizen\\nhas a credible fear are both executive determinations,\\nmeaning that the requirements of habeas are “more urgent.”\\nCf. Boumediene, 553 U.S. at 783. While the duration of\\n\\f THURAISSIGIAM v. USDHS 45\\n\\nThuraissigiam’s detention may seem to cut against review,\\nthe Court has recognized that an excluded person’s\\n“movements are restrained by authority of the United States.”\\nMezei, 345 U.S. at 213. Moreover, “it would be difficult to\\nsay that [Thuraissigiam] was not imprisoned, theoretically as\\nwell as practically, when to turn him back meant that he must\\nget into a vessel against his wish and be carried to [Sri\\nLanka].” Chin Yow v. United States, 208 U.S. 8, 12 (1908).\\nThe finality era cases also demonstrate that habeas is a viable\\nmeans of reviewing exclusion and removal orders.\\n\\n Most important, habeas review provides important\\noversight of whether DHS complied with the required\\ncredible fear procedures.21 Under the existing administrative\\nscheme, there are no rigorous adversarial proceedings prior\\nto a negative credible fear determination. First, the credible\\nfear interview is initiated only after the CBP officer identifies\\na noncitizen who fears persecution and refers that individual\\nto a USCIS officer. See 8 C.F.R. § 235.3(b)(4); see also\\nRefugee and Human Rights Amicus Br. 11–12. A noncitizen\\ncan consult with someone at his own expense before his\\nasylum officer interview, but only as long as such\\nconsultation does not “unreasonably delay the process and is\\nat no expense to the government.” 8 C.F.R. § 208.30(d)(4).\\nBefore the IJ hearing, a noncitizen in expedited removal may\\n\\n\\n 21\\n Section 1252(e)(2) also restricts judicial oversight of whether the\\nagency properly placed a person in expedited removal in the first place:\\n“The troubling reality of the expedited removal procedure is that a CBP\\nofficer can create the § 1182(a)(7) charge by deciding to convert the\\nperson’s status from a non-immigrant with valid papers to an intending\\nimmigrant without the proper papers, and then that same officer, free from\\nthe risk of judicial oversight, can confirm his or her suspicions of the\\nperson’s intentions and find the person guilty of that charge.” Khan,\\n608 F.3d at 329.\\n\\f46 THURAISSIGIAM v. USDHS\\n\\nagain consult with someone at his own expense, but the\\nperiod to obtain such assistance is extremely abbreviated: an\\nIJ “shall conclude the review to the maximum extent\\npracticable within 24 hours” of the supervisory officer’s\\napproval of the asylum officer’s determination. 8 C.F.R.\\n§ 1003.42(c), (e). Such review may take place “in person or\\nvia telephonic or video connection.” Jaya Ramji, Legislating\\nAway International Law: The Refugee Provisions of the\\nIllegal Immigration Reform and Immigrant Responsibility\\nAct, 37 Stan. J. Int’l L. 117, 134 –41 (2001). There is also no\\nrequirement that the IJ provide reasons for her decision.\\nIndeed, in this case, the IJ simply checked a box on a form\\nstating that the immigration officer’s decision was\\n“Affirmed.”\\n\\n These meager procedural protections are compounded by\\nthe fact that § 1252(e)(2) prevents any judicial review of\\nwhether DHS complied with the procedures in an individual\\ncase, or applied the correct legal standards.22 We think it\\nobvious that the constitutional minimum – whether\\nThuraissigiam was detained pursuant to the “erroneous\\ninterpretation or application of relevant law” – is not satisfied\\nby such a scheme.23 Our conclusion is bolstered by the Third\\n\\n\\n 22\\n One amicus brief describes reports that the agency does not always\\nfollow the required procedures. See Refugee and Human Rights Amicus\\nBr. 16–27; see also Michele R. Pistone & John J. Hoeffner, Rules Are\\nMade to Be Broken: How the Process of Expedited Removal Fails Asylum\\nSeekers, 20 Geo. Immigr. L.J. 167, 175–93 (2006) (describing procedural\\nerrors commonly committed during the expedited removal process). If\\ntrue, those reports only underscore the need for judicial review.\\n 23\\n A petitioner’s perceived ultimate desire – as Judge Hardiman put\\nit in Castro, to “alter their status in the United States in the hope of\\navoiding release to their homelands,” 835 F.3d at 450–51 (Hardiman, J.,\\n\\f THURAISSIGIAM v. USDHS 47\\n\\nCircuit’s recent decision in Osorio-Martinez. As Osorio-\\nMartinez put it, § 1252(e)(2) fails to provide “even [that]\\n‘uncontroversial’ baseline of review” required by\\nBoumediene. 893 F.3d at 177. Because the statute prevented\\nthe district court from considering whether the agency\\nlawfully applied the expedited removal statute, it a fortiori\\nprecluded review of “the erroneous application or\\ninterpretation of relevant law.” Id. (citing Boumediene,\\n553 U.S. at 779). So too here, because § 1252(e)(2) prevents\\na court from reviewing claims of procedural error relating to\\na negative credible fear determination, it precludes review of\\nthe agency’s application of relevant law and thus raises\\nserious Suspension Clause questions.24 Plenary power\\nconcerns cannot in all circumstances overwhelm the\\n“fundamental procedural protections of habeas corpus . . . , a\\nright of first importance.” Boumediene, 553 U.S. at 798.\\n\\nIV. The Canon of Constitutional Avoidance\\n\\n We further decline to interpret § 1252(e)(2) to avoid the\\nserious Suspension Clause problems engendered by the\\nstatute. The constitutional avoidance canon applies “if an\\notherwise acceptable construction of a statute would raise\\n\\n\\nconcurring dubitante) – is not relevant where a petitioner challenges the\\nfairness of specific procedures leading to an expedited removal order.\\n 24\\n Because Thuraissigiam’s petition does not present the question, we\\ndo not consider one amicus’ argument that “there is a compelling case for\\nallowing habeas courts to review factual challenges to an expedited\\nremoval order.” Scholars of Habeas Corpus Law Amicus Br. 18.\\n“Generally . . . the court will not consider arguments raised only in amicus\\nbriefs.” United States v. Wahchumwah, 710 F.3d 862, 868 (9th Cir. 2013)\\n(citation omitted); see also Russian River Watershed Prot. Comm. v. City\\nof Santa Rosa, 142 F.3d 1136, 1141 n.1 (9th Cir. 1998).\\n\\f48 THURAISSIGIAM v. USDHS\\n\\nserious constitutional problems, and where an alternative\\ninterpretation of the statute is ‘fairly possible.’” St. Cyr,\\n533 U.S. at 299–300 (citation omitted); see also Ramadan,\\n479 F.3d at 654 (“The Supreme Court has been careful to\\nconstrue statutes in light of the Suspension Clause.”).\\nHowever, for us to apply the canon, the statute in question\\nmust be “susceptible of more than one construction.” Clark\\nv. Martinez, 543 U.S. 371, 385 (2005).\\n\\n As explained at length above, we and other courts have\\nconsistently interpreted § 1252(e)(2) to foreclose review of\\nclaims like Thuraissigiam’s. Section 1252(a)(2)(A)(i)\\nprecludes review of “any other cause or claim arising from or\\nrelating to the implementation of or operation of” an\\nexpedited removal order, which clearly bars claims relating\\nto procedural error. We do not think the statute can bear a\\nreading that avoids the constitutional problems it creates.\\n\\n • ! •\\n\\n Therefore, we hold that § 1252(e)(2) violates the\\nSuspension Clause as applied to Thuraissigiam, although we\\ndo not profess to decide in this opinion what right or rights\\nThuraissigiam may vindicate via use of the writ. The district\\ncourt has jurisdiction and, on remand, should exercise that\\njurisdiction to consider Thuraissigiam’s legal challenges to\\nthe procedures leading to his expedited removal order.\\n\\n The judgment of the district court is REVERSED and the\\ncase is REMANDED for further proceedings consistent with\\nthis opinion.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4374951/\", \"author_raw\": \"TASHIMA, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"A WALLACE TASHIMA"},"judge2":{"kind":"string","value":"M MARGARET MCKEOWN"},"judge3":{"kind":"string","value":"RICHARD A PAEZ"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4597698/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Vijayakumar Thuraissigiam v. Usdhs\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-07\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nVIJAYAKUMAR THURAISSIGIAM, No. 18-55313\n Petitioner-Appellant,\n D.C. No.\n v. CV 18-135 AJB\n\nU.S. DEPARTMENT OF HOMELAND\nSECURITY; U.S. CUSTOMS AND\nBORDER PROTECTION; U.S. OPINION\nCITIZENSHIP AND IMMIGRATION\nSERVICES; U.S. IMMIGRATION AND\nCUSTOMS ENFORCEMENT; KIRSTJEN\nNIELSEN, Secretary of DHS;\nWILLIAM P. BARR, Attorney\nGeneral; KEVIN K. MCALEENAN,\nActing Commissioner of CBP;\nTHOMAS HOMAN; L. FRANCIS\nCISSNA, Director of USCIS; PETE\nFLORES, San Diego Field Director,\nCBP; GREGORY ARCHAMBEAULT,\nSan Diego Field Office Director,\nICE; FRED FIGUEROA, Warden, Otay\nMesa Detention Center,\n Respondents-Appellees.\n\n\n Appeal from the United States District Court\n for the Southern District of California\n Anthony J. Battaglia, District Judge, Presiding\n\f2 THURAISSIGIAM v. USDHS\n\n Argued and Submitted May 17, 2018\n Portland, Oregon\n\n Filed March 7, 2019\n\n Before: A. Wallace Tashima, M. Margaret McKeown,\n and Richard A. Paez, Circuit Judges.\n\n Opinion by Judge Tashima\n\n\n SUMMARY*\n\n\n Immigration\n\n In an action in which Vijayakumar Thuraissigiam filed a\nhabeas petition to challenge procedures leading to his\nexpedited removal order, the panel reversed the district\ncourt’s dismissal of the petition for lack of subject matter\njurisdiction, held that 8 U.S.C. § 1252(e)(2) violates the\nSuspension Clause as applied to Thuraissigiam, and\nremanded.\n\n Under 8 U.S.C. § 1225(b)(1)(A)(i), when a U.S. Customs\nand Border Protection (“CBP”) officer determines that a\nnoncitizen arriving at a port of entry is inadmissible for\nmisrepresenting a material fact or lacking necessary\ndocumentation, the officer must place the noncitizen in so-\ncalled “expedited removal” proceedings. Expedited removal\nalso applies to inadmissible noncitizens arrested within 100\n\n *\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\f THURAISSIGIAM v. USDHS 3\n\nmiles of the border and unable to prove that they have been\nin the United States for more than the prior two weeks.\n\n The Department of Homeland Security (“DHS”) removes\nnoncitizens eligible for expedited removal without further\nhearing or review, subject to only one exception: If, in an\ninterview with a CBP officer, the noncitizen indicates an\nintent to apply for asylum or a fear of persecution, DHS must\nrefer the noncitizen for an interview with an asylum officer.\nIf the asylum officer finds no credible fear of persecution, the\nnoncitizen will be removed. A supervisor reviews the asylum\nofficer’s credible fear determination, and the noncitizen may\nalso request de novo review by an immigration judge.\n\n Congress sharply circumscribed judicial review of the\nexpedited removal process such that “no court shall have\njurisdiction to review . . . any individual determination [or]\n. . . the application of [§ 1225(b)(1)] to individual aliens”\noutside of the review permitted by the habeas review\nprovision, § 1252(e). 8 U.S.C. § 1252(a)(2)(A)(iii). Under\n§ 1252(e)(2), a person in expedited removal proceedings may\nfile a habeas petition in federal district court to contest three\nDHS determinations: whether the person is a noncitizen,\nwhether he “was ordered removed” via expedited removal,\nand whether he is a lawful permanent resident or has another\nstatus exempting him from expedited removal.\n\n Thuraissigiam is a native and citizen of Sri Lanka and a\nTamil, an ethnic minority group in Sri Lanka. After crossing\ninto the United States, he was arrested 25 yards north of the\nMexican border, and placed in expedited removal\nproceedings. He was referred for a credible fear interview\nafter he indicated a fear of persecution in Sri Lanka, but an\nasylum officer determined that Thuraissigiam had not\n\f4 THURAISSIGIAM v. USDHS\n\nestablished a credible fear of persecution. A supervisor\napproved the decision, and an immigration judge affirmed the\nnegative credible fear finding in a check-box decision and\nreturned the case to DHS for Thuraissigiam’s removal.\n\n Thuraissigiam filed a habeas petition in federal district\ncourt, arguing that his expedited removal order violated his\nstatutory, regulatory, and constitutional rights. The district\ncourt dismissed the petition for lack of subject matter\njurisdiction, concluding that § 1252(e) did not authorize\njurisdiction over Thuraissigiam’s claims and rejecting his\nSuspension Clause arguments.\n\n The panel concluded, in line with this court’s precedents,\nthat § 1252(e)(2) does not authorize jurisdiction over\nThuraissigiam’s petition because § 1252(e)(2) limits a district\ncourt to reviewing three basic factual determinations related\nto an expedited removal order, and Thuraissigiam’s petition\ndoes not challenge any of those determinations.\n\n The panel next considered whether the habeas review\navailable to Thuraissigiam under § 1252(e)(2) satisfied the\nrequirements of the Suspension Clause. In doing so, the\npanel observed that Boumediene v. Bush, 553 U.S. 723\n(2008), provides an analytical template for evaluating a\nSuspension Clause challenge: at step one, the court examines\nwhether the Suspension Clause applies to the petitioner; and,\nif so, at step two, the court examines whether the substitute\nprocedure provides review that satisfies the Clause.\n\n The panel also observed that INS v. St. Cyr, 533 U.S. 289,\n302 (2001), illuminates how to approach both Boumediene\nsteps. The panel explained that, like Boumedine, St. Cyr\nlooked to the 1789-era historical application of the writ of\n\f THURAISSIGIAM v. USDHS 5\n\nhabeas corpus, but that St. Cyr also looked to habeas\nprecedents from the co-called “finality era,” a period from\n1891 to 1952 during which the statutory scheme precluded\njudicial intervention in immigration enforcement, except as\nrequired by the Constitution.\n\n Drawing from Boumedine and St. Cyr, the panel\nconcluded that it must evaluate Thuraissigiam’s Suspension\nClause challenge in two steps: First, to determine whether\nThuraissigiam may invoke the Suspension Clause, the court\nexamines 1789-era practice, the finality era cases, and other\nrelevant precedents. Second, the court asks whether\n§ 1252(e)(2) provides Thuraissigiam a “meaningful\nopportunity to demonstrate that he is being held pursuant to\n‘the erroneous application or interpretation’ of relevant law.”\nBoumediene, 553 U.S. at 779 (quoting St. Cyr, 533 U.S. at\n302).\n\n In applying Boumediene step one, which concerns the\nreach of the Suspension Clause, the panel concluded that\nThuraissigiam, who was arrested within the United States,\nmay invoke the Suspension Clause. In so concluding, the\npanel observed that, in the finality era, the Court permitted\neven arriving noncitizens to invoke habeas review.\n\n The panel considered at Boumediene step two whether\nhabeas review under § 1252(e) is so limited so as effectively\nto suspend the writ as applied to Thuraissigiam. In doing so,\nthe panel rejected the government’s contention that because,\nin its view, Thuraissigiam lacks due process rights, there are\nno rights for the Suspension Clause to protect, explaining that\nBoumediene foreclosed that argument by holding that,\nwhether or not due process was satisfied, the Suspension\nClause might require more. Reviewing the relevant\n\f6 THURAISSIGIAM v. USDHS\n\nprecedent, the panel concluded that the Suspension Clause\nrequires review of Thuraissigiam’s claims.\n\n The panel held that 8 U.S.C. § 1252(e)(2) violates the\nSuspension Clause as applied to Thuraissigiam. Noting the\nmeager procedural protections afforded by the administrative\nscheme governing credible fear determinations in this\ncontext, and the fact that § 1252(a)(2) prevents any judicial\nreview of whether DHS complied with the procedures in an\nindividual case or applied the correct standard, the panel\nconcluded that it is obvious that the constitutional minimum\n– whether Thuraissigiam was detained pursuant to the\n“erroneous interpretation or application of relevant law” – is\nnot satisfied by the scheme set out by § 1252(e)(2).\n\n The panel further declined to interpret § 1252(e)(2) to\navoid the serious Suspension Clause problems engendered by\nthe statute, explaining that the constitutional avoidance canon\ndoes not apply because the statute cannot bear a reading that\navoids the constitutional problems it creates.\n\n Finally, the panel observed that it did not profess to\ndecide in this opinion what right or rights Thuraissigiam may\nvindicate via use of the writ, and remanded the case for the\ndistrict court to exercise jurisdiction to consider\nThuraissigiam’s legal challenges to the procedures leading to\nhis expedited removal order.\n\f THURAISSIGIAM v. USDHS 7\n\n COUNSEL\n\nLee P. Gelernt (argued) and David Hausman, American Civil\nLiberties Union Foundation Immigrants’ Rights Project, New\nYork, New York; Jennifer Chang Newell and Cody Wofsy,\nAmerican Civil Liberties Union Foundation Immigrants’\nRights Project, San Francisco, California; David Loy, ACLU\nFoundation of San Diego & Imperial Counties, San Diego,\nCalifornia; for Petitioner-Appellant.\n\nJoshua S. Press (argued), and Joseph A. Darrow, Trial\nAttorneys; Erez Reuveni, Assistant Director; William C.\nPeachey, Director; Office of Immigration Litigation, Civil\nDivision, United States Department of Justice, Washington,\nD.C.; for Respondent-Appellees.\n\nMatthew E. Price, Jenner & Block, New York, New York;\nBlaine Bookey, Karen Musalo, and Eunice Lee, Center for\nGender & Refugee Studies, University of California Hastings\nCollege of the Law, San Francisco, Calfornia; for Amici\nCuriae Refugee and Human Rights Organizations and\nScholars.\n\nEthan D. Dettmer, Soolean Choy, Priyah Kaul, and Eli M.\nLazarus, Gibson Dunn & Crutcher, San Francisco, California;\nJoshua S. Lipshutz, Gibson Dunn & Crutcher, Washington,\nD.C.; for Amici Curiae Scholars of Immigration Law.\n\nNoah A. Levine, Wilmer Cutler Pickering Hale & Dorr, New\nYork, New York, for Amici Curiae Scholars of Habeas\nCorpus Law.\n\f8 THURAISSIGIAM v. USDHS\n\nAnjali Srinivasan and Leo L. Lam, Keker Van Nest & Peters,\nSan Francisco, California, for Amici Curiae Scholars of Sri\nLankan Politics.\n\n\n OPINION\n\nTASHIMA, Circuit Judge:\n\n Vijayakumar Thuraissigiam filed a habeas petition in\ndistrict court pursuant to 8 U.S.C. § 1252(e)(2) to challenge\nthe procedures leading to his expedited removal order. The\ncourt dismissed the petition for lack of subject matter\njurisdiction. We reverse. Although § 1252(e)(2) does not\nauthorize jurisdiction over the claims in Thuraissigiam’s\npetition, the Suspension Clause, U.S. Const. art. I, § 9, cl. 2,\nrequires that Thuraissigiam have a “meaningful opportunity\nto demonstrate that he is being held pursuant to ‘the\nerroneous application or interpretation’ of relevant law.”\nBoumediene v. Bush, 553 U.S. 723, 779 (2008) (quoting INS\nv. St. Cyr, 533 U.S. 289, 302 (2001)). Because § 1252(e)(2)\ndoes not provide that meaningful opportunity, the provision\nviolates the Suspension Clause as applied to Thuraissigiam.\n\n BACKGROUND\n\nI. Statutory Background\n\n When a U.S. Customs and Border Protection (“CBP”)\nofficer determines that a noncitizen arriving at a port of entry\nis inadmissible for misrepresenting a material fact or lacking\n\f THURAISSIGIAM v. USDHS 9\n\nnecessary documentation,1 the officer must place the\nnoncitizen in so-called “expedited removal” proceedings.\n8 U.S.C. § 1225(b)(1)(A)(i). By regulation, the Department\nof Homeland Security (“DHS”), of which CBP is a\nconstituent agency, also applies expedited removal to\ninadmissible noncitizens arrested within 100 miles of the\nborder and unable to prove that they have been in the United\nStates for more than the prior two weeks. Designating Aliens\nfor Expedited Removal, 69 Fed. Reg. 48877-01, 48879-80\n(Aug. 11, 2004);2 see also 8 U.S.C. § 1225(b)(1)(A)(iii)(II).\n\n DHS removes noncitizens eligible for expedited removal\n“without further hearing or review,” subject to only one\nexception. 8 U.S.C. § 1225(b)(1)(A)(i). If, in an interview\nwith a CBP officer, the noncitizen indicates an intent to apply\nfor asylum or a fear of persecution, DHS must refer the\nnoncitizen for an interview with an asylum officer. Id.\n§ 1225(b)(1)(A)(ii); 8 C.F.R. § 208.30. If that asylum officer\ndetermines that the noncitizen’s fear of persecution is\n\n\n 1\n See 8 U.S.C. § 1182(a)(6)(C) (misrepresentation bar); id.\n§ 1182(a)(7) (documentation bar).\n 2\n Congress gave the Attorney General the authority to extend\nexpedited removal to some or all inadmissible noncitizens who cannot\nprove that they have been in the United States for more than two years\nprior; thus, the current regime does not represent the full exercise of\nexecutive authority permitted by statute. 8 U.S.C. § 1225(b)(1)(A)(iii).\nDHS also applies expedited removal to noncitizens who entered the\nUnited States by sea and who have not been in the United States for two\nyears. See Notice Designating Aliens Subject to Expedited Removal\nUnder Section 235(b)(1)(A)(iii) of the Immigration and Nationality Act,\n67 Fed. Reg. 68,924, 68,924–25 (Nov. 13, 2002). The current regime\nmay, however, expand; a January 2017 executive order instructs the\nSecretary of DHS to apply expedited removal to the fullest extent of the\nlaw. See Exec. Order No. 13,767, 82 Fed. Reg. 8793 (Jan. 25, 2017).\n\f10 THURAISSIGIAM v. USDHS\n\ncredible, the noncitizen is referred to non-expedited removal\nproceedings, in which the noncitizen may apply for asylum or\nother forms of relief from removal. See 8 U.S.C.\n§ 1225(b)(1)(B)(ii); 8 C.F.R. § 208.30(f); 8 C.F.R.\n§ 1003.42(f). If the asylum officer finds no credible fear of\npersecution, the noncitizen will be removed. 8 U.S.C.\n§ 1225(b)(1)(B)(iii). A supervisor reviews the asylum\nofficer’s credible fear determination, 8 C.F.R.\n§§ 208.30(e)(7), 235.3(b)(2), (b)(7), and a noncitizen may\nalso request de novo review by an immigration judge (“IJ”).\n8 U.S.C. § 1225(b)(1)(B)(iii)(III); 8 C.F.R. § 1003.42. In\n2016, DHS conducted over 141,000 expedited removals. See\nRefugee and Human Rights Amicus Br. 10. All individuals\nplaced in expedited removal proceedings are subject to\nmandatory detention pending a final determination of credible\nfear of persecution or until they are removed. 8 U.S.C.\n§ 1225(b)(1)(B)(iii)(IV).\n\n Congress sharply circumscribed judicial review of the\nexpedited removal process. “[N]o court shall have\njurisdiction to review . . . any individual determination [or]\n. . . the application of [§ 1225(b)(1)] to individual aliens”\noutside of the review permitted by the habeas review\nprovision, § 1252(e). 8 U.S.C. § 1252(a)(2)(A)(iii). Under\n§ 1252(e)(2), a person in expedited removal proceedings may\nfile a habeas petition in federal district court to contest three\nDHS determinations: whether the person is a noncitizen,\nwhether he “was ordered removed” via expedited removal,\nand whether he is a lawful permanent resident or has another\nstatus exempting him from expedited removal. Id.\n§ 1252(e)(2)(A)–(C). Review of whether a petitioner “was\nordered removed” is “limited to whether such an order in fact\nwas issued and whether it relates to the petitioner. Id.\n§ 1252(e)(5). “There shall be no review of whether the alien\n\f THURAISSIGIAM v. USDHS 11\n\nis actually inadmissible or entitled to any relief from\nremoval.” Id.; see also 8 C.F.R. § 1003.42(f) (“No appeal\nshall lie from a review of an adverse credible fear\ndetermination made by an immigration judge.”).3\n\nII. Factual Background\n\n Thuraissigiam is a native and citizen of Sri Lanka and a\nTamil, an ethnic minority group in Sri Lanka. See Scholars\nof Sri Lankan Politics Amicus Br. 3. Thuraissigiam fled his\nhome country in June 2016 and made his way to Mexico. On\nFebruary 17, 2017, Thuraissigiam crossed the border into the\nUnited States. Late that night, he was arrested by a CBP\nofficer four miles west of the San Ysidro, California, port of\nentry, 25 yards north of the border.\n\n DHS placed Thuraissigiam in expedited removal\nproceedings. Pursuant to 8 U.S.C. § 1225(b)(1)(A)(ii), CBP\nreferred Thuraissigiam for an interview with an asylum\nofficer after he indicated a fear of persecution in Sri Lanka.\nOn March 9, an asylum officer from the United States\nCitizenship and Immigration Services (“USCIS”) interviewed\nThuraissigiam and determined that he had not established a\ncredible fear of persecution. A supervisor approved the\n\n\n 3\n Under 8 U.S.C. § 1252(e)(3), a person may challenge the\nconstitutionality and legality of the expedited removal provisions,\nregulations implementing those provisions, or written policies to\nimplement the provisions. Such challenges, however, must be brought\nwithin 60 days after implementation and only in the District of Columbia.\nId. § 1252(e)(3)(A)–(B). Various expedited removal provisions and\nimplementing regulations survived a § 1252(e)(3) challenge in American\nImmigration Lawyers Ass’n v. Reno, 18 F. Supp. 2d 38 (D.D.C. 1998),\nalthough the plaintiffs did not raise a Suspension Clause argument about\nthe extent of habeas review. See id. at 41.\n\f12 THURAISSIGIAM v. USDHS\n\ndecision. Thuraissigiam then requested review by an IJ, who\naffirmed the negative credible fear finding in a check-box\ndecision and returned the case to DHS for Thuraissigiam’s\nremoval.\n\nIII. District Court Proceedings\n\n In January 2018, Thuraissigiam filed a habeas petition in\nfederal district court, naming as respondents DHS, several of\nits constituent agencies, and individual agency officials.\nThuraissigiam argued that his “expedited removal order\nviolated his statutory, regulatory, and constitutional rights,”\nsought to vacate the order, and requested relief in the form of\na “new, meaningful opportunity to apply for asylum and other\nrelief from removal.” Thuraissigiam alleged that in Sri Lanka\nhe had been harassed for supporting a Tamil political\ncandidate. In 2007, he was “detained and beaten” by Sri\nLankan army officers, and told not to support the candidate.\nIn 2014, after Thuraissigiam continued to support the\ncandidate, government intelligence officers kidnapped,\nbound, and beat him during an interrogation about his\npolitical activities. Thuraissigiam alleged that he “was\nlowered into a well, simulating drowning, threatened with\ndeath, and then suffocated, causing him to lose\nconsciousness.”\n\n Thuraissigiam also made various factual allegations about\nthe expedited removal procedures to which he was subject\nafter being apprehended. For one, he alleged that the asylum\nofficer failed to “elicit all relevant and useful information\nbearing on whether the applicant has a credible fear of\npersecution or torture” in violation of 8 C.F.R. § 208.30(d)\nand “failed to consider relevant country conditions evidence”\nin violation of 8 U.S.C. § 1225(b)(1)(B)(v) and 8 C.F.R.\n\f THURAISSIGIAM v. USDHS 13\n\n§ 208.30(e)(2). Thuraissigiam also alleged that there were\n“communication problems” between the asylum officer,\nThuraissigiam, and the translator, in violation of 8 C.F.R.\n§ 208.30(d)(1)–(2). Thuraissigiam alleged that the IJ hearing\nincluded the same procedural and substantive flaws, and that\nat both hearings, he was unaware whether “information he\noffered would be shared with the Sri Lankan government.”\nThuraissigiam’s petition asserted two causes of action:\n\n First, DHS’ credible fear screening deprived\nThuraissigiam “of a meaningful right to apply for asylum”\nand other forms of relief, in violation of 8 U.S.C.\n§ 1225(b)(1), its implementing regulations, and the United\nStates Convention Against Torture, implemented in the\nForeign Affairs Reform and Restructuring Act of 1998\n(“FARRA”), Pub. L. No. 105-277, div. G., Title XXII,\n§ 2242, 112 Stat. 2681 (1998). The asylum officer and IJ also\nviolated those statutes “by applying an incorrect legal\nstandard” to Thuraissigiam’s credible fear application.\n\n Second, the asylum officer and IJ violated\nThuraissigiam’s rights under the Due Process Clause of the\nFifth Amendment by “not providing him with a meaningful\nopportunity to establish his claims, failing to comply with the\napplicable statutory and regulatory requirements, and in not\nproviding him with a reasoned explanation for their\ndecisions.”\n\n The district court dismissed the habeas petition for lack of\nsubject matter jurisdiction. Thuraissigiam v. U.S. Dep’t of\nHomeland Sec., 287 F. Supp. 3d 1077 (S.D. Cal. 2018).\nRelying on our precedents, the district court concluded that\n8 U.S.C. § 1252(e) did not authorize jurisdiction over the\nclaims in Thuraissigiam’s petition. Id. at 1082. Next, the\n\f14 THURAISSIGIAM v. USDHS\n\ncourt rejected Thuraissigiam’s Suspension Clause arguments.\nAlthough the court concluded that Thuraissigiam could\ninvoke the Suspension Clause, it held that the statute’s “strict\nrestraints” on habeas review of expedited removal orders did\nnot effectively suspend the writ of habeas corpus and were\ntherefore constitutionally sound. Id. at 1082–83.4\n\n Thuraissigiam timely appealed the district court’s\ndismissal and moved for a stay of removal pending appeal.\nA motions panel of our court initially denied Thuraissigiam’s\nstay motion, but later vacated that order and stayed\nThuraissigiam’s removal pending appeal.\n\n STANDARD OF REVIEW\n\n We have jurisdiction under 28 U.S.C. § 1291. We review\nde novo the district court’s dismissal of Thuraissigiam’s\nhabeas petition for lack of subject matter jurisdiction. Bishop\nPaiute Tribe v. Inyo County, 863 F.3d 1144, 1151 (9th Cir.\n2017); see also Garcia de Rincon v. Dep’t of Homeland Sec.,\n539 F.3d 1133, 1136 (9th Cir. 2008).\n\n DISCUSSION\n\n We must decide whether a federal district court has\njurisdiction to review the claims in Thuraissigiam’s petition.\nWe first inquire whether 8 U.S.C. § 1252(e)(2) authorizes\njurisdiction over Thuraissigiam’s petition. Concluding that\n§ 1252(e)(2) does not authorize jurisdiction, we then address\n\n\n\n 4\n The district court also denied various stay motions that\nThuraissigiam had filed, concluding that they were moot due to the\npetition’s dismissal. 287 F. Supp. 3d at 1078.\n\f THURAISSIGIAM v. USDHS 15\n\nwhether the provision restricting habeas review violates the\nSuspension Clause.\n\nI. Jurisdiction Under 8 U.S.C. § 1252(e)(2)\n\n Thuraissigiam contends that 8 U.S.C. § 1252(e)(2)(B)\nauthorizes review of the statutory, regulatory, and\nconstitutional claims raised in his habeas petition. We\ndisagree. Section 1252(e)(2), including Subsection (B), limits\na district court to reviewing three basic factual determinations\nrelated to an expedited removal order. Because\nThuraissigiam’s petition does not challenge any of those\ndeterminations, § 1252(e)(2) does not authorize jurisdiction\nover the petition.\n\n A court applying habeas review under § 1252(e)(2) is\nlimited to determining:\n\n (A) whether the petitioner is an alien,\n\n (B) whether the petitioner was ordered\n removed under such section, and\n\n (C) whether the petitioner can prove by a\n preponderance of the evidence that the\n petitioner is an alien lawfully admitted for\n permanent residence, has been admitted as a\n refugee under section 1157 of this title, or has\n been granted asylum under section 1158 of\n this title . . . .\n\nCongress also provided express limitations on review under\nSubsection (B):\n\f16 THURAISSIGIAM v. USDHS\n\n In determining whether an alien has been\n ordered removed under section 1225(b)(1) of\n this title, the court’s inquiry shall be limited to\n whether such an order in fact was issued and\n whether it relates to the petitioner. There\n shall be no review of whether the alien is\n actually inadmissible or entitled to any relief\n from removal.\n\nId. § 1252(e)(5). Nonetheless, Thuraissigiam stakes his claim\non Subsection (B).\n\n We considered and rejected a nearly identical argument\nin Garcia de Rincon. The petitioner in Garcia de Rincon\nraised a due process challenge to an expedited removal order.\n539 F.3d at 1136. Characterizing § 1252(e) as among the\n“most stringent” jurisdiction-limiting provisions in the\nimmigration statutes, we held that § 1252(e)(2) permits\nreview only of “habeas petitions alleging that the petitioner\nis not an alien or was never subject to an expedited removal\norder.” Id. at 1135, 1139. We therefore lacked jurisdiction\nbecause the petitioner’s due process claims were not\nencompassed by those enumerated grounds. Id. at 1139.\nLikewise, Thuraissigiam’s claims of procedural violations are\nplainly not claims about whether Thuraissigiam “was never\nsubject to an expedited removal order.” 8 U.S.C. § 1252(e).\n\n Thuraissigiam contends that such a reading of Subsection\n(B) renders superfluous the prohibition in § 1252(e)(5)\nagainst “review of whether the alien is actually inadmissible\nor entitled to any relief from removal.” However,\n§ 1252(e)(5) functions not to repeat § 1252(e)(2)(B), but to\nexplain it. Moreover, Thuraissigiam’s petition is barred by\nthe first sentence in § 1252(e)(5), not the second sentence.\n\f THURAISSIGIAM v. USDHS 17\n\nBecause he asks the district court to review the government’s\nprocedures, those claims are beyond the scope of “whether\nsuch an [expedited removal] order in fact was issued and\nwhether it relates to the petitioner.”5 8 U.S.C. § 1252(e); see\nalso United States v. Barajas-Alvarado, 655 F.3d 1077, 1082\n(9th Cir. 2011) (reaffirming that jurisdiction under\n§ 1252(e)(2) “does not extend to review of the claim that an\nalien was wrongfully deprived of the administrative review\npermitted under the statute and applicable regulations”).\n\n Thuraissigiam relies on Smith v. U.S. Customs & Border\nProtection, 741 F.3d 1016 (9th Cir. 2014), to contend that we\nhave adopted a more expansive view of Subsection (B).\nSpecifically, he contends that Smith reviewed “whether the\npetitioner belonged in the expedited removal system,” and\nthat a court may thus review his petition. Smith, however,\ndoes not support Thuraissigiam’s argument. In Smith, CBP\nplaced the petitioner, a Canadian citizen arriving at the\nborder, in expedited removal proceedings for lacking certain\ndocuments. Id. at 1019. The petitioner alleged that CBP\nexceeded its authority under the expedited removal statute\nbecause certain document requirements are waived for\nCanadians, and argued that Subsection (B) authorized review.\nId. at 1019, 1021. “Accepting [petitioner’s] theory at face\nvalue,” we reviewed whether CBP in fact classified him as an\n“intending immigrant.” Id. at 1021. Concluding that CBP had\ndone so, we held that § 1252(e)(2) “permit[ted] us to go no\n\n\n 5\n Thuraissigiam also makes a structural argument, contending that\nbecause Congress provided for some review of asylum claims even in\nexpedited removal cases, Congress must not have intended to strip judicial\nreview to “police the boundaries of those limits.” This argument ignores\nthe plain language of the statute, which evidences Congress’ intent to do\njust that.\n\f18 THURAISSIGIAM v. USDHS\n\nfurther” and did not discuss the merits of CBP’s\nclassification. Id. at 1021–22 & n.4.6 Therefore, Smith\nreviewed only how CBP classified the petitioner, which is\nfairly encompassed by whether “[the petitioner] was ordered\nremoved” under the expedited removal provision. Id. at 1022\n(internal quotation marks omitted). By contrast,\nThuraissigiam asks the district court to pass judgment on the\nprocedures leading to his removal order. The limited review\nprovided under § 1252(e)(2) does not encompass such\nclaims.7\n\n Therefore, in line with our precedents, we conclude that\n§ 1252(e) does not authorize habeas review of\nThuraissigiam’s petition. We do not here address\nThuraissigiam’s request that we apply the canon of\nconstitutional avoidance to interpret § 1252(e) to provide\njurisdiction over his legal claims. That canon only comes\ninto play if we conclude that § 1252(e) raises serious\nconstitutional questions; thus, we first address\nThuraissigiam’s Suspension Clause argument before\ncontemplating the application of that canon. See St. Cyr,\n533 U.S. at 299–300, 314 (explaining constitutional\navoidance canon and applying it upon concluding that the\nstatute in question raised serious constitutional questions).\n\n\n 6\n The Smith petitioner also contended that § 1252(e)(2) violated the\nSuspension Clause, but we did not reach the argument because we held\nthat the statute permitted limited review of his petition. 741 F.3d at 1022\nn.6.\n 7\n We have held that in appeals from convictions for criminal reentry,\na defendant may collaterally attack a removal order that forms the basis\nfor his conviction. See United States v. Ochoa-Oregel, 904 F.3d 682, 686\n(9th Cir. 2018) (considering collateral attack on expedited removal order).\nBut that rule does not apply to this case.\n\f THURAISSIGIAM v. USDHS 19\n\nII. Suspension Clause\n\n The Suspension Clause mandates, “The Privilege of the\nWrit of Habeas Corpus shall not be suspended, unless when\nin Cases of Rebellion or Invasion the public Safety may\nrequire it.” U.S. Const. art. I, § 9, cl. 2. Our nation’s\nfounders viewed the writ as a “vital instrument” to secure\nindividual liberty. Boumediene, 553 U.S. at 743. “The\nClause protects the rights of the detained by a means\nconsistent with the essential design of the Constitution. It\nensures that, except during periods of formal suspension, the\nJudiciary will have a time-tested device, the writ, to maintain\nthe ‘delicate balance of governance’ that is itself the surest\nsafeguard of liberty.” Id. at 745 (citing Hamdi v. Rumsfeld,\n542 U.S. 507, 536 (2004) (opinion of O’Connor, J.)). The\nSuspension Clause prevents Congress from passing a statute\nthat effectively suspends the writ absent rebellion or invasion.\nSee Felker v. Turpin, 518 U.S. 651, 663–64 (1996). Thus, the\nquestion in this case is whether 8 U.S.C. § 1252(e)(2)\neffectively suspends the writ. Put another way, the question\nis whether the habeas review available to Thuraissigiam\nunder § 1252(e)(2) satisfies the requirements of the\nSuspension Clause.\n\n The Supreme Court has not yet answered that question.\nIn fact, the Court has rarely addressed who may invoke the\nSuspension Clause and the extent of review the Clause\nrequires. For example, only in Boumediene has the Court\nconcluded that a statute violated the Suspension Clause.\nGerald L. Neuman, The Habeas Corpus Suspension Clause\nAfter Boumediene v. Bush, 110 Colum. L. Rev. 537, 538\n(2010). In the Court’s other most recent Suspension Clause\ncase, St. Cyr, the Court, after extensive analysis of the\nSuspension Clause issues at play, interpreted the statute to\n\f20 THURAISSIGIAM v. USDHS\n\navoid those issues. 533 U.S. at 336–37. Of the federal courts\nof appeals, only the Third Circuit has addressed the precise\nquestion before us, whether § 1252(e)(2) as applied to\nnoncitizen petitioners in expedited removal violates the\nSuspension Clause. See Castro v. U.S. Dep’t of Homeland\nSec., 835 F.3d 422 (3d Cir. 2016) (concluding that, due to\ntheir status, such petitioners could not invoke the Suspension\nClause), cert. denied, 137 S. Ct. 1581 (2017).\n\n Boumediene traced the writ of habeas corpus to its origins\nas a tool of the English crown, citing the detailed historical\naccount in Paul D. Halliday and G. Edward White, The\nSuspension Clause: English Text, Imperial Contexts, and\nAmerican Implications, 94 Va. L. Rev. 575 (2008). 553 U.S.\nat 740. As Halliday and White explain, the writ in England\nwas the vehicle “to determine the rightness of constraints\nimposed on the bodies of the king’s subjects of all kinds.” 94\nVa. L. Rev. at 607. The writ was on occasion suspended in\nEngland. Id. at 619; see Boumediene, 553 U.S. at 741.\nAccording to Boumediene, that history “no doubt confirmed\n[the Framers’] view that pendular swings to and away from\nindividual liberty were endemic to undivided, uncontrolled\npower.” Id. at 742; see also Amanda L. Tyler, A “Second\nMagna Carta”: The English Habeas Corpus Act and the\nStatutory Origins of the Habeas Privilege, 91 Notre Dame L.\nRev. 1949, 1985–86 (2016) (describing how suspensions of\nthe writ in colonial America motivated the States’ desire to\nimport similar habeas protections after gaining\nindependence); Halliday & White, 94 Va. L. Rev. at 671\n(highlighting the Framers’ desire to restore “the traditional\norder of writs and suspensions”).\n\n As Boumediene summed it up, the Suspension Clause is\nrooted in the Framers’ first-hand experience “that the\n\f THURAISSIGIAM v. USDHS 21\n\ncommon-law writ all too often had been insufficient to guard\nagainst the abuse of monarchial power.” 553 U.S. at 739–40.\nThe Clause, therefore, is “not merely about suspending the\nprivilege of the writ of habeas corpus, but about the meaning\nof the ‘privilege of the writ’ itself.” Halliday & White,\n94 Va. L. Rev. at 699. “Indeed, common law habeas corpus\nwas, above all, an adaptable remedy . . . [whose] precise\napplication and scope changed depending upon the\ncircumstances.” Boumediene, 553 U.S. at 779–80 (citing,\ninter alia, Richard H. Fallon, Jr. & Daniel J. Meltzer, Habeas\nCorpus Jurisdiction, Substantive Rights, and the War on\nTerror, 120 Harv. L. Rev. 2029, 2102 (2007)).\n\n In examining how the Supreme Court has defined the\nSuspension Clause’s requirements, Boumediene is our\nstarting point, even if it does not provide a direct answer to\nThuraissigiam’s challenge. Boumediene and its predecessors,\nlike St. Cyr, do provide an analytical blueprint. We therefore\nreview those precedents before deciding how best to apply\ntheir principles to this appeal.\n\n A. Boumediene v. Bush\n\n In Boumediene, the Supreme Court struck down a War on\nTerror-era law after detainees at the Guantanamo Bay prison\nin Cuba brought a Suspension Clause challenge. 553 U.S. at\n732–33. In the wake of the September 11, 2001, attacks, the\nU.S. Department of Defense created Combatant Status\nReview Tribunals (“CSRTs”) to decide if detainees were\n“enemy combatants.” Id. at 733. The Boumediene\npetitioners, who had all appeared before CSRTs and been\ndeemed enemy combatants, sought a writ of habeas corpus\nunder the general habeas statute, 28 U.S.C. § 2241. Id. at\n734. After protracted litigation, Congress passed the\n\f22 THURAISSIGIAM v. USDHS\n\nDetainee Treatment Act of 2005 (“DTA”), which amended\n§ 2241 to bar judicial review of habeas petitions filed by\nGuantanamo detainees and to vest review of CSRT decisions\nexclusively in the D.C. Circuit. Id. at 735 (citing DTA\n§ 1005(e), 119 Stat. 2742). Section 7 of the Military\nCommissions Act of 2006 (“MCA”) made those provisions\nretroactive. Id. at 736. See generally Hamad v. Gates,\n732 F.3d 990, 996–99 (9th Cir. 2013) (describing\nBoumediene’s place in the line of Guantanamo detainee\ncases). The Court took a two-step approach to evaluating the\ndetainees’ challenge to the MCA.\n\n At step one, the Court evaluated whether the Guantanamo\ndetainees – as enemy combatants detained on foreign soil –\ncould even invoke the Suspension Clause. See Boumediene,\n553 U.S. at 739. In so doing, the Court affirmed that\nalthough the writ’s protections may have expanded since the\nConstitution’s drafting, “at the absolute minimum,” the\nClause protects the writ as it existed in 1789. Id. at 746\n(citing St. Cyr, 533 U.S. at 301). The Court therefore\nexamined historical authorities to determine the scope of the\nwrit in 1789, and whether it ran to “an enemy alien detained\nabroad.” Id. at 752. Although noting that “at common law a\npetitioner’s status as an alien was not a categorical bar to\nhabeas corpus,” the Court concluded that the historical record\ndid not provide a definitive answer. Id. at 747, 752. Instead,\nthe Court turned to its extraterritoriality precedents and from\nthem concluded that “questions of extraterritoriality turn on\nobjective factors and practical concerns, not formalism.” Id.\nat 764. Boumediene drew from Johnson v. Eisentrager,\n339 U.S. 763 (1950), another case about the extraterritorial\napplication of the Suspension Clause, three non-exclusive\nfactors relevant to the Clause’s extraterritorial scope:\n\f THURAISSIGIAM v. USDHS 23\n\n (1) the citizenship and status of the detainee\n and the adequacy of the process through\n which that status determination was made;\n (2) the nature of the sites where apprehension\n and then detention took place; and (3) the\n practical obstacles inherent in resolving the\n prisoner’s entitlement to the writ.\n\n553 U.S. at 766.8 Applying those factors, the Court\nconcluded that the detainees could invoke the Suspension\nClause. Id. at 771.\n\n At step two, the Court considered whether Congress had\nsuspended the writ without an adequate substitute. The Court\nacknowledged that there are “few precedents addressing what\nfeatures an adequate substitute for habeas corpus must\ncontain.” Id. at 772. For example, the Court had previously\nupheld provisions of the Antiterrorism and Effective Death\nPenalty Act of 1996 (“AEDPA”) against a Suspension Clause\nchallenge because the provisions “did not constitute a\nsubstantial departure from common-law habeas procedures.”\nId. at 774 (citing Felker, 518 U.S. at 664); see also Neuman,\n110 Colum. L. Rev. at 542 (stating that “what matters is the\nsubstance, not the form, of the Great Writ,” and that\n“Congress can rename or reconfigure the procedure by which\ncourts examine the lawfulness of detention,” as long as the\nsubstitute is adequate).\n\n 8\n In Rasul v. Bush, 542 U.S. 466 (2004), the Court had first discussed\nEisentrager’s applicability to the question of who may invoke the\nSuspension Clause. Id. at 487 (Kennedy, J., concurring) (“A faithful\napplication of Eisentrager, then, requires an initial inquiry into the general\ncircumstances of the detention to determine whether the Court has the\nauthority to entertain the petition and to grant relief after considering all\nof the facts presented.”).\n\f24 THURAISSIGIAM v. USDHS\n\n In Boumediene, the Court gleaned from its precedents two\n“easily identified attributes of any constitutionally adequate\nhabeas corpus proceeding.” 553 U.S. at 779. First, the\n“privilege of habeas corpus entitles the prisoner to a\nmeaningful opportunity to demonstrate that he is being held\npursuant to ‘the erroneous application or interpretation’ of\nrelevant law.” Id. (quoting St. Cyr, 533 U.S. at 302). Second,\n“the habeas court must have the power to order the\nconditional release of an individual unlawfully detained.” Id.\nBeyond those minimum requirements, “depending on the\ncircumstances, more may be required.” Id.\n\n The Court further emphasized that “the necessary scope\nof habeas review in part depends upon the rigor of any earlier\nproceedings.” Id. at 781; see also id. at 786 (noting that\n“habeas corpus review may be more circumscribed if the\nunderlying detention proceedings are more thorough”). For\nthat reason, courts sitting in habeas afford deference when\nreviewing another court’s decision, but when a petitioner is\n“detained by executive order . . . the need for collateral\nreview is most pressing.” Id. at 783. To be effective, the\n“habeas court must have sufficient authority to conduct a\nmeaningful review of both the cause for detention and the\nExecutive’s power to detain.” Id. Applying those principles\nto the CSRTs and D.C. Circuit review, the Court concluded\nthat the MCA did not provide an adequate substitute because\nthe D.C. Circuit could not “consider newly discovered\nevidence that could not have been made part of the CSRT\nrecord.” Id. at 790. The Court then concluded that it was not\npossible to read into the statute provisions for the procedures\nnecessary to satisfy the Suspension Clause, and therefore held\nit unconstitutional. Id. at 792.\n\f THURAISSIGIAM v. USDHS 25\n\n Boumediene provides an analytical template for\nevaluating a Suspension Clause challenge: at step one, we\nexamine whether the Suspension Clause applies to the\npetitioner; and, if so, at step two, we examine whether the\nsubstitute procedure provides review that satisfies the Clause.\nHow more specifically to apply that template is less clear,\ngiven that the Court generated its three-factor test at step one\nin light of the extraterritoriality question in Boumediene. See\nid. at 764, 766. Those factors, as both parties acknowledge,\ndo not map precisely onto this case because Thuraissigiam\nwas apprehended and is detained on U.S. soil.9 Yet, the\nmanner in which the Court divined those factors informs our\napproach here. Boumediene relied on Eisentrager and related\ncases, but also looked to 1789-era application of the writ to\ndetermine whether petitioners similarly situated to\nGuantanamo detainees had been able to invoke the Clause.\nAlthough the Court emphasized that the history was not\ndispositive, it made clear that “settled precedents or legal\ncommentaries in 1789 . . . can be instructive.” Id. at 739.10\n\n\n\n\n 9\n We too have applied the three Boumediene factors more readily\nwhen asking whether a noncitizen outside the United States – again,\nunlike Thuraissigiam – can claim the Constitution’s protections. See\nIbrahim v. Dep’t of Homeland Sec., 669 F.3d 983, 995 (9th Cir. 2012)\n(evaluating extraterritoriality in context of First and Fifth Amendment\nclaims).\n 10\n Indeed, the Supreme Court’s proposition that the Suspension Clause\nat least protects the writ as it existed in 1789 “necessarily invites reference\nto history when interpreting and applying the Suspension Clause.”\nAmanda L. Tyler, Habeas Corpus in Wartime 9 (2017); see also Omar v.\nMcHugh, 646 F.3d 13, 19 (D.C. Cir. 2011) (Kavanaugh, J.) (“[H]istory\nmatters: In habeas cases, we seek guidance from history ‘addressing the\nspecific question before us.’” (quoting Boumediene, 553 U.S. at 746)).\n\f26 THURAISSIGIAM v. USDHS\n\n At step two, Boumediene held that, at a minimum, the\nSuspension Clause entitles a petitioner “to a meaningful\nopportunity to demonstrate he is being held to ‘the erroneous\napplication or interpretation’ of relevant law.” Id. at 779\n(quoting St. Cyr, 533 U.S. at 302). In considering whether\nthe Clause required more in the circumstances of\nBoumediene, the Court impliedly considered the rigor and\ncharacter of the proceedings preceding habeas review. Also\nrelevant to Thuraissigiam’s case, the Court affirmed that the\nSuspension Clause protects “a right of first importance,” even\nin circumstances – such as national security, in Boumediene\n– where the executive’s power is at its zenith. Id. at 797–98.\n\n B. INS v. St. Cyr\n\n St. Cyr, which predated Boumediene by several years,\nsheds additional light on the Court’s approach to Suspension\nClause questions. The petitioner, St. Cyr, was a lawful\npermanent resident admitted to the United States in 1986.\n533 U.S. at 293. In 1996, St. Cyr pleaded guilty to a criminal\ncharge that made him removable, although under pre-AEDPA\nlaw (applicable at the time of his conviction), he was eligible\nfor a discretionary waiver from the Attorney General. Id.\nAfter AEDPA was passed, the government began removal\nproceedings, with the Attorney General interpreting AEDPA\nto have removed his discretion to grant St. Cyr a waiver. Id.\nSt. Cyr filed a habeas petition alleging that the Attorney\nGeneral’s interpretation was erroneous because St. Cyr’s\nconviction predated AEDPA. Id. After the district court and\nSecond Circuit agreed with St. Cyr, the government argued to\n\f THURAISSIGIAM v. USDHS 27\n\nthe Supreme Court that the courts lacked jurisdiction to\nreview the Attorney General’s interpretation. Id. at 297–98.\n\n The Court stated that the government’s position had to\novercome several presumptions, chief among them the\n“strong presumption in favor of judicial review of\nadministrative action and the longstanding rule requiring a\nclear statement of congressional intent to repeal habeas\njurisdiction.” Id. at 298 (citing Ex parte Yerger, 75 U.S.\n(8 Wall.) 85, 102 (1869)). To address whether the statute\nraised serious Suspension Clause questions, the Court started\nfrom the principle that “[b]ecause of [the] Clause, some\n‘judicial intervention in deportation cases’ is unquestionably\n‘required by the Constitution.’” Id. at 300 (citing Heikkila v.\nBarber, 345 U.S. 229, 235 (1953)). Because “at the absolute\nminimum, the Suspension Clause protects the writ ‘as it\nexisted in 1789,’” the Court looked at the writ’s application\nbefore and after the drafting of the Constitution. Id. at 301\n(quoting Felker, 518 U.S. at 663–64). Legal and historical\nauthorities indicated that in both England and the United\nStates “the writ of habeas corpus has served as a means of\nreviewing the legality of Executive detention, and it is in that\ncontext that its protections have been strongest.” Id.\nMoreover, the writ was available both to “nonenemy aliens as\nwell as citizens” and “encompassed detentions based on\nerrors of law, including the erroneous application or\ninterpretation of statutes.” Id. at 301–02.\n\f28 THURAISSIGIAM v. USDHS\n\n St. Cyr also looked to the so-called “finality era,”11 during\nwhich the statutory scheme precluded judicial intervention in\nimmigration enforcement, except as required by the\nConstitution. Id. at 304–06. Despite that statutory bar, the\nCourt in the finality era “allow[ed] for review on habeas of\nquestions of law.” Id. at 304. Accordingly, the government’s\nreading of the statute – to prohibit any judicial review of the\nAttorney General’s interpretation – raised “Suspension\nClause questions that . . . are difficult and significant.” Id.\nMore directly, “to conclude that the writ is no longer\navailable in this context would represent a departure from\nhistorical practice in immigration law.” Id. at 305. After\ncanvassing that historical practice, and noting that it was\nconsistent with the writ’s “common-law antecedents,” the\nCourt concluded that St. Cyr could have brought his habeas\nclaims under that regime. Id. at 308. Thus, due to the serious\nconstitutional questions raised, and because Congress had not\nprovided a “clear, unambiguous, and express” intent to\npreclude habeas jurisdiction over questions of law, the Court\nconcluded that the statutes at issue did not repeal habeas\njurisdiction. Id. at 314.\n\n St. Cyr further illuminates how to approach both\nBoumediene steps. Like Boumediene, St. Cyr looked to the\n1789-era historical application of the writ. St. Cyr also\nlooked to the finality era because it provides evidence of what\n\n 11\n The “finality era” refers to “an approximately sixty-year period\nwhen federal immigration law rendered final (hence, the ‘finality’ era) the\nExecutive’s decisions to admit, exclude, or deport noncitizens. This\nperiod began with the passage of the Immigration Act of 1891, ch. 551,\n26 Stat. 1084, and concluded when Congress enacted the Immigration and\nNationality Act of 1952, Pub. L. No. 82-414, 66 Stat. 163, which\npermitted judicial review of deportation orders through declaratory\njudgment actions in federal district courts.” Castro, 835 F.3d at 436.\n\f THURAISSIGIAM v. USDHS 29\n\ndegree of habeas review is required under the Suspension\nClause and to whom such review is guaranteed in the\nimmigration enforcement context. St. Cyr’s resort to prior\nhabeas cases aligns with Boumediene’s similar reliance on\nEisentrager to resolve ambiguities in the 1789-era application\nof the writ. That St. Cyr ultimately avoided the Suspension\nClause question does not diminish its wisdom or relevance as\nan example of the Court’s analytical approach to Suspension\nClause questions. Consistent with Boumediene and St. Cyr,\nwe conclude that both the common-law history of the writ\nand the Court’s finality era cases are relevant to what and\nwhom the Suspension Clause protects. See also Flores-\nMiramontes v. INS, 212 F.3d 1133, 1141–43 (9th Cir. 2000)\n(relying on common-law history and finality era cases in\naddressing Suspension Clause challenge); see also Trinidad\ny Garcia v. Thomas, 683 F.3d 952, 960 (9th Cir. 2012) (en\nbanc) (Thomas, J., concurring) (discussing finality era cases\nas evidence of rights protected by the Suspension Clause).\n\n C. The Third Circuit’s Decision in Castro\n\n Before addressing Thuraissigiam’s Suspension Clause\nchallenge, we discuss the Third Circuit’s decision in Castro,\nwhich involved an analogous challenge to § 1252(e).12 The\n\n\n 12\n The government’s contention that the Second and Seventh Circuits\nhave addressed the question before us is incorrect. Neither case addresses\nthe Suspension Clause. See Shunaula v. Holder, 732 F.3d 143, 147 (2d\nCir. 2013) (addressing due process challenge to § 1252(e)(2) and\n§ 1252(a)(2)(A)); Khan v. Holder, 608 F.3d 325, 328 (7th Cir. 2010)\n(addressing § 1252(e)(2) in light of that circuit’s “safety valve” doctrine\nfor “judicial correction of bizarre miscarriages of justice”). Likewise, the\ncase cited in the government’s Rule 28(j) letter, Hamama v. Adducci,\n912 F.3d 869 (6th Cir. 2018), does not address the Suspension Clause in\nthe context of the procedures leading up to an expedited removal order.\n\f30 THURAISSIGIAM v. USDHS\n\nThird Circuit concluded that § 1252(e) does not violate the\nSuspension Clause as applied to 28 asylum-seeking families\nwho, like Thuraissigiam, raised constitutional, statutory, and\nregulatory claims relating to their negative credible fear\ndeterminations. 835 F.3d at 425, 428. The families were all\napprehended shortly after entering the country, placed in\nexpedited removal, and found not to have credible fear. Id.\nat 427–28. As we do, the Third Circuit rejected the argument\nthat § 1252(e)(2) provides jurisdiction over claims of legal\nerror. Id. at 434.\n\n Turning to the petitioners’ Suspension Clause challenge,\nthe court opined that the Supreme Court’s habeas cases are\n“perhaps even competing” with the plenary power doctrine.\nId. After reviewing Boumediene and St. Cyr, Castro\ndiscussed the Court’s “commitment to the full breadth of\n[that] doctrine, at least as to aliens at the border seeking\ninitial admission to the country.” Id. at 443. Castro\napproached step one of Boumediene by reference to the\npetitioners’ status in light of Landon v. Plascencia, 459 U.S.\n21 (1982), a case addressing due process, not habeas, rights.\nCastro concluded that petitioners, as “recent surreptitious\nentrants,” should be treated for constitutional purposes as\n“alien[s] seeking initial admission to the United States.”\n835 F.3d at 448. In Landon, the Court stated that such a\nnoncitizen “has no constitutional rights regarding his\napplication” for entry into the country. 459 U.S. at 32.\nAccordingly, the Third Circuit concluded that the petitioners’\nchallenge failed at step one, and did not address whether\n§ 1252(e) was an adequate habeas substitute. 835 F.3d at\n446. The court acknowledged that its discussion of the\npetitioners’ status “appear[ed] to ignore” Supreme Court\nprecedent relating to the due process rights of noncitizens\nphysically present in the country, but concluded that no case\n\f THURAISSIGIAM v. USDHS 31\n\nhad clearly held that “arriving aliens” were entitled to due\nprocess protections. Id. at 447–48.13\n\n We disagree with Castro’s resolution of how Boumediene\nand St. Cyr require us to approach a Suspension Clause\nchallenge. As explained at length above, the Court’s mode of\nanalysis in both of those cases addressed the scope of the\nSuspension Clause by reference to the writ as it stood in 1789\nand relevant habeas corpus precedents. Castro explained that\nit did not rely on St. Cyr’s description of the Court’s habeas\napproach in immigration cases in the finality era by\nemphasizing that, unlike the Castro petitioners, St. Cyr was\na lawful permanent resident, and that St. Cyr discussed only\nwhat the Suspension Clause might protect. Id. at 446.\n\n That St. Cyr did not affirmatively hold that the\nSuspension Clause was violated does not render its\ndescription of the finality era cases incorrect or its approach\nirrelevant. Moreover, Castro’s decision to rely instead on\nLandon is misplaced. Landon held that a permanent resident\nwho traveled abroad and was detained when attempting to\nreenter the United States should be placed in exclusion\n\n\n\n\n 13\n After argument, the Third Circuit decided Osorio-Martinez v.\nAttorney General, 893 F.3d 153 (3d Cir. 2018), involving four juvenile\npetitioners from Castro. After their original habeas petitions were\ndismissed, the juveniles had been granted Special Immigrant Juvenile\n(“SIJ”) status under 8 U.S.C. § 1108(a)(27)(J). Id. at 160. Applying\nCastro, the Third Circuit held that § 1252(e) was an unconstitutional\nsuspension of the writ as applied to the petitioners, by virtue of their\n“significant ties to this country” and the constitutional and statutory rights\nflowing to SIJ designees under 8 U.S.C. § 1255(a) & (h)(1). Id. at 167.\n\f32 THURAISSIGIAM v. USDHS\n\nproceedings rather than deportation. 459 U.S. at 22.14\nAddressing the petitioner’s due process challenge to her\nexclusion proceedings, the Court noted it had “long held that\nan alien seeking initial admission to the United States\nrequests a privilege and has no constitutional rights regarding\nhis application.” Id. As explained by Judge Hardiman, the\nCourt in Landon did not “purport to resolve a jurisdictional\nquestion raising the possibility of an unconstitutional\nsuspension of the writ of habeas corpus”; rather it addressed\nonly the due process rights of a permanent resident. Castro,\n835 F.3d at 450 (Hardiman, J., concurring dubitante); see\nLandon, 459 U.S. at 32–35. Landon could not and did not\naddress the much different question of whether a petitioner\nlike Thuraissigiam may invoke the Suspension Clause.15\n\n Although often conflated, the rights protected by the\nSuspension Clause are not identical to those under the Fifth\nAmendment’s guarantee of due process. See Lee Kovarsky,\n\n\n 14\n At the time, Congress provided that removable noncitizens in the\nUnited States were subject to deportation and those seeking initial entry\nwere subject to exclusion. Id. at 25. Now all noncitizens are subject to\nremoval, whether via 8 U.S.C. § 1225(b)(1) expedited removal or the\nremoval procedures under 8 U.S.C. § 1229a.\n 15\n Regardless, we disagree with the government’s contention and\nCastro’s conclusion that a person like Thuraissigiam lacks all procedural\ndue process rights. See 835 F.3d at 447–48. The Supreme Court has been\nclear that presence matters to due process. See, e.g., Mathews v. Diaz,\n426 U.S. 67, 77 (1976); Zadvydas v. Davis, 533 U.S. 678, 693 (2001).\nAnd we have held that a noncitizen situated almost exactly like\nThuraissigiam had a constitutional right “to expedited removal\nproceedings that conformed to the dictates of due process.” United States\nv. Raya-Vaca, 771 F.3d 1195, 1203 (9th Cir. 2014); see also Immigration\nScholars Amicus Br. (explaining why Thuraissigiam has procedural due\nprocess rights).\n\f THURAISSIGIAM v. USDHS 33\n\nCustodial and Collateral Process: A Response to Professor\nGarrett, 98 Cornell L. Rev. Online 1, 1 (2013) (“Due process\nand the habeas privilege are distinct constitutional\nphenomena, [but] federal courts almost pathologically\nconfuse them.”). It is true that, historically, the Fifth\nAmendment’s due process guarantee and the Suspension\nClause have been applied in tandem, as their applicability was\nrarely disputed. See Mary Van Houten, The Post-\nBoumediene Paradox: Habeas Corpus or Due Process?,\n67 Stan. L. Rev. Online 9, 10 (2014) (observing that these\nprovisions “were almost always jointly applied before\nBoumediene”). But this fact does not mean these rights\nshould be elided, as made clear by the fact that the\nConstitution, ratified two-and-a-half years before the Fifth\nAmendment, see Bute v. People of State of Ill., 333 U.S. 640,\n650 (1948), presupposed the existence of the writ of habeas\ncorpus, see Boumediene, 553 U.S. at 739 (“Protection for the\nprivilege of habeas corpus was one of the few safeguards of\nliberty specified in a Constitution that, at the outset, had no\nBill of Rights.”). Indeed, the writ “is almost the only remedy\nmentioned in the Constitution” as originally ratified. Fallon\n& Meltzer, 120 Harv. L. Rev. at 2037.\n\n Boumediene itself clearly recognized the distinction\nbetween the Fifth Amendment’s due process rights and the\nSuspension Clause – providing further reason not to treat\nLandon’s discussion of due process rights as having any\nbearing on the application of the Suspension Clause. In\nBoumediene, the Court decided that the Guantanamo\ndetainees could invoke the Suspension Clause without\naddressing whether they had due process rights or whether\nthe CSRTs satisfied due process. 553 U.S. at 785; see also id.\nat 739 (starting from the proposition that “protection for the\nprivilege of habeas corpus was one of the few safeguards of\n\f34 THURAISSIGIAM v. USDHS\n\nliberty specified in a Constitution that, at the outset, had no\nBill of Rights[]”); Flores-Miramontes, 212 F.3d at 1142\n(noting that habeas was available at common law prior to the\ndrafting of the Constitution). The Court in Boumediene\ntherefore explicitly declined to link due process rights and\nSuspension Clause rights. See Hamad, 732 F.3d at 999\n(noting that Boumediene did not address whether the due\nprocess clause applied to the Guantanamo detainees); see also\nKiyemba v. Obama, 555 F.3d 1022, 1027 (D.C. Cir. 2009)\n(concluding on habeas review that Guantanamo detainees\nlacked due process rights), vacated by 559 U.S. 131 (2010),\nreinstated by 605 F.3d 1046, 1047 (D.C. Cir. 2010). Landon,\na due process case, is not relevant to whether Thuraissigiam\ncan invoke the Suspension Clause. For that reason, we\ndecline to follow Castro’s approach and reject the\ngovernment’s argument that Thuraissigiam’s purported lack\nof due process rights is determinative of whether he can\ninvoke the Suspension Clause.\n\n Instead, in accordance with Boumediene, we evaluate\nThuraissigiam’s Suspension Clause challenge in two steps:\nFirst, to determine whether Thuraissigiam may invoke the\nSuspension Clause, we examine 1789-era practice, the\nfinality era cases, and other relevant precedents. Second, we\nask whether § 1252(e)(2) provides Thuraissigiam a\n“meaningful opportunity to demonstrate that he is being held\npursuant to ‘the erroneous application or interpretation’ of\nrelevant law.” Boumediene, 553 U.S. at 779. At step two, we\nkeep in mind that the character of the earlier proceedings\nbears on the level of habeas review required. Id. at 781.\n\f THURAISSIGIAM v. USDHS 35\n\nIII. Application\n\n A. Garcia de Rincon and Pena\n\n At the outset, the government contends that our decisions\nin Garcia de Rincon, 539 F.3d 1133, and Pena v. Lynch,\n815 F.3d 452 (9th Cir. 2016), require us to affirm. Although\nin both cases we rejected arguments that § 1252(e)(2)\nauthorized jurisdiction, neither case answered the\nconstitutional question before us today.\n\n In Garcia de Rincon, the petitioner was a noncitizen\nliving in the United States who was stopped at the border\nattempting to reenter after a visit to Mexico, and placed in\nexpedited removal. 539 F.3d at 1135. After rejecting the\npetitioner’s statutory challenge, we dismissed her argument\n– “although . . . not articulated” as such – that the Suspension\nClause required review of her petition. Id. at 1141. The\nprecise question considered was whether “the INA provides\nno adequate substitute for habeas review and therefore\nsuspends the writ” – a Boumediene step two question,\nalthough Garcia de Rincon never addressed Boumediene,\nwhich had been decided months earlier. Id. We concluded\nthat Li v. Eddy, 259 F.3d 1132 (9th Cir. 2001), vacated on\nreh’g as moot, 324 F.3d 1109 (9th Cir. 2003), discredited the\npetitioner’s “generalized due process argument,” the only\nright she sought to vindicate via her petition. Id. Garcia de\nRincon says nothing about whether Thuraissigiam can invoke\nthe Suspension Clause, whether the Clause requires habeas\nreview of statutory or legal claims, or what the Clause\nrequires for a petitioner like Thuraissigiam who is within the\nUnited States. Instead, the case addressed only whether\n§ 1252(e)(2) suspends the writ when a petitioner lacks due\nprocess rights. Put in Boumediene step-two terms, the due\n\f36 THURAISSIGIAM v. USDHS\n\nprocess clause was not “relevant law” for the Garcia de\nRincon petitioner.16\n\n Pena also did not settle the question before us. In Pena,\na noncitizen placed in expedited removal filed a petition for\nreview of the Board of Immigration Appeals’ dismissal of his\nappeal from an IJ’s decision affirming a negative credible\nfear determination. 815 F.3d at 454. Because the petition\nwas not brought under § 1252(e)(2), we concluded that we\nlacked jurisdiction. Id. at 457. We went on to note that in\nWebster v. Doe, 486 U.S. 592, 603 (1988), the Court had\n“suggested that a litigant may be unconstitutionally denied a\nforum when there is absolutely no avenue for judicial review\nof a colorable claim of constitutional deprivation.” 815 F.3d\nat 456 (emphasis in original). We concluded that Pena’s\npetition did not raise Webster concerns because he lacked a\ncolorable constitutional claim,17 and further noted that\n§ 1252(e)(2) provides “some avenues of judicial review.” Id.\nat 456–57. All that Pena says, therefore, is that § 1252(e)\ndoes not implicate the Webster doctrine when a petitioner\nfails to raise colorable constitutional claims. Pena never\naddressed the Suspension Clause.\n\n Because neither Garcia de Rincon nor Pena addressed\nwhether § 1252(e)(2) unlawfully suspends the writ as applied\n\n\n 16\n In case there were any doubt, Smith subsequently reserved the\nquestion of whether, as applied to a noncitizen in expedited removal, the\nSuspension Clause requires review beyond that provided for in\n§ 1252(e)(2). See 741 F.3d at 1022 n.6. That reservation necessarily\ndetermined that Garcia de Rincon had not settled the question.\n 17\n Pena claimed that the IJ violated due process by failing to elicit a\nvoluntary waiver of his right to counsel, but we noted that that claim was\ncontradicted by the record. Id. at 455–56.\n\f THURAISSIGIAM v. USDHS 37\n\nto a petitioner like Thuraissigiam, we reject the government’s\nargument that those cases alone require us to affirm.\n\n B. Reach of the Suspension Clause\n\n At Boumediene step one, we must consider the reach of\nthe Suspension Clause, or, in other words, whether\nThuraissigiam is “barred from seeking the writ or invoking\nthe protections of the Suspension Clause . . . because of [his]\nstatus . . . .” Boumediene, 553 U.S. at 739. In Boumediene,\nthe Court answered this question by reference to its\nprecedents and the common law history of the writ. We\ntherefore do the same.18\n\n\n 18\n As described above, the Court in Boumediene generated a three-\nfactor test at step one in light of the extraterritoriality question presented.\nThis test does not clearly fit in the present case, given that Thuraissigiam\nwas apprehended and detained in the United States. See 553 U.S. at 764,\n766. However, even were we to apply Boumediene’s three-factor test\nhere, it would, as in Boumediene, support application of the Suspension\nClause.\n\n The first factor, Thuraissigiam’s “citizenship and status” and “the\nadequacy of the process through which that status determination was\nmade,” Id. at 766, weighs in favor of applying the Suspension Clause.\nThuraissigiam is a foreign national who contests his status – he contends\nthat he has a credible fear of persecution and therefore qualifies as a\nrefugee entitled to asylum. Like the CRST process at issue in\nBoumediene, the determination as to whether a noncitizen has a credible\nfear is not made via a “rigorous adversarial process to test the legality of\n[his] detention.” Id. at 767. The determination is made by an asylum\nofficer, 8 C.F.R. § 208.30(d), and although the noncitizen may consult\nothers and even have them present a statement at the end of the interview,\n8 C.F.R. § 208.30(d)(4), other hallmarks of the adversarial process are\nlacking. If the noncitizen then chooses to contest an asylum officer’s\nnegative credible fear determination, the noncitizen is entitled only to\ncursory review by an IJ. 8 C.F.R. § 208.30(g); 8 C.F.R. § 1208.39(g)(2).\n\f38 THURAISSIGIAM v. USDHS\n\n As explained, in St. Cyr, the Court canvassed cases from\nEngland and historical accounts to conclude that the writ was\navailable before 1789 to “nonenemy aliens as well as to\ncitizens.” 533 U.S. at 301; accord Boumediene, 553 U.S. at\n747 (“We know that at common law a petitioner’s status as an\n\n\nCritically, unlike in Boumediene, a noncitizen cannot seek review of the\ncredible fear determination in an Article III court. See 8 U.S.C.\n§ 1252(e)(2); cf. Boumediene, 553 U.S. at 767 (“And although the\ndetainee can seek review of his status determination in the Court of\nAppeals, that review process cannot cure all defects in the earlier\nproceedings.”). Accordingly, the procedural protections available to\nThuraissigiam and other noncitizens in expedited removal “fall well short\nof the procedures and adversarial mechanisms that would eliminate the\nneed for habeas corpus review.” Boumediene, 553 U.S. at 767 (internal\nquotation marks omitted).\n\n As to the second factor, there is no question that Thuraissigiam was\napprehended and detained within the sovereign territory of the United\nStates. This factor weighs strongly in favor of finding Thuraissigiam has\nrights under the Suspension Clause. See id. at 768–69. The government\ninsists that the nature and length of a noncitizen’s detention is relevant to\nthis factor. Not so. Boumediene only invokes these considerations under\nstep two. The second factor (under step one) is wholly focused on the\nlevel and duration of control exerted by the United States over the territory\n– which is not at issue here, where the territory is the United States.\nBoumediene, 553 U.S. at 768–69.\n\n As in Boumediene, the third factor is somewhat equivocal: “there are\ncosts to holding the Suspension Clause applicable in a case of [asylum-\nseekers in expedited removal proceedings.]” Id. at 769. But\n“[c]ompliance with any judicial process requires some incremental\nexpenditure of resources,” and direct review by the courts already exists\nand functions in non-expedited removal proceedings. Id. Thus, here, as\nin Boumediene, “[w]hile we are sensitive to [the government’s] concerns,\nwe do not find them dispositive.” Id.\n\n Consequently, Boumediene’s extraterritorial step one factors, if they\nwere relevant here, would support application of the Suspension Clause.\n\f THURAISSIGIAM v. USDHS 39\n\nalien was not a categorical bar to habeas corpus relief.”);\nRasul, 542 U.S. at 481 (“At common law, courts exercised\nhabeas jurisdiction over the claims of aliens detained within\nsovereign territory of the realm. . . .”); see also Gerald L.\nNeuman, Habeas Corpus, Executive Detention, and the\nRemoval of Aliens, 98 Colum. L. Rev. 961, 989–90 (1998)\n(collecting cases).\n\n After the adoption of the Constitution and its Suspension\nClause, courts in the United States applied the same\napproach. For example, in Ex parte D’Olivera, 7 F. Cas. 853\n(C.C.D. Mass. 1813) (No. 3,967), a federal court in\nMassachusetts permitted an arrested noncitizen seaman to\ninvoke habeas. In later years, the Supreme Court continued\nto hold that habeas was available to noncitizens – even\nexcluded noncitizens stopped at the border. United States v.\nJung Ah Lung, 124 U.S. 621, 628–32 (1888); see also\nNeuman, 98 Colum. L. Rev. at 1006. Cases throughout the\nfinality era, from the 1890s to the 1950s, which carry\nsignificant weight here, held firm to this constitutional\npremise. In Nishimura Ekiu v. United States, 142 U.S. 651\n(1892), the Court affirmed that despite the finality law, “[a]n\nalien immigrant, prevented from landing by any such officer\nclaiming authority to do so under an act of congress, and\nthereby restrained of his liberty, is doubtless entitled to a writ\nof habeas corpus to ascertain whether the restraint is lawful.”\nId. at 660.\n\n The Court continued that approach later in the finality era.\nGegiow v. Uhl, 239 U.S. 3, 9 (1915) (“The courts are not\nforbidden by the [finality] statute to consider whether the\nreasons, when they are given, agree with the requirements of\nthe act.”). In Gegiow, for example, the Court reversed the\ngovernment’s legal conclusion that the petitioner was subject\n\f40 THURAISSIGIAM v. USDHS\n\nto exclusion as a public charge based on a lack of labor\nopportunities in his immediate destination. Id. at 9–10; see\nalso Shaughnessy v. United States ex rel. Mezei, 345 U.S.\n206, 212–13 (1953) (stating that even though a noncitizen\nwho had not entered the country lacks due process, he “may\nby habeas corpus test the validity of his exclusion”); United\nStates ex rel. Knauff v. Shaughnessy, 338 U.S. 537, 544\n(1950) (addressing, but rejecting, noncitizen’s “contention\nthat the regulations were not ‘reasonable’ as they were\nrequired to be [under a federal statute]”). In United States ex\nrel. Accardi v. Shaughnessy, the Court granted habeas on\nlegal grounds to a noncitizen who entered from Canada\n“without immigration inspection and without an immigration\nvisa.” 347 U.S. 260, 262, 268 (1954). In Heikkila, the Court\nexplained that the Constitution was the source for habeas\nreview during the finality era, 345 U.S. at 234–35, and in St.\nCyr, the Court clarified that the Suspension Clause was the\nspecific source of such review. See 533 U.S. at 304. Indeed,\nthe government points to no alternative reading.\n\n More broadly, the government offers no convincing\nreason to discount the finality era, nor does it offer a\ncompeting account of the common-law scope of the writ or of\nthe finality era. The government, citing Mezei, 345 U.S. at\n214, answers Boumediene step one by contending that\nThuraissigiam, “as an alien apprehended immediately after\ncrossing the border illegally, is no different from other aliens\nat the border, and is therefore ‘assimilated to [that] status’ for\nconstitutional purposes.” However, Mezei spoke only of such\nassimilation for the purposes of due process, and it otherwise\naffirmed the principle that habeas is available even when a\npetitioner lacks due process rights. Id. at 213. And, crucially,\nBoumediene never linked Suspension Clause rights to due\nprocess rights. The government provides no authority from\n\f THURAISSIGIAM v. USDHS 41\n\nSuspension Clause cases to support its contention that\nThuraissigiam lacks Suspension Clause rights. Because in\nthe finality era the Court permitted even arriving noncitizens\nto invoke habeas review, we conclude that Thuraissigiam,\nwho was arrested within the United States, may invoke the\nSuspension Clause.19\n\n C. Compliance with the Suspension Clause\n\n Having concluded that Thuraissigiam may invoke the\nSuspension Clause, we must consider at Boumediene step two\nwhether habeas review under § 1252(e) is so limited so as\neffectively to suspend the writ as applied to Thuraissigiam.\nAt a minimum, the Suspension Clause “entitles the\n[petitioner] to a meaningful opportunity to demonstrate that\nhe is being held pursuant to ‘the erroneous application or\ninterpretation of relevant law.’” Boumediene, 553 U.S. at 779\n(quoting St. Cyr, 553 U.S. at 302).\n\n Congress may modify the scope of habeas review so long\nas the review “is neither inadequate nor ineffective to test the\nlegality of a person’s detention.” Swain v. Pressley, 430 U.S.\n372, 381 (1977); see Crater v. Galarza, 491 F.3d 1119,\n1124–25 (9th Cir. 2007) (noting, in the context of a challenge\nto AEDPA, that not all restrictions on habeas review\neffectively suspend the writ). We bear in mind that “[a]t its\nhistorical core, the writ of habeas corpus has served as a\n\n\n 19\n In so doing, we reject any argument that only noncitizens who have\n“been lawfully admitted” may invoke the Suspension Clause. Because the\nwrit is an indispensable separation of powers mechanism, “[t]he test for\ndetermining the scope of this provision must not be subject to\nmanipulation by those whose power it is designed to restrain.”\nBoumediene, 553 U.S. at 765–66.\n\f42 THURAISSIGIAM v. USDHS\n\nmeans of reviewing the legality of Executive detention, and\nit is in that context that its protections have been strongest.”\nSt. Cyr, 533 U.S. at 300–01; see also Boumediene, 553 U.S.\nat 783 (noting that in cases of executive detention, “the need\nfor habeas corpus is more urgent”). Therefore, when\nevaluating whether a substitute is adequate, we consider “the\nrigor of any earlier proceedings” and “the intended duration\nof the detention and the reasons for it.” Id. at 781, 783.\n\n The government urges a different approach to step two.\nThe government contends that Thuraissigiam’s status matters\nto the extent of review the Suspension Clause requires. The\ngovernment even suggests we should apply the Boumediene\nstep one extraterritorial factors to determine whether\n§ 1252(e)(2) provides sufficient review. However, those\nfactors have no bearing on step two; only step one considers\nthe petitioner’s status. See Boumediene, 533 U.S. at 773–93\n(considering, without any reference to Guantanamo\ndetainees’ status, whether the DTA was an adequate habeas\nstatute by assessing “the sum total of procedural protections\nafforded to the detainee at all stages, direct and collateral”).\nBoth logically and as applied in Boumediene, the\ncircumstances relevant to step two – the extent of review the\nSuspension Clause requires – are those relating to the\ndetainer, not the detainee. We also reject the government’s\ncontention that because, in its view, Thuraissigiam lacks due\nprocess rights, there are no rights for the Suspension Clause\nto protect. Boumediene foreclosed that argument by holding\nthat, whether or not due process was satisfied, the Suspension\nClause might require more. 553 U.S. at 785.\n\n As a reminder, Thuraissigiam’s petition contends that the\ngovernment denied him a “fair procedure,” “appl[ied] an\nincorrect legal standard” to his credible fear contentions, and\n\f THURAISSIGIAM v. USDHS 43\n\n“fail[ed] to comply with the applicable statutory and\nregulatory requirements.” The core of his claim is that the\ngovernment failed to follow the required procedures and\napply the correct legal standards when evaluating his credible\nfear claim. As Thuraissigiam’s brief states: “Petitioner’s\nclaims merely assert his right to the meaningful credible fear\nprocedure to which he is entitled under the immigration\nstatute, regulations, and Constitution.” We therefore consider\nwhether the Suspension Clause requires review of those\nclaims.20 We conclude that the Clause requires such review\nin Thuraissigiam’s case and that because § 1252(e)(2) fails to\nprovide a meaningful opportunity for such review, it raises\nserious Suspension Clause questions.\n\n At step two, the finality era again informs our analysis of\nwhat the Suspension Clause requires when a removal order is\nchallenged. Finality era precedent establishes that the Court\nregularly reviewed on habeas “claims for statutory as well as\nconstitutional error in deportation proceedings” and “claims\n\n\n 20\n Thuraissigiam’s petition indicates that he might be asking a federal\ncourt to review the agency’s credible fear determination, as he contends\nthat he “can show a significant possibility of prevailing on his claims for\nasylum and other forms of relief.” The government accordingly contends\nthat Thuraissigiam’s petition instead requests “ultimate application of a\nlegal standard to factual determinations and weighing of evidence\nunderlying the Executive’s negative credible-fear findings.” However, we\nread Thuraissigiam’s petition to be explaining why, in his view, DHS’\nprocedural errors matter, particularly given his express assertion that he\nonly wants review of the procedural errors. We therefore do not consider\nhere whether the Suspension Clause requires judicial review of DHS’\ncredible fear determination on the merits. Cf. Ortiz-Alfaro v. Holder,\n694 F.3d 955, 958 (9th Cir. 2012) (noting, based on St. Cyr, that\n“depriving [petitioner] the opportunity for judicial review of a\ndetermination that he lacks a reasonable fear of persecution could raise\nserious constitutional concerns”).\n\f44 THURAISSIGIAM v. USDHS\n\nthat deportation hearings were conducted unfairly.” Flores-\nMiramontes, 212 F.3d at 1143 (citing, inter alia, Fong Haw\nTan v. Phelan, 333 U.S. 6, 8–10 (1948) (interpreting statute\non habeas)); Kessler v. Strecker, 307 U.S. 22, 33–34 (1939)\n(same)). In Gegiow, the Court also reviewed the executive’s\napplication of a legal standard to undisputed facts, concluding\nthat the government had incorrectly determined that the\npetitioner was likely to become a public charge. 239 U.S. at\n9–10. Similarly, we have interpreted the nature of habeas\nreview, encompassing inquiry into “the erroneous application\nor interpretation of statutes,” St. Cyr, 533 U.S. at 302, to\nrequire that “mixed questions of fact and law – those\ninvolving an application of law to undisputed fact . . . be\nprovided meaningful judicial review.” Ramadan v. Gonzales,\n479 F.3d 646, 652 (9th Cir. 2007) (per curiam).\n\n Thuraissigiam and amici point us to other examples of the\nCourt reviewing not just pure legal questions like the one at\nissue in St. Cyr, but also the application of a legal standard to\nundisputed facts. See, e.g., Hansen v. Haff, 291 U.S. 559,\n562–63 (1934) (rejecting government determination that\n“petitioner’s entry was for the purpose” of immoral\nrelations); Mahler v. Eby, 264 U.S. 32, 44 (1924) (holding\nthat the government failed to comply “with all the statutory\nrequirements”). Those cases suggest that the Suspension\nClause requires review of legal and mixed questions of law\nand fact related to removal orders, including expedited\nremoval orders.\n\n As in Boumediene, the decision to place a noncitizen in\nexpedited removal and the finding of whether that noncitizen\nhas a credible fear are both executive determinations,\nmeaning that the requirements of habeas are “more urgent.”\nCf. Boumediene, 553 U.S. at 783. While the duration of\n\f THURAISSIGIAM v. USDHS 45\n\nThuraissigiam’s detention may seem to cut against review,\nthe Court has recognized that an excluded person’s\n“movements are restrained by authority of the United States.”\nMezei, 345 U.S. at 213. Moreover, “it would be difficult to\nsay that [Thuraissigiam] was not imprisoned, theoretically as\nwell as practically, when to turn him back meant that he must\nget into a vessel against his wish and be carried to [Sri\nLanka].” Chin Yow v. United States, 208 U.S. 8, 12 (1908).\nThe finality era cases also demonstrate that habeas is a viable\nmeans of reviewing exclusion and removal orders.\n\n Most important, habeas review provides important\noversight of whether DHS complied with the required\ncredible fear procedures.21 Under the existing administrative\nscheme, there are no rigorous adversarial proceedings prior\nto a negative credible fear determination. First, the credible\nfear interview is initiated only after the CBP officer identifies\na noncitizen who fears persecution and refers that individual\nto a USCIS officer. See 8 C.F.R. § 235.3(b)(4); see also\nRefugee and Human Rights Amicus Br. 11–12. A noncitizen\ncan consult with someone at his own expense before his\nasylum officer interview, but only as long as such\nconsultation does not “unreasonably delay the process and is\nat no expense to the government.” 8 C.F.R. § 208.30(d)(4).\nBefore the IJ hearing, a noncitizen in expedited removal may\n\n\n 21\n Section 1252(e)(2) also restricts judicial oversight of whether the\nagency properly placed a person in expedited removal in the first place:\n“The troubling reality of the expedited removal procedure is that a CBP\nofficer can create the § 1182(a)(7) charge by deciding to convert the\nperson’s status from a non-immigrant with valid papers to an intending\nimmigrant without the proper papers, and then that same officer, free from\nthe risk of judicial oversight, can confirm his or her suspicions of the\nperson’s intentions and find the person guilty of that charge.” Khan,\n608 F.3d at 329.\n\f46 THURAISSIGIAM v. USDHS\n\nagain consult with someone at his own expense, but the\nperiod to obtain such assistance is extremely abbreviated: an\nIJ “shall conclude the review to the maximum extent\npracticable within 24 hours” of the supervisory officer’s\napproval of the asylum officer’s determination. 8 C.F.R.\n§ 1003.42(c), (e). Such review may take place “in person or\nvia telephonic or video connection.” Jaya Ramji, Legislating\nAway International Law: The Refugee Provisions of the\nIllegal Immigration Reform and Immigrant Responsibility\nAct, 37 Stan. J. Int’l L. 117, 134 –41 (2001). There is also no\nrequirement that the IJ provide reasons for her decision.\nIndeed, in this case, the IJ simply checked a box on a form\nstating that the immigration officer’s decision was\n“Affirmed.”\n\n These meager procedural protections are compounded by\nthe fact that § 1252(e)(2) prevents any judicial review of\nwhether DHS complied with the procedures in an individual\ncase, or applied the correct legal standards.22 We think it\nobvious that the constitutional minimum – whether\nThuraissigiam was detained pursuant to the “erroneous\ninterpretation or application of relevant law” – is not satisfied\nby such a scheme.23 Our conclusion is bolstered by the Third\n\n\n 22\n One amicus brief describes reports that the agency does not always\nfollow the required procedures. See Refugee and Human Rights Amicus\nBr. 16–27; see also Michele R. Pistone & John J. Hoeffner, Rules Are\nMade to Be Broken: How the Process of Expedited Removal Fails Asylum\nSeekers, 20 Geo. Immigr. L.J. 167, 175–93 (2006) (describing procedural\nerrors commonly committed during the expedited removal process). If\ntrue, those reports only underscore the need for judicial review.\n 23\n A petitioner’s perceived ultimate desire – as Judge Hardiman put\nit in Castro, to “alter their status in the United States in the hope of\navoiding release to their homelands,” 835 F.3d at 450–51 (Hardiman, J.,\n\f THURAISSIGIAM v. USDHS 47\n\nCircuit’s recent decision in Osorio-Martinez. As Osorio-\nMartinez put it, § 1252(e)(2) fails to provide “even [that]\n‘uncontroversial’ baseline of review” required by\nBoumediene. 893 F.3d at 177. Because the statute prevented\nthe district court from considering whether the agency\nlawfully applied the expedited removal statute, it a fortiori\nprecluded review of “the erroneous application or\ninterpretation of relevant law.” Id. (citing Boumediene,\n553 U.S. at 779). So too here, because § 1252(e)(2) prevents\na court from reviewing claims of procedural error relating to\na negative credible fear determination, it precludes review of\nthe agency’s application of relevant law and thus raises\nserious Suspension Clause questions.24 Plenary power\nconcerns cannot in all circumstances overwhelm the\n“fundamental procedural protections of habeas corpus . . . , a\nright of first importance.” Boumediene, 553 U.S. at 798.\n\nIV. The Canon of Constitutional Avoidance\n\n We further decline to interpret § 1252(e)(2) to avoid the\nserious Suspension Clause problems engendered by the\nstatute. The constitutional avoidance canon applies “if an\notherwise acceptable construction of a statute would raise\n\n\nconcurring dubitante) – is not relevant where a petitioner challenges the\nfairness of specific procedures leading to an expedited removal order.\n 24\n Because Thuraissigiam’s petition does not present the question, we\ndo not consider one amicus’ argument that “there is a compelling case for\nallowing habeas courts to review factual challenges to an expedited\nremoval order.” Scholars of Habeas Corpus Law Amicus Br. 18.\n“Generally . . . the court will not consider arguments raised only in amicus\nbriefs.” United States v. Wahchumwah, 710 F.3d 862, 868 (9th Cir. 2013)\n(citation omitted); see also Russian River Watershed Prot. Comm. v. City\nof Santa Rosa, 142 F.3d 1136, 1141 n.1 (9th Cir. 1998).\n\f48 THURAISSIGIAM v. USDHS\n\nserious constitutional problems, and where an alternative\ninterpretation of the statute is ‘fairly possible.’” St. Cyr,\n533 U.S. at 299–300 (citation omitted); see also Ramadan,\n479 F.3d at 654 (“The Supreme Court has been careful to\nconstrue statutes in light of the Suspension Clause.”).\nHowever, for us to apply the canon, the statute in question\nmust be “susceptible of more than one construction.” Clark\nv. Martinez, 543 U.S. 371, 385 (2005).\n\n As explained at length above, we and other courts have\nconsistently interpreted § 1252(e)(2) to foreclose review of\nclaims like Thuraissigiam’s. Section 1252(a)(2)(A)(i)\nprecludes review of “any other cause or claim arising from or\nrelating to the implementation of or operation of” an\nexpedited removal order, which clearly bars claims relating\nto procedural error. We do not think the statute can bear a\nreading that avoids the constitutional problems it creates.\n\n • ! •\n\n Therefore, we hold that § 1252(e)(2) violates the\nSuspension Clause as applied to Thuraissigiam, although we\ndo not profess to decide in this opinion what right or rights\nThuraissigiam may vindicate via use of the writ. The district\ncourt has jurisdiction and, on remand, should exercise that\njurisdiction to consider Thuraissigiam’s legal challenges to\nthe procedures leading to his expedited removal order.\n\n The judgment of the district court is REVERSED and the\ncase is REMANDED for further proceedings consistent with\nthis opinion.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Vijayakumar Thuraissigiam v. Usdhs\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-07\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nVIJAYAKUMAR THURAISSIGIAM, No. 18-55313\\n Petitioner-Appellant,\\n D.C. No.\\n v. CV 18-135 AJB\\n\\nU.S. DEPARTMENT OF HOMELAND\\nSECURITY; U.S. CUSTOMS AND\\nBORDER PROTECTION; U.S. OPINION\\nCITIZENSHIP AND IMMIGRATION\\nSERVICES; U.S. IMMIGRATION AND\\nCUSTOMS ENFORCEMENT; KIRSTJEN\\nNIELSEN, Secretary of DHS;\\nWILLIAM P. BARR, Attorney\\nGeneral; KEVIN K. MCALEENAN,\\nActing Commissioner of CBP;\\nTHOMAS HOMAN; L. FRANCIS\\nCISSNA, Director of USCIS; PETE\\nFLORES, San Diego Field Director,\\nCBP; GREGORY ARCHAMBEAULT,\\nSan Diego Field Office Director,\\nICE; FRED FIGUEROA, Warden, Otay\\nMesa Detention Center,\\n Respondents-Appellees.\\n\\n\\n Appeal from the United States District Court\\n for the Southern District of California\\n Anthony J. Battaglia, District Judge, Presiding\\n\\f2 THURAISSIGIAM v. USDHS\\n\\n Argued and Submitted May 17, 2018\\n Portland, Oregon\\n\\n Filed March 7, 2019\\n\\n Before: A. Wallace Tashima, M. Margaret McKeown,\\n and Richard A. Paez, Circuit Judges.\\n\\n Opinion by Judge Tashima\\n\\n\\n SUMMARY*\\n\\n\\n Immigration\\n\\n In an action in which Vijayakumar Thuraissigiam filed a\\nhabeas petition to challenge procedures leading to his\\nexpedited removal order, the panel reversed the district\\ncourt’s dismissal of the petition for lack of subject matter\\njurisdiction, held that 8 U.S.C. § 1252(e)(2) violates the\\nSuspension Clause as applied to Thuraissigiam, and\\nremanded.\\n\\n Under 8 U.S.C. § 1225(b)(1)(A)(i), when a U.S. Customs\\nand Border Protection (“CBP”) officer determines that a\\nnoncitizen arriving at a port of entry is inadmissible for\\nmisrepresenting a material fact or lacking necessary\\ndocumentation, the officer must place the noncitizen in so-\\ncalled “expedited removal” proceedings. Expedited removal\\nalso applies to inadmissible noncitizens arrested within 100\\n\\n *\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f THURAISSIGIAM v. USDHS 3\\n\\nmiles of the border and unable to prove that they have been\\nin the United States for more than the prior two weeks.\\n\\n The Department of Homeland Security (“DHS”) removes\\nnoncitizens eligible for expedited removal without further\\nhearing or review, subject to only one exception: If, in an\\ninterview with a CBP officer, the noncitizen indicates an\\nintent to apply for asylum or a fear of persecution, DHS must\\nrefer the noncitizen for an interview with an asylum officer.\\nIf the asylum officer finds no credible fear of persecution, the\\nnoncitizen will be removed. A supervisor reviews the asylum\\nofficer’s credible fear determination, and the noncitizen may\\nalso request de novo review by an immigration judge.\\n\\n Congress sharply circumscribed judicial review of the\\nexpedited removal process such that “no court shall have\\njurisdiction to review . . . any individual determination [or]\\n. . . the application of [§ 1225(b)(1)] to individual aliens”\\noutside of the review permitted by the habeas review\\nprovision, § 1252(e). 8 U.S.C. § 1252(a)(2)(A)(iii). Under\\n§ 1252(e)(2), a person in expedited removal proceedings may\\nfile a habeas petition in federal district court to contest three\\nDHS determinations: whether the person is a noncitizen,\\nwhether he “was ordered removed” via expedited removal,\\nand whether he is a lawful permanent resident or has another\\nstatus exempting him from expedited removal.\\n\\n Thuraissigiam is a native and citizen of Sri Lanka and a\\nTamil, an ethnic minority group in Sri Lanka. After crossing\\ninto the United States, he was arrested 25 yards north of the\\nMexican border, and placed in expedited removal\\nproceedings. He was referred for a credible fear interview\\nafter he indicated a fear of persecution in Sri Lanka, but an\\nasylum officer determined that Thuraissigiam had not\\n\\f4 THURAISSIGIAM v. USDHS\\n\\nestablished a credible fear of persecution. A supervisor\\napproved the decision, and an immigration judge affirmed the\\nnegative credible fear finding in a check-box decision and\\nreturned the case to DHS for Thuraissigiam’s removal.\\n\\n Thuraissigiam filed a habeas petition in federal district\\ncourt, arguing that his expedited removal order violated his\\nstatutory, regulatory, and constitutional rights. The district\\ncourt dismissed the petition for lack of subject matter\\njurisdiction, concluding that § 1252(e) did not authorize\\njurisdiction over Thuraissigiam’s claims and rejecting his\\nSuspension Clause arguments.\\n\\n The panel concluded, in line with this court’s precedents,\\nthat § 1252(e)(2) does not authorize jurisdiction over\\nThuraissigiam’s petition because § 1252(e)(2) limits a district\\ncourt to reviewing three basic factual determinations related\\nto an expedited removal order, and Thuraissigiam’s petition\\ndoes not challenge any of those determinations.\\n\\n The panel next considered whether the habeas review\\navailable to Thuraissigiam under § 1252(e)(2) satisfied the\\nrequirements of the Suspension Clause. In doing so, the\\npanel observed that Boumediene v. Bush, 553 U.S. 723\\n(2008), provides an analytical template for evaluating a\\nSuspension Clause challenge: at step one, the court examines\\nwhether the Suspension Clause applies to the petitioner; and,\\nif so, at step two, the court examines whether the substitute\\nprocedure provides review that satisfies the Clause.\\n\\n The panel also observed that INS v. St. Cyr, 533 U.S. 289,\\n302 (2001), illuminates how to approach both Boumediene\\nsteps. The panel explained that, like Boumedine, St. Cyr\\nlooked to the 1789-era historical application of the writ of\\n\\f THURAISSIGIAM v. USDHS 5\\n\\nhabeas corpus, but that St. Cyr also looked to habeas\\nprecedents from the co-called “finality era,” a period from\\n1891 to 1952 during which the statutory scheme precluded\\njudicial intervention in immigration enforcement, except as\\nrequired by the Constitution.\\n\\n Drawing from Boumedine and St. Cyr, the panel\\nconcluded that it must evaluate Thuraissigiam’s Suspension\\nClause challenge in two steps: First, to determine whether\\nThuraissigiam may invoke the Suspension Clause, the court\\nexamines 1789-era practice, the finality era cases, and other\\nrelevant precedents. Second, the court asks whether\\n§ 1252(e)(2) provides Thuraissigiam a “meaningful\\nopportunity to demonstrate that he is being held pursuant to\\n‘the erroneous application or interpretation’ of relevant law.”\\nBoumediene, 553 U.S. at 779 (quoting St. Cyr, 533 U.S. at\\n302).\\n\\n In applying Boumediene step one, which concerns the\\nreach of the Suspension Clause, the panel concluded that\\nThuraissigiam, who was arrested within the United States,\\nmay invoke the Suspension Clause. In so concluding, the\\npanel observed that, in the finality era, the Court permitted\\neven arriving noncitizens to invoke habeas review.\\n\\n The panel considered at Boumediene step two whether\\nhabeas review under § 1252(e) is so limited so as effectively\\nto suspend the writ as applied to Thuraissigiam. In doing so,\\nthe panel rejected the government’s contention that because,\\nin its view, Thuraissigiam lacks due process rights, there are\\nno rights for the Suspension Clause to protect, explaining that\\nBoumediene foreclosed that argument by holding that,\\nwhether or not due process was satisfied, the Suspension\\nClause might require more. Reviewing the relevant\\n\\f6 THURAISSIGIAM v. USDHS\\n\\nprecedent, the panel concluded that the Suspension Clause\\nrequires review of Thuraissigiam’s claims.\\n\\n The panel held that 8 U.S.C. § 1252(e)(2) violates the\\nSuspension Clause as applied to Thuraissigiam. Noting the\\nmeager procedural protections afforded by the administrative\\nscheme governing credible fear determinations in this\\ncontext, and the fact that § 1252(a)(2) prevents any judicial\\nreview of whether DHS complied with the procedures in an\\nindividual case or applied the correct standard, the panel\\nconcluded that it is obvious that the constitutional minimum\\n– whether Thuraissigiam was detained pursuant to the\\n“erroneous interpretation or application of relevant law” – is\\nnot satisfied by the scheme set out by § 1252(e)(2).\\n\\n The panel further declined to interpret § 1252(e)(2) to\\navoid the serious Suspension Clause problems engendered by\\nthe statute, explaining that the constitutional avoidance canon\\ndoes not apply because the statute cannot bear a reading that\\navoids the constitutional problems it creates.\\n\\n Finally, the panel observed that it did not profess to\\ndecide in this opinion what right or rights Thuraissigiam may\\nvindicate via use of the writ, and remanded the case for the\\ndistrict court to exercise jurisdiction to consider\\nThuraissigiam’s legal challenges to the procedures leading to\\nhis expedited removal order.\\n\\f THURAISSIGIAM v. USDHS 7\\n\\n COUNSEL\\n\\nLee P. Gelernt (argued) and David Hausman, American Civil\\nLiberties Union Foundation Immigrants’ Rights Project, New\\nYork, New York; Jennifer Chang Newell and Cody Wofsy,\\nAmerican Civil Liberties Union Foundation Immigrants’\\nRights Project, San Francisco, California; David Loy, ACLU\\nFoundation of San Diego & Imperial Counties, San Diego,\\nCalifornia; for Petitioner-Appellant.\\n\\nJoshua S. Press (argued), and Joseph A. Darrow, Trial\\nAttorneys; Erez Reuveni, Assistant Director; William C.\\nPeachey, Director; Office of Immigration Litigation, Civil\\nDivision, United States Department of Justice, Washington,\\nD.C.; for Respondent-Appellees.\\n\\nMatthew E. Price, Jenner & Block, New York, New York;\\nBlaine Bookey, Karen Musalo, and Eunice Lee, Center for\\nGender & Refugee Studies, University of California Hastings\\nCollege of the Law, San Francisco, Calfornia; for Amici\\nCuriae Refugee and Human Rights Organizations and\\nScholars.\\n\\nEthan D. Dettmer, Soolean Choy, Priyah Kaul, and Eli M.\\nLazarus, Gibson Dunn & Crutcher, San Francisco, California;\\nJoshua S. Lipshutz, Gibson Dunn & Crutcher, Washington,\\nD.C.; for Amici Curiae Scholars of Immigration Law.\\n\\nNoah A. Levine, Wilmer Cutler Pickering Hale & Dorr, New\\nYork, New York, for Amici Curiae Scholars of Habeas\\nCorpus Law.\\n\\f8 THURAISSIGIAM v. USDHS\\n\\nAnjali Srinivasan and Leo L. Lam, Keker Van Nest & Peters,\\nSan Francisco, California, for Amici Curiae Scholars of Sri\\nLankan Politics.\\n\\n\\n OPINION\\n\\nTASHIMA, Circuit Judge:\\n\\n Vijayakumar Thuraissigiam filed a habeas petition in\\ndistrict court pursuant to 8 U.S.C. § 1252(e)(2) to challenge\\nthe procedures leading to his expedited removal order. The\\ncourt dismissed the petition for lack of subject matter\\njurisdiction. We reverse. Although § 1252(e)(2) does not\\nauthorize jurisdiction over the claims in Thuraissigiam’s\\npetition, the Suspension Clause, U.S. Const. art. I, § 9, cl. 2,\\nrequires that Thuraissigiam have a “meaningful opportunity\\nto demonstrate that he is being held pursuant to ‘the\\nerroneous application or interpretation’ of relevant law.”\\nBoumediene v. Bush, 553 U.S. 723, 779 (2008) (quoting INS\\nv. St. Cyr, 533 U.S. 289, 302 (2001)). Because § 1252(e)(2)\\ndoes not provide that meaningful opportunity, the provision\\nviolates the Suspension Clause as applied to Thuraissigiam.\\n\\n BACKGROUND\\n\\nI. Statutory Background\\n\\n When a U.S. Customs and Border Protection (“CBP”)\\nofficer determines that a noncitizen arriving at a port of entry\\nis inadmissible for misrepresenting a material fact or lacking\\n\\f THURAISSIGIAM v. USDHS 9\\n\\nnecessary documentation,1 the officer must place the\\nnoncitizen in so-called “expedited removal” proceedings.\\n8 U.S.C. § 1225(b)(1)(A)(i). By regulation, the Department\\nof Homeland Security (“DHS”), of which CBP is a\\nconstituent agency, also applies expedited removal to\\ninadmissible noncitizens arrested within 100 miles of the\\nborder and unable to prove that they have been in the United\\nStates for more than the prior two weeks. Designating Aliens\\nfor Expedited Removal, 69 Fed. Reg. 48877-01, 48879-80\\n(Aug. 11, 2004);2 see also 8 U.S.C. § 1225(b)(1)(A)(iii)(II).\\n\\n DHS removes noncitizens eligible for expedited removal\\n“without further hearing or review,” subject to only one\\nexception. 8 U.S.C. § 1225(b)(1)(A)(i). If, in an interview\\nwith a CBP officer, the noncitizen indicates an intent to apply\\nfor asylum or a fear of persecution, DHS must refer the\\nnoncitizen for an interview with an asylum officer. Id.\\n§ 1225(b)(1)(A)(ii); 8 C.F.R. § 208.30. If that asylum officer\\ndetermines that the noncitizen’s fear of persecution is\\n\\n\\n 1\\n See 8 U.S.C. § 1182(a)(6)(C) (misrepresentation bar); id.\\n§ 1182(a)(7) (documentation bar).\\n 2\\n Congress gave the Attorney General the authority to extend\\nexpedited removal to some or all inadmissible noncitizens who cannot\\nprove that they have been in the United States for more than two years\\nprior; thus, the current regime does not represent the full exercise of\\nexecutive authority permitted by statute. 8 U.S.C. § 1225(b)(1)(A)(iii).\\nDHS also applies expedited removal to noncitizens who entered the\\nUnited States by sea and who have not been in the United States for two\\nyears. See Notice Designating Aliens Subject to Expedited Removal\\nUnder Section 235(b)(1)(A)(iii) of the Immigration and Nationality Act,\\n67 Fed. Reg. 68,924, 68,924–25 (Nov. 13, 2002). The current regime\\nmay, however, expand; a January 2017 executive order instructs the\\nSecretary of DHS to apply expedited removal to the fullest extent of the\\nlaw. See Exec. Order No. 13,767, 82 Fed. Reg. 8793 (Jan. 25, 2017).\\n\\f10 THURAISSIGIAM v. USDHS\\n\\ncredible, the noncitizen is referred to non-expedited removal\\nproceedings, in which the noncitizen may apply for asylum or\\nother forms of relief from removal. See 8 U.S.C.\\n§ 1225(b)(1)(B)(ii); 8 C.F.R. § 208.30(f); 8 C.F.R.\\n§ 1003.42(f). If the asylum officer finds no credible fear of\\npersecution, the noncitizen will be removed. 8 U.S.C.\\n§ 1225(b)(1)(B)(iii). A supervisor reviews the asylum\\nofficer’s credible fear determination, 8 C.F.R.\\n§§ 208.30(e)(7), 235.3(b)(2), (b)(7), and a noncitizen may\\nalso request de novo review by an immigration judge (“IJ”).\\n8 U.S.C. § 1225(b)(1)(B)(iii)(III); 8 C.F.R. § 1003.42. In\\n2016, DHS conducted over 141,000 expedited removals. See\\nRefugee and Human Rights Amicus Br. 10. All individuals\\nplaced in expedited removal proceedings are subject to\\nmandatory detention pending a final determination of credible\\nfear of persecution or until they are removed. 8 U.S.C.\\n§ 1225(b)(1)(B)(iii)(IV).\\n\\n Congress sharply circumscribed judicial review of the\\nexpedited removal process. “[N]o court shall have\\njurisdiction to review . . . any individual determination [or]\\n. . . the application of [§ 1225(b)(1)] to individual aliens”\\noutside of the review permitted by the habeas review\\nprovision, § 1252(e). 8 U.S.C. § 1252(a)(2)(A)(iii). Under\\n§ 1252(e)(2), a person in expedited removal proceedings may\\nfile a habeas petition in federal district court to contest three\\nDHS determinations: whether the person is a noncitizen,\\nwhether he “was ordered removed” via expedited removal,\\nand whether he is a lawful permanent resident or has another\\nstatus exempting him from expedited removal. Id.\\n§ 1252(e)(2)(A)–(C). Review of whether a petitioner “was\\nordered removed” is “limited to whether such an order in fact\\nwas issued and whether it relates to the petitioner. Id.\\n§ 1252(e)(5). “There shall be no review of whether the alien\\n\\f THURAISSIGIAM v. USDHS 11\\n\\nis actually inadmissible or entitled to any relief from\\nremoval.” Id.; see also 8 C.F.R. § 1003.42(f) (“No appeal\\nshall lie from a review of an adverse credible fear\\ndetermination made by an immigration judge.”).3\\n\\nII. Factual Background\\n\\n Thuraissigiam is a native and citizen of Sri Lanka and a\\nTamil, an ethnic minority group in Sri Lanka. See Scholars\\nof Sri Lankan Politics Amicus Br. 3. Thuraissigiam fled his\\nhome country in June 2016 and made his way to Mexico. On\\nFebruary 17, 2017, Thuraissigiam crossed the border into the\\nUnited States. Late that night, he was arrested by a CBP\\nofficer four miles west of the San Ysidro, California, port of\\nentry, 25 yards north of the border.\\n\\n DHS placed Thuraissigiam in expedited removal\\nproceedings. Pursuant to 8 U.S.C. § 1225(b)(1)(A)(ii), CBP\\nreferred Thuraissigiam for an interview with an asylum\\nofficer after he indicated a fear of persecution in Sri Lanka.\\nOn March 9, an asylum officer from the United States\\nCitizenship and Immigration Services (“USCIS”) interviewed\\nThuraissigiam and determined that he had not established a\\ncredible fear of persecution. A supervisor approved the\\n\\n\\n 3\\n Under 8 U.S.C. § 1252(e)(3), a person may challenge the\\nconstitutionality and legality of the expedited removal provisions,\\nregulations implementing those provisions, or written policies to\\nimplement the provisions. Such challenges, however, must be brought\\nwithin 60 days after implementation and only in the District of Columbia.\\nId. § 1252(e)(3)(A)–(B). Various expedited removal provisions and\\nimplementing regulations survived a § 1252(e)(3) challenge in American\\nImmigration Lawyers Ass’n v. Reno, 18 F. Supp. 2d 38 (D.D.C. 1998),\\nalthough the plaintiffs did not raise a Suspension Clause argument about\\nthe extent of habeas review. See id. at 41.\\n\\f12 THURAISSIGIAM v. USDHS\\n\\ndecision. Thuraissigiam then requested review by an IJ, who\\naffirmed the negative credible fear finding in a check-box\\ndecision and returned the case to DHS for Thuraissigiam’s\\nremoval.\\n\\nIII. District Court Proceedings\\n\\n In January 2018, Thuraissigiam filed a habeas petition in\\nfederal district court, naming as respondents DHS, several of\\nits constituent agencies, and individual agency officials.\\nThuraissigiam argued that his “expedited removal order\\nviolated his statutory, regulatory, and constitutional rights,”\\nsought to vacate the order, and requested relief in the form of\\na “new, meaningful opportunity to apply for asylum and other\\nrelief from removal.” Thuraissigiam alleged that in Sri Lanka\\nhe had been harassed for supporting a Tamil political\\ncandidate. In 2007, he was “detained and beaten” by Sri\\nLankan army officers, and told not to support the candidate.\\nIn 2014, after Thuraissigiam continued to support the\\ncandidate, government intelligence officers kidnapped,\\nbound, and beat him during an interrogation about his\\npolitical activities. Thuraissigiam alleged that he “was\\nlowered into a well, simulating drowning, threatened with\\ndeath, and then suffocated, causing him to lose\\nconsciousness.”\\n\\n Thuraissigiam also made various factual allegations about\\nthe expedited removal procedures to which he was subject\\nafter being apprehended. For one, he alleged that the asylum\\nofficer failed to “elicit all relevant and useful information\\nbearing on whether the applicant has a credible fear of\\npersecution or torture” in violation of 8 C.F.R. § 208.30(d)\\nand “failed to consider relevant country conditions evidence”\\nin violation of 8 U.S.C. § 1225(b)(1)(B)(v) and 8 C.F.R.\\n\\f THURAISSIGIAM v. USDHS 13\\n\\n§ 208.30(e)(2). Thuraissigiam also alleged that there were\\n“communication problems” between the asylum officer,\\nThuraissigiam, and the translator, in violation of 8 C.F.R.\\n§ 208.30(d)(1)–(2). Thuraissigiam alleged that the IJ hearing\\nincluded the same procedural and substantive flaws, and that\\nat both hearings, he was unaware whether “information he\\noffered would be shared with the Sri Lankan government.”\\nThuraissigiam’s petition asserted two causes of action:\\n\\n First, DHS’ credible fear screening deprived\\nThuraissigiam “of a meaningful right to apply for asylum”\\nand other forms of relief, in violation of 8 U.S.C.\\n§ 1225(b)(1), its implementing regulations, and the United\\nStates Convention Against Torture, implemented in the\\nForeign Affairs Reform and Restructuring Act of 1998\\n(“FARRA”), Pub. L. No. 105-277, div. G., Title XXII,\\n§ 2242, 112 Stat. 2681 (1998). The asylum officer and IJ also\\nviolated those statutes “by applying an incorrect legal\\nstandard” to Thuraissigiam’s credible fear application.\\n\\n Second, the asylum officer and IJ violated\\nThuraissigiam’s rights under the Due Process Clause of the\\nFifth Amendment by “not providing him with a meaningful\\nopportunity to establish his claims, failing to comply with the\\napplicable statutory and regulatory requirements, and in not\\nproviding him with a reasoned explanation for their\\ndecisions.”\\n\\n The district court dismissed the habeas petition for lack of\\nsubject matter jurisdiction. Thuraissigiam v. U.S. Dep’t of\\nHomeland Sec., 287 F. Supp. 3d 1077 (S.D. Cal. 2018).\\nRelying on our precedents, the district court concluded that\\n8 U.S.C. § 1252(e) did not authorize jurisdiction over the\\nclaims in Thuraissigiam’s petition. Id. at 1082. Next, the\\n\\f14 THURAISSIGIAM v. USDHS\\n\\ncourt rejected Thuraissigiam’s Suspension Clause arguments.\\nAlthough the court concluded that Thuraissigiam could\\ninvoke the Suspension Clause, it held that the statute’s “strict\\nrestraints” on habeas review of expedited removal orders did\\nnot effectively suspend the writ of habeas corpus and were\\ntherefore constitutionally sound. Id. at 1082–83.4\\n\\n Thuraissigiam timely appealed the district court’s\\ndismissal and moved for a stay of removal pending appeal.\\nA motions panel of our court initially denied Thuraissigiam’s\\nstay motion, but later vacated that order and stayed\\nThuraissigiam’s removal pending appeal.\\n\\n STANDARD OF REVIEW\\n\\n We have jurisdiction under 28 U.S.C. § 1291. We review\\nde novo the district court’s dismissal of Thuraissigiam’s\\nhabeas petition for lack of subject matter jurisdiction. Bishop\\nPaiute Tribe v. Inyo County, 863 F.3d 1144, 1151 (9th Cir.\\n2017); see also Garcia de Rincon v. Dep’t of Homeland Sec.,\\n539 F.3d 1133, 1136 (9th Cir. 2008).\\n\\n DISCUSSION\\n\\n We must decide whether a federal district court has\\njurisdiction to review the claims in Thuraissigiam’s petition.\\nWe first inquire whether 8 U.S.C. § 1252(e)(2) authorizes\\njurisdiction over Thuraissigiam’s petition. Concluding that\\n§ 1252(e)(2) does not authorize jurisdiction, we then address\\n\\n\\n\\n 4\\n The district court also denied various stay motions that\\nThuraissigiam had filed, concluding that they were moot due to the\\npetition’s dismissal. 287 F. Supp. 3d at 1078.\\n\\f THURAISSIGIAM v. USDHS 15\\n\\nwhether the provision restricting habeas review violates the\\nSuspension Clause.\\n\\nI. Jurisdiction Under 8 U.S.C. § 1252(e)(2)\\n\\n Thuraissigiam contends that 8 U.S.C. § 1252(e)(2)(B)\\nauthorizes review of the statutory, regulatory, and\\nconstitutional claims raised in his habeas petition. We\\ndisagree. Section 1252(e)(2), including Subsection (B), limits\\na district court to reviewing three basic factual determinations\\nrelated to an expedited removal order. Because\\nThuraissigiam’s petition does not challenge any of those\\ndeterminations, § 1252(e)(2) does not authorize jurisdiction\\nover the petition.\\n\\n A court applying habeas review under § 1252(e)(2) is\\nlimited to determining:\\n\\n (A) whether the petitioner is an alien,\\n\\n (B) whether the petitioner was ordered\\n removed under such section, and\\n\\n (C) whether the petitioner can prove by a\\n preponderance of the evidence that the\\n petitioner is an alien lawfully admitted for\\n permanent residence, has been admitted as a\\n refugee under section 1157 of this title, or has\\n been granted asylum under section 1158 of\\n this title . . . .\\n\\nCongress also provided express limitations on review under\\nSubsection (B):\\n\\f16 THURAISSIGIAM v. USDHS\\n\\n In determining whether an alien has been\\n ordered removed under section 1225(b)(1) of\\n this title, the court’s inquiry shall be limited to\\n whether such an order in fact was issued and\\n whether it relates to the petitioner. There\\n shall be no review of whether the alien is\\n actually inadmissible or entitled to any relief\\n from removal.\\n\\nId. § 1252(e)(5). Nonetheless, Thuraissigiam stakes his claim\\non Subsection (B).\\n\\n We considered and rejected a nearly identical argument\\nin Garcia de Rincon. The petitioner in Garcia de Rincon\\nraised a due process challenge to an expedited removal order.\\n539 F.3d at 1136. Characterizing § 1252(e) as among the\\n“most stringent” jurisdiction-limiting provisions in the\\nimmigration statutes, we held that § 1252(e)(2) permits\\nreview only of “habeas petitions alleging that the petitioner\\nis not an alien or was never subject to an expedited removal\\norder.” Id. at 1135, 1139. We therefore lacked jurisdiction\\nbecause the petitioner’s due process claims were not\\nencompassed by those enumerated grounds. Id. at 1139.\\nLikewise, Thuraissigiam’s claims of procedural violations are\\nplainly not claims about whether Thuraissigiam “was never\\nsubject to an expedited removal order.” 8 U.S.C. § 1252(e).\\n\\n Thuraissigiam contends that such a reading of Subsection\\n(B) renders superfluous the prohibition in § 1252(e)(5)\\nagainst “review of whether the alien is actually inadmissible\\nor entitled to any relief from removal.” However,\\n§ 1252(e)(5) functions not to repeat § 1252(e)(2)(B), but to\\nexplain it. Moreover, Thuraissigiam’s petition is barred by\\nthe first sentence in § 1252(e)(5), not the second sentence.\\n\\f THURAISSIGIAM v. USDHS 17\\n\\nBecause he asks the district court to review the government’s\\nprocedures, those claims are beyond the scope of “whether\\nsuch an [expedited removal] order in fact was issued and\\nwhether it relates to the petitioner.”5 8 U.S.C. § 1252(e); see\\nalso United States v. Barajas-Alvarado, 655 F.3d 1077, 1082\\n(9th Cir. 2011) (reaffirming that jurisdiction under\\n§ 1252(e)(2) “does not extend to review of the claim that an\\nalien was wrongfully deprived of the administrative review\\npermitted under the statute and applicable regulations”).\\n\\n Thuraissigiam relies on Smith v. U.S. Customs & Border\\nProtection, 741 F.3d 1016 (9th Cir. 2014), to contend that we\\nhave adopted a more expansive view of Subsection (B).\\nSpecifically, he contends that Smith reviewed “whether the\\npetitioner belonged in the expedited removal system,” and\\nthat a court may thus review his petition. Smith, however,\\ndoes not support Thuraissigiam’s argument. In Smith, CBP\\nplaced the petitioner, a Canadian citizen arriving at the\\nborder, in expedited removal proceedings for lacking certain\\ndocuments. Id. at 1019. The petitioner alleged that CBP\\nexceeded its authority under the expedited removal statute\\nbecause certain document requirements are waived for\\nCanadians, and argued that Subsection (B) authorized review.\\nId. at 1019, 1021. “Accepting [petitioner’s] theory at face\\nvalue,” we reviewed whether CBP in fact classified him as an\\n“intending immigrant.” Id. at 1021. Concluding that CBP had\\ndone so, we held that § 1252(e)(2) “permit[ted] us to go no\\n\\n\\n 5\\n Thuraissigiam also makes a structural argument, contending that\\nbecause Congress provided for some review of asylum claims even in\\nexpedited removal cases, Congress must not have intended to strip judicial\\nreview to “police the boundaries of those limits.” This argument ignores\\nthe plain language of the statute, which evidences Congress’ intent to do\\njust that.\\n\\f18 THURAISSIGIAM v. USDHS\\n\\nfurther” and did not discuss the merits of CBP’s\\nclassification. Id. at 1021–22 & n.4.6 Therefore, Smith\\nreviewed only how CBP classified the petitioner, which is\\nfairly encompassed by whether “[the petitioner] was ordered\\nremoved” under the expedited removal provision. Id. at 1022\\n(internal quotation marks omitted). By contrast,\\nThuraissigiam asks the district court to pass judgment on the\\nprocedures leading to his removal order. The limited review\\nprovided under § 1252(e)(2) does not encompass such\\nclaims.7\\n\\n Therefore, in line with our precedents, we conclude that\\n§ 1252(e) does not authorize habeas review of\\nThuraissigiam’s petition. We do not here address\\nThuraissigiam’s request that we apply the canon of\\nconstitutional avoidance to interpret § 1252(e) to provide\\njurisdiction over his legal claims. That canon only comes\\ninto play if we conclude that § 1252(e) raises serious\\nconstitutional questions; thus, we first address\\nThuraissigiam’s Suspension Clause argument before\\ncontemplating the application of that canon. See St. Cyr,\\n533 U.S. at 299–300, 314 (explaining constitutional\\navoidance canon and applying it upon concluding that the\\nstatute in question raised serious constitutional questions).\\n\\n\\n 6\\n The Smith petitioner also contended that § 1252(e)(2) violated the\\nSuspension Clause, but we did not reach the argument because we held\\nthat the statute permitted limited review of his petition. 741 F.3d at 1022\\nn.6.\\n 7\\n We have held that in appeals from convictions for criminal reentry,\\na defendant may collaterally attack a removal order that forms the basis\\nfor his conviction. See United States v. Ochoa-Oregel, 904 F.3d 682, 686\\n(9th Cir. 2018) (considering collateral attack on expedited removal order).\\nBut that rule does not apply to this case.\\n\\f THURAISSIGIAM v. USDHS 19\\n\\nII. Suspension Clause\\n\\n The Suspension Clause mandates, “The Privilege of the\\nWrit of Habeas Corpus shall not be suspended, unless when\\nin Cases of Rebellion or Invasion the public Safety may\\nrequire it.” U.S. Const. art. I, § 9, cl. 2. Our nation’s\\nfounders viewed the writ as a “vital instrument” to secure\\nindividual liberty. Boumediene, 553 U.S. at 743. “The\\nClause protects the rights of the detained by a means\\nconsistent with the essential design of the Constitution. It\\nensures that, except during periods of formal suspension, the\\nJudiciary will have a time-tested device, the writ, to maintain\\nthe ‘delicate balance of governance’ that is itself the surest\\nsafeguard of liberty.” Id. at 745 (citing Hamdi v. Rumsfeld,\\n542 U.S. 507, 536 (2004) (opinion of O’Connor, J.)). The\\nSuspension Clause prevents Congress from passing a statute\\nthat effectively suspends the writ absent rebellion or invasion.\\nSee Felker v. Turpin, 518 U.S. 651, 663–64 (1996). Thus, the\\nquestion in this case is whether 8 U.S.C. § 1252(e)(2)\\neffectively suspends the writ. Put another way, the question\\nis whether the habeas review available to Thuraissigiam\\nunder § 1252(e)(2) satisfies the requirements of the\\nSuspension Clause.\\n\\n The Supreme Court has not yet answered that question.\\nIn fact, the Court has rarely addressed who may invoke the\\nSuspension Clause and the extent of review the Clause\\nrequires. For example, only in Boumediene has the Court\\nconcluded that a statute violated the Suspension Clause.\\nGerald L. Neuman, The Habeas Corpus Suspension Clause\\nAfter Boumediene v. Bush, 110 Colum. L. Rev. 537, 538\\n(2010). In the Court’s other most recent Suspension Clause\\ncase, St. Cyr, the Court, after extensive analysis of the\\nSuspension Clause issues at play, interpreted the statute to\\n\\f20 THURAISSIGIAM v. USDHS\\n\\navoid those issues. 533 U.S. at 336–37. Of the federal courts\\nof appeals, only the Third Circuit has addressed the precise\\nquestion before us, whether § 1252(e)(2) as applied to\\nnoncitizen petitioners in expedited removal violates the\\nSuspension Clause. See Castro v. U.S. Dep’t of Homeland\\nSec., 835 F.3d 422 (3d Cir. 2016) (concluding that, due to\\ntheir status, such petitioners could not invoke the Suspension\\nClause), cert. denied, 137 S. Ct. 1581 (2017).\\n\\n Boumediene traced the writ of habeas corpus to its origins\\nas a tool of the English crown, citing the detailed historical\\naccount in Paul D. Halliday and G. Edward White, The\\nSuspension Clause: English Text, Imperial Contexts, and\\nAmerican Implications, 94 Va. L. Rev. 575 (2008). 553 U.S.\\nat 740. As Halliday and White explain, the writ in England\\nwas the vehicle “to determine the rightness of constraints\\nimposed on the bodies of the king’s subjects of all kinds.” 94\\nVa. L. Rev. at 607. The writ was on occasion suspended in\\nEngland. Id. at 619; see Boumediene, 553 U.S. at 741.\\nAccording to Boumediene, that history “no doubt confirmed\\n[the Framers’] view that pendular swings to and away from\\nindividual liberty were endemic to undivided, uncontrolled\\npower.” Id. at 742; see also Amanda L. Tyler, A “Second\\nMagna Carta”: The English Habeas Corpus Act and the\\nStatutory Origins of the Habeas Privilege, 91 Notre Dame L.\\nRev. 1949, 1985–86 (2016) (describing how suspensions of\\nthe writ in colonial America motivated the States’ desire to\\nimport similar habeas protections after gaining\\nindependence); Halliday & White, 94 Va. L. Rev. at 671\\n(highlighting the Framers’ desire to restore “the traditional\\norder of writs and suspensions”).\\n\\n As Boumediene summed it up, the Suspension Clause is\\nrooted in the Framers’ first-hand experience “that the\\n\\f THURAISSIGIAM v. USDHS 21\\n\\ncommon-law writ all too often had been insufficient to guard\\nagainst the abuse of monarchial power.” 553 U.S. at 739–40.\\nThe Clause, therefore, is “not merely about suspending the\\nprivilege of the writ of habeas corpus, but about the meaning\\nof the ‘privilege of the writ’ itself.” Halliday & White,\\n94 Va. L. Rev. at 699. “Indeed, common law habeas corpus\\nwas, above all, an adaptable remedy . . . [whose] precise\\napplication and scope changed depending upon the\\ncircumstances.” Boumediene, 553 U.S. at 779–80 (citing,\\ninter alia, Richard H. Fallon, Jr. & Daniel J. Meltzer, Habeas\\nCorpus Jurisdiction, Substantive Rights, and the War on\\nTerror, 120 Harv. L. Rev. 2029, 2102 (2007)).\\n\\n In examining how the Supreme Court has defined the\\nSuspension Clause’s requirements, Boumediene is our\\nstarting point, even if it does not provide a direct answer to\\nThuraissigiam’s challenge. Boumediene and its predecessors,\\nlike St. Cyr, do provide an analytical blueprint. We therefore\\nreview those precedents before deciding how best to apply\\ntheir principles to this appeal.\\n\\n A. Boumediene v. Bush\\n\\n In Boumediene, the Supreme Court struck down a War on\\nTerror-era law after detainees at the Guantanamo Bay prison\\nin Cuba brought a Suspension Clause challenge. 553 U.S. at\\n732–33. In the wake of the September 11, 2001, attacks, the\\nU.S. Department of Defense created Combatant Status\\nReview Tribunals (“CSRTs”) to decide if detainees were\\n“enemy combatants.” Id. at 733. The Boumediene\\npetitioners, who had all appeared before CSRTs and been\\ndeemed enemy combatants, sought a writ of habeas corpus\\nunder the general habeas statute, 28 U.S.C. § 2241. Id. at\\n734. After protracted litigation, Congress passed the\\n\\f22 THURAISSIGIAM v. USDHS\\n\\nDetainee Treatment Act of 2005 (“DTA”), which amended\\n§ 2241 to bar judicial review of habeas petitions filed by\\nGuantanamo detainees and to vest review of CSRT decisions\\nexclusively in the D.C. Circuit. Id. at 735 (citing DTA\\n§ 1005(e), 119 Stat. 2742). Section 7 of the Military\\nCommissions Act of 2006 (“MCA”) made those provisions\\nretroactive. Id. at 736. See generally Hamad v. Gates,\\n732 F.3d 990, 996–99 (9th Cir. 2013) (describing\\nBoumediene’s place in the line of Guantanamo detainee\\ncases). The Court took a two-step approach to evaluating the\\ndetainees’ challenge to the MCA.\\n\\n At step one, the Court evaluated whether the Guantanamo\\ndetainees – as enemy combatants detained on foreign soil –\\ncould even invoke the Suspension Clause. See Boumediene,\\n553 U.S. at 739. In so doing, the Court affirmed that\\nalthough the writ’s protections may have expanded since the\\nConstitution’s drafting, “at the absolute minimum,” the\\nClause protects the writ as it existed in 1789. Id. at 746\\n(citing St. Cyr, 533 U.S. at 301). The Court therefore\\nexamined historical authorities to determine the scope of the\\nwrit in 1789, and whether it ran to “an enemy alien detained\\nabroad.” Id. at 752. Although noting that “at common law a\\npetitioner’s status as an alien was not a categorical bar to\\nhabeas corpus,” the Court concluded that the historical record\\ndid not provide a definitive answer. Id. at 747, 752. Instead,\\nthe Court turned to its extraterritoriality precedents and from\\nthem concluded that “questions of extraterritoriality turn on\\nobjective factors and practical concerns, not formalism.” Id.\\nat 764. Boumediene drew from Johnson v. Eisentrager,\\n339 U.S. 763 (1950), another case about the extraterritorial\\napplication of the Suspension Clause, three non-exclusive\\nfactors relevant to the Clause’s extraterritorial scope:\\n\\f THURAISSIGIAM v. USDHS 23\\n\\n (1) the citizenship and status of the detainee\\n and the adequacy of the process through\\n which that status determination was made;\\n (2) the nature of the sites where apprehension\\n and then detention took place; and (3) the\\n practical obstacles inherent in resolving the\\n prisoner’s entitlement to the writ.\\n\\n553 U.S. at 766.8 Applying those factors, the Court\\nconcluded that the detainees could invoke the Suspension\\nClause. Id. at 771.\\n\\n At step two, the Court considered whether Congress had\\nsuspended the writ without an adequate substitute. The Court\\nacknowledged that there are “few precedents addressing what\\nfeatures an adequate substitute for habeas corpus must\\ncontain.” Id. at 772. For example, the Court had previously\\nupheld provisions of the Antiterrorism and Effective Death\\nPenalty Act of 1996 (“AEDPA”) against a Suspension Clause\\nchallenge because the provisions “did not constitute a\\nsubstantial departure from common-law habeas procedures.”\\nId. at 774 (citing Felker, 518 U.S. at 664); see also Neuman,\\n110 Colum. L. Rev. at 542 (stating that “what matters is the\\nsubstance, not the form, of the Great Writ,” and that\\n“Congress can rename or reconfigure the procedure by which\\ncourts examine the lawfulness of detention,” as long as the\\nsubstitute is adequate).\\n\\n 8\\n In Rasul v. Bush, 542 U.S. 466 (2004), the Court had first discussed\\nEisentrager’s applicability to the question of who may invoke the\\nSuspension Clause. Id. at 487 (Kennedy, J., concurring) (“A faithful\\napplication of Eisentrager, then, requires an initial inquiry into the general\\ncircumstances of the detention to determine whether the Court has the\\nauthority to entertain the petition and to grant relief after considering all\\nof the facts presented.”).\\n\\f24 THURAISSIGIAM v. USDHS\\n\\n In Boumediene, the Court gleaned from its precedents two\\n“easily identified attributes of any constitutionally adequate\\nhabeas corpus proceeding.” 553 U.S. at 779. First, the\\n“privilege of habeas corpus entitles the prisoner to a\\nmeaningful opportunity to demonstrate that he is being held\\npursuant to ‘the erroneous application or interpretation’ of\\nrelevant law.” Id. (quoting St. Cyr, 533 U.S. at 302). Second,\\n“the habeas court must have the power to order the\\nconditional release of an individual unlawfully detained.” Id.\\nBeyond those minimum requirements, “depending on the\\ncircumstances, more may be required.” Id.\\n\\n The Court further emphasized that “the necessary scope\\nof habeas review in part depends upon the rigor of any earlier\\nproceedings.” Id. at 781; see also id. at 786 (noting that\\n“habeas corpus review may be more circumscribed if the\\nunderlying detention proceedings are more thorough”). For\\nthat reason, courts sitting in habeas afford deference when\\nreviewing another court’s decision, but when a petitioner is\\n“detained by executive order . . . the need for collateral\\nreview is most pressing.” Id. at 783. To be effective, the\\n“habeas court must have sufficient authority to conduct a\\nmeaningful review of both the cause for detention and the\\nExecutive’s power to detain.” Id. Applying those principles\\nto the CSRTs and D.C. Circuit review, the Court concluded\\nthat the MCA did not provide an adequate substitute because\\nthe D.C. Circuit could not “consider newly discovered\\nevidence that could not have been made part of the CSRT\\nrecord.” Id. at 790. The Court then concluded that it was not\\npossible to read into the statute provisions for the procedures\\nnecessary to satisfy the Suspension Clause, and therefore held\\nit unconstitutional. Id. at 792.\\n\\f THURAISSIGIAM v. USDHS 25\\n\\n Boumediene provides an analytical template for\\nevaluating a Suspension Clause challenge: at step one, we\\nexamine whether the Suspension Clause applies to the\\npetitioner; and, if so, at step two, we examine whether the\\nsubstitute procedure provides review that satisfies the Clause.\\nHow more specifically to apply that template is less clear,\\ngiven that the Court generated its three-factor test at step one\\nin light of the extraterritoriality question in Boumediene. See\\nid. at 764, 766. Those factors, as both parties acknowledge,\\ndo not map precisely onto this case because Thuraissigiam\\nwas apprehended and is detained on U.S. soil.9 Yet, the\\nmanner in which the Court divined those factors informs our\\napproach here. Boumediene relied on Eisentrager and related\\ncases, but also looked to 1789-era application of the writ to\\ndetermine whether petitioners similarly situated to\\nGuantanamo detainees had been able to invoke the Clause.\\nAlthough the Court emphasized that the history was not\\ndispositive, it made clear that “settled precedents or legal\\ncommentaries in 1789 . . . can be instructive.” Id. at 739.10\\n\\n\\n\\n\\n 9\\n We too have applied the three Boumediene factors more readily\\nwhen asking whether a noncitizen outside the United States – again,\\nunlike Thuraissigiam – can claim the Constitution’s protections. See\\nIbrahim v. Dep’t of Homeland Sec., 669 F.3d 983, 995 (9th Cir. 2012)\\n(evaluating extraterritoriality in context of First and Fifth Amendment\\nclaims).\\n 10\\n Indeed, the Supreme Court’s proposition that the Suspension Clause\\nat least protects the writ as it existed in 1789 “necessarily invites reference\\nto history when interpreting and applying the Suspension Clause.”\\nAmanda L. Tyler, Habeas Corpus in Wartime 9 (2017); see also Omar v.\\nMcHugh, 646 F.3d 13, 19 (D.C. Cir. 2011) (Kavanaugh, J.) (“[H]istory\\nmatters: In habeas cases, we seek guidance from history ‘addressing the\\nspecific question before us.’” (quoting Boumediene, 553 U.S. at 746)).\\n\\f26 THURAISSIGIAM v. USDHS\\n\\n At step two, Boumediene held that, at a minimum, the\\nSuspension Clause entitles a petitioner “to a meaningful\\nopportunity to demonstrate he is being held to ‘the erroneous\\napplication or interpretation’ of relevant law.” Id. at 779\\n(quoting St. Cyr, 533 U.S. at 302). In considering whether\\nthe Clause required more in the circumstances of\\nBoumediene, the Court impliedly considered the rigor and\\ncharacter of the proceedings preceding habeas review. Also\\nrelevant to Thuraissigiam’s case, the Court affirmed that the\\nSuspension Clause protects “a right of first importance,” even\\nin circumstances – such as national security, in Boumediene\\n– where the executive’s power is at its zenith. Id. at 797–98.\\n\\n B. INS v. St. Cyr\\n\\n St. Cyr, which predated Boumediene by several years,\\nsheds additional light on the Court’s approach to Suspension\\nClause questions. The petitioner, St. Cyr, was a lawful\\npermanent resident admitted to the United States in 1986.\\n533 U.S. at 293. In 1996, St. Cyr pleaded guilty to a criminal\\ncharge that made him removable, although under pre-AEDPA\\nlaw (applicable at the time of his conviction), he was eligible\\nfor a discretionary waiver from the Attorney General. Id.\\nAfter AEDPA was passed, the government began removal\\nproceedings, with the Attorney General interpreting AEDPA\\nto have removed his discretion to grant St. Cyr a waiver. Id.\\nSt. Cyr filed a habeas petition alleging that the Attorney\\nGeneral’s interpretation was erroneous because St. Cyr’s\\nconviction predated AEDPA. Id. After the district court and\\nSecond Circuit agreed with St. Cyr, the government argued to\\n\\f THURAISSIGIAM v. USDHS 27\\n\\nthe Supreme Court that the courts lacked jurisdiction to\\nreview the Attorney General’s interpretation. Id. at 297–98.\\n\\n The Court stated that the government’s position had to\\novercome several presumptions, chief among them the\\n“strong presumption in favor of judicial review of\\nadministrative action and the longstanding rule requiring a\\nclear statement of congressional intent to repeal habeas\\njurisdiction.” Id. at 298 (citing Ex parte Yerger, 75 U.S.\\n(8 Wall.) 85, 102 (1869)). To address whether the statute\\nraised serious Suspension Clause questions, the Court started\\nfrom the principle that “[b]ecause of [the] Clause, some\\n‘judicial intervention in deportation cases’ is unquestionably\\n‘required by the Constitution.’” Id. at 300 (citing Heikkila v.\\nBarber, 345 U.S. 229, 235 (1953)). Because “at the absolute\\nminimum, the Suspension Clause protects the writ ‘as it\\nexisted in 1789,’” the Court looked at the writ’s application\\nbefore and after the drafting of the Constitution. Id. at 301\\n(quoting Felker, 518 U.S. at 663–64). Legal and historical\\nauthorities indicated that in both England and the United\\nStates “the writ of habeas corpus has served as a means of\\nreviewing the legality of Executive detention, and it is in that\\ncontext that its protections have been strongest.” Id.\\nMoreover, the writ was available both to “nonenemy aliens as\\nwell as citizens” and “encompassed detentions based on\\nerrors of law, including the erroneous application or\\ninterpretation of statutes.” Id. at 301–02.\\n\\f28 THURAISSIGIAM v. USDHS\\n\\n St. Cyr also looked to the so-called “finality era,”11 during\\nwhich the statutory scheme precluded judicial intervention in\\nimmigration enforcement, except as required by the\\nConstitution. Id. at 304–06. Despite that statutory bar, the\\nCourt in the finality era “allow[ed] for review on habeas of\\nquestions of law.” Id. at 304. Accordingly, the government’s\\nreading of the statute – to prohibit any judicial review of the\\nAttorney General’s interpretation – raised “Suspension\\nClause questions that . . . are difficult and significant.” Id.\\nMore directly, “to conclude that the writ is no longer\\navailable in this context would represent a departure from\\nhistorical practice in immigration law.” Id. at 305. After\\ncanvassing that historical practice, and noting that it was\\nconsistent with the writ’s “common-law antecedents,” the\\nCourt concluded that St. Cyr could have brought his habeas\\nclaims under that regime. Id. at 308. Thus, due to the serious\\nconstitutional questions raised, and because Congress had not\\nprovided a “clear, unambiguous, and express” intent to\\npreclude habeas jurisdiction over questions of law, the Court\\nconcluded that the statutes at issue did not repeal habeas\\njurisdiction. Id. at 314.\\n\\n St. Cyr further illuminates how to approach both\\nBoumediene steps. Like Boumediene, St. Cyr looked to the\\n1789-era historical application of the writ. St. Cyr also\\nlooked to the finality era because it provides evidence of what\\n\\n 11\\n The “finality era” refers to “an approximately sixty-year period\\nwhen federal immigration law rendered final (hence, the ‘finality’ era) the\\nExecutive’s decisions to admit, exclude, or deport noncitizens. This\\nperiod began with the passage of the Immigration Act of 1891, ch. 551,\\n26 Stat. 1084, and concluded when Congress enacted the Immigration and\\nNationality Act of 1952, Pub. L. No. 82-414, 66 Stat. 163, which\\npermitted judicial review of deportation orders through declaratory\\njudgment actions in federal district courts.” Castro, 835 F.3d at 436.\\n\\f THURAISSIGIAM v. USDHS 29\\n\\ndegree of habeas review is required under the Suspension\\nClause and to whom such review is guaranteed in the\\nimmigration enforcement context. St. Cyr’s resort to prior\\nhabeas cases aligns with Boumediene’s similar reliance on\\nEisentrager to resolve ambiguities in the 1789-era application\\nof the writ. That St. Cyr ultimately avoided the Suspension\\nClause question does not diminish its wisdom or relevance as\\nan example of the Court’s analytical approach to Suspension\\nClause questions. Consistent with Boumediene and St. Cyr,\\nwe conclude that both the common-law history of the writ\\nand the Court’s finality era cases are relevant to what and\\nwhom the Suspension Clause protects. See also Flores-\\nMiramontes v. INS, 212 F.3d 1133, 1141–43 (9th Cir. 2000)\\n(relying on common-law history and finality era cases in\\naddressing Suspension Clause challenge); see also Trinidad\\ny Garcia v. Thomas, 683 F.3d 952, 960 (9th Cir. 2012) (en\\nbanc) (Thomas, J., concurring) (discussing finality era cases\\nas evidence of rights protected by the Suspension Clause).\\n\\n C. The Third Circuit’s Decision in Castro\\n\\n Before addressing Thuraissigiam’s Suspension Clause\\nchallenge, we discuss the Third Circuit’s decision in Castro,\\nwhich involved an analogous challenge to § 1252(e).12 The\\n\\n\\n 12\\n The government’s contention that the Second and Seventh Circuits\\nhave addressed the question before us is incorrect. Neither case addresses\\nthe Suspension Clause. See Shunaula v. Holder, 732 F.3d 143, 147 (2d\\nCir. 2013) (addressing due process challenge to § 1252(e)(2) and\\n§ 1252(a)(2)(A)); Khan v. Holder, 608 F.3d 325, 328 (7th Cir. 2010)\\n(addressing § 1252(e)(2) in light of that circuit’s “safety valve” doctrine\\nfor “judicial correction of bizarre miscarriages of justice”). Likewise, the\\ncase cited in the government’s Rule 28(j) letter, Hamama v. Adducci,\\n912 F.3d 869 (6th Cir. 2018), does not address the Suspension Clause in\\nthe context of the procedures leading up to an expedited removal order.\\n\\f30 THURAISSIGIAM v. USDHS\\n\\nThird Circuit concluded that § 1252(e) does not violate the\\nSuspension Clause as applied to 28 asylum-seeking families\\nwho, like Thuraissigiam, raised constitutional, statutory, and\\nregulatory claims relating to their negative credible fear\\ndeterminations. 835 F.3d at 425, 428. The families were all\\napprehended shortly after entering the country, placed in\\nexpedited removal, and found not to have credible fear. Id.\\nat 427–28. As we do, the Third Circuit rejected the argument\\nthat § 1252(e)(2) provides jurisdiction over claims of legal\\nerror. Id. at 434.\\n\\n Turning to the petitioners’ Suspension Clause challenge,\\nthe court opined that the Supreme Court’s habeas cases are\\n“perhaps even competing” with the plenary power doctrine.\\nId. After reviewing Boumediene and St. Cyr, Castro\\ndiscussed the Court’s “commitment to the full breadth of\\n[that] doctrine, at least as to aliens at the border seeking\\ninitial admission to the country.” Id. at 443. Castro\\napproached step one of Boumediene by reference to the\\npetitioners’ status in light of Landon v. Plascencia, 459 U.S.\\n21 (1982), a case addressing due process, not habeas, rights.\\nCastro concluded that petitioners, as “recent surreptitious\\nentrants,” should be treated for constitutional purposes as\\n“alien[s] seeking initial admission to the United States.”\\n835 F.3d at 448. In Landon, the Court stated that such a\\nnoncitizen “has no constitutional rights regarding his\\napplication” for entry into the country. 459 U.S. at 32.\\nAccordingly, the Third Circuit concluded that the petitioners’\\nchallenge failed at step one, and did not address whether\\n§ 1252(e) was an adequate habeas substitute. 835 F.3d at\\n446. The court acknowledged that its discussion of the\\npetitioners’ status “appear[ed] to ignore” Supreme Court\\nprecedent relating to the due process rights of noncitizens\\nphysically present in the country, but concluded that no case\\n\\f THURAISSIGIAM v. USDHS 31\\n\\nhad clearly held that “arriving aliens” were entitled to due\\nprocess protections. Id. at 447–48.13\\n\\n We disagree with Castro’s resolution of how Boumediene\\nand St. Cyr require us to approach a Suspension Clause\\nchallenge. As explained at length above, the Court’s mode of\\nanalysis in both of those cases addressed the scope of the\\nSuspension Clause by reference to the writ as it stood in 1789\\nand relevant habeas corpus precedents. Castro explained that\\nit did not rely on St. Cyr’s description of the Court’s habeas\\napproach in immigration cases in the finality era by\\nemphasizing that, unlike the Castro petitioners, St. Cyr was\\na lawful permanent resident, and that St. Cyr discussed only\\nwhat the Suspension Clause might protect. Id. at 446.\\n\\n That St. Cyr did not affirmatively hold that the\\nSuspension Clause was violated does not render its\\ndescription of the finality era cases incorrect or its approach\\nirrelevant. Moreover, Castro’s decision to rely instead on\\nLandon is misplaced. Landon held that a permanent resident\\nwho traveled abroad and was detained when attempting to\\nreenter the United States should be placed in exclusion\\n\\n\\n\\n\\n 13\\n After argument, the Third Circuit decided Osorio-Martinez v.\\nAttorney General, 893 F.3d 153 (3d Cir. 2018), involving four juvenile\\npetitioners from Castro. After their original habeas petitions were\\ndismissed, the juveniles had been granted Special Immigrant Juvenile\\n(“SIJ”) status under 8 U.S.C. § 1108(a)(27)(J). Id. at 160. Applying\\nCastro, the Third Circuit held that § 1252(e) was an unconstitutional\\nsuspension of the writ as applied to the petitioners, by virtue of their\\n“significant ties to this country” and the constitutional and statutory rights\\nflowing to SIJ designees under 8 U.S.C. § 1255(a) & (h)(1). Id. at 167.\\n\\f32 THURAISSIGIAM v. USDHS\\n\\nproceedings rather than deportation. 459 U.S. at 22.14\\nAddressing the petitioner’s due process challenge to her\\nexclusion proceedings, the Court noted it had “long held that\\nan alien seeking initial admission to the United States\\nrequests a privilege and has no constitutional rights regarding\\nhis application.” Id. As explained by Judge Hardiman, the\\nCourt in Landon did not “purport to resolve a jurisdictional\\nquestion raising the possibility of an unconstitutional\\nsuspension of the writ of habeas corpus”; rather it addressed\\nonly the due process rights of a permanent resident. Castro,\\n835 F.3d at 450 (Hardiman, J., concurring dubitante); see\\nLandon, 459 U.S. at 32–35. Landon could not and did not\\naddress the much different question of whether a petitioner\\nlike Thuraissigiam may invoke the Suspension Clause.15\\n\\n Although often conflated, the rights protected by the\\nSuspension Clause are not identical to those under the Fifth\\nAmendment’s guarantee of due process. See Lee Kovarsky,\\n\\n\\n 14\\n At the time, Congress provided that removable noncitizens in the\\nUnited States were subject to deportation and those seeking initial entry\\nwere subject to exclusion. Id. at 25. Now all noncitizens are subject to\\nremoval, whether via 8 U.S.C. § 1225(b)(1) expedited removal or the\\nremoval procedures under 8 U.S.C. § 1229a.\\n 15\\n Regardless, we disagree with the government’s contention and\\nCastro’s conclusion that a person like Thuraissigiam lacks all procedural\\ndue process rights. See 835 F.3d at 447–48. The Supreme Court has been\\nclear that presence matters to due process. See, e.g., Mathews v. Diaz,\\n426 U.S. 67, 77 (1976); Zadvydas v. Davis, 533 U.S. 678, 693 (2001).\\nAnd we have held that a noncitizen situated almost exactly like\\nThuraissigiam had a constitutional right “to expedited removal\\nproceedings that conformed to the dictates of due process.” United States\\nv. Raya-Vaca, 771 F.3d 1195, 1203 (9th Cir. 2014); see also Immigration\\nScholars Amicus Br. (explaining why Thuraissigiam has procedural due\\nprocess rights).\\n\\f THURAISSIGIAM v. USDHS 33\\n\\nCustodial and Collateral Process: A Response to Professor\\nGarrett, 98 Cornell L. Rev. Online 1, 1 (2013) (“Due process\\nand the habeas privilege are distinct constitutional\\nphenomena, [but] federal courts almost pathologically\\nconfuse them.”). It is true that, historically, the Fifth\\nAmendment’s due process guarantee and the Suspension\\nClause have been applied in tandem, as their applicability was\\nrarely disputed. See Mary Van Houten, The Post-\\nBoumediene Paradox: Habeas Corpus or Due Process?,\\n67 Stan. L. Rev. Online 9, 10 (2014) (observing that these\\nprovisions “were almost always jointly applied before\\nBoumediene”). But this fact does not mean these rights\\nshould be elided, as made clear by the fact that the\\nConstitution, ratified two-and-a-half years before the Fifth\\nAmendment, see Bute v. People of State of Ill., 333 U.S. 640,\\n650 (1948), presupposed the existence of the writ of habeas\\ncorpus, see Boumediene, 553 U.S. at 739 (“Protection for the\\nprivilege of habeas corpus was one of the few safeguards of\\nliberty specified in a Constitution that, at the outset, had no\\nBill of Rights.”). Indeed, the writ “is almost the only remedy\\nmentioned in the Constitution” as originally ratified. Fallon\\n& Meltzer, 120 Harv. L. Rev. at 2037.\\n\\n Boumediene itself clearly recognized the distinction\\nbetween the Fifth Amendment’s due process rights and the\\nSuspension Clause – providing further reason not to treat\\nLandon’s discussion of due process rights as having any\\nbearing on the application of the Suspension Clause. In\\nBoumediene, the Court decided that the Guantanamo\\ndetainees could invoke the Suspension Clause without\\naddressing whether they had due process rights or whether\\nthe CSRTs satisfied due process. 553 U.S. at 785; see also id.\\nat 739 (starting from the proposition that “protection for the\\nprivilege of habeas corpus was one of the few safeguards of\\n\\f34 THURAISSIGIAM v. USDHS\\n\\nliberty specified in a Constitution that, at the outset, had no\\nBill of Rights[]”); Flores-Miramontes, 212 F.3d at 1142\\n(noting that habeas was available at common law prior to the\\ndrafting of the Constitution). The Court in Boumediene\\ntherefore explicitly declined to link due process rights and\\nSuspension Clause rights. See Hamad, 732 F.3d at 999\\n(noting that Boumediene did not address whether the due\\nprocess clause applied to the Guantanamo detainees); see also\\nKiyemba v. Obama, 555 F.3d 1022, 1027 (D.C. Cir. 2009)\\n(concluding on habeas review that Guantanamo detainees\\nlacked due process rights), vacated by 559 U.S. 131 (2010),\\nreinstated by 605 F.3d 1046, 1047 (D.C. Cir. 2010). Landon,\\na due process case, is not relevant to whether Thuraissigiam\\ncan invoke the Suspension Clause. For that reason, we\\ndecline to follow Castro’s approach and reject the\\ngovernment’s argument that Thuraissigiam’s purported lack\\nof due process rights is determinative of whether he can\\ninvoke the Suspension Clause.\\n\\n Instead, in accordance with Boumediene, we evaluate\\nThuraissigiam’s Suspension Clause challenge in two steps:\\nFirst, to determine whether Thuraissigiam may invoke the\\nSuspension Clause, we examine 1789-era practice, the\\nfinality era cases, and other relevant precedents. Second, we\\nask whether § 1252(e)(2) provides Thuraissigiam a\\n“meaningful opportunity to demonstrate that he is being held\\npursuant to ‘the erroneous application or interpretation’ of\\nrelevant law.” Boumediene, 553 U.S. at 779. At step two, we\\nkeep in mind that the character of the earlier proceedings\\nbears on the level of habeas review required. Id. at 781.\\n\\f THURAISSIGIAM v. USDHS 35\\n\\nIII. Application\\n\\n A. Garcia de Rincon and Pena\\n\\n At the outset, the government contends that our decisions\\nin Garcia de Rincon, 539 F.3d 1133, and Pena v. Lynch,\\n815 F.3d 452 (9th Cir. 2016), require us to affirm. Although\\nin both cases we rejected arguments that § 1252(e)(2)\\nauthorized jurisdiction, neither case answered the\\nconstitutional question before us today.\\n\\n In Garcia de Rincon, the petitioner was a noncitizen\\nliving in the United States who was stopped at the border\\nattempting to reenter after a visit to Mexico, and placed in\\nexpedited removal. 539 F.3d at 1135. After rejecting the\\npetitioner’s statutory challenge, we dismissed her argument\\n– “although . . . not articulated” as such – that the Suspension\\nClause required review of her petition. Id. at 1141. The\\nprecise question considered was whether “the INA provides\\nno adequate substitute for habeas review and therefore\\nsuspends the writ” – a Boumediene step two question,\\nalthough Garcia de Rincon never addressed Boumediene,\\nwhich had been decided months earlier. Id. We concluded\\nthat Li v. Eddy, 259 F.3d 1132 (9th Cir. 2001), vacated on\\nreh’g as moot, 324 F.3d 1109 (9th Cir. 2003), discredited the\\npetitioner’s “generalized due process argument,” the only\\nright she sought to vindicate via her petition. Id. Garcia de\\nRincon says nothing about whether Thuraissigiam can invoke\\nthe Suspension Clause, whether the Clause requires habeas\\nreview of statutory or legal claims, or what the Clause\\nrequires for a petitioner like Thuraissigiam who is within the\\nUnited States. Instead, the case addressed only whether\\n§ 1252(e)(2) suspends the writ when a petitioner lacks due\\nprocess rights. Put in Boumediene step-two terms, the due\\n\\f36 THURAISSIGIAM v. USDHS\\n\\nprocess clause was not “relevant law” for the Garcia de\\nRincon petitioner.16\\n\\n Pena also did not settle the question before us. In Pena,\\na noncitizen placed in expedited removal filed a petition for\\nreview of the Board of Immigration Appeals’ dismissal of his\\nappeal from an IJ’s decision affirming a negative credible\\nfear determination. 815 F.3d at 454. Because the petition\\nwas not brought under § 1252(e)(2), we concluded that we\\nlacked jurisdiction. Id. at 457. We went on to note that in\\nWebster v. Doe, 486 U.S. 592, 603 (1988), the Court had\\n“suggested that a litigant may be unconstitutionally denied a\\nforum when there is absolutely no avenue for judicial review\\nof a colorable claim of constitutional deprivation.” 815 F.3d\\nat 456 (emphasis in original). We concluded that Pena’s\\npetition did not raise Webster concerns because he lacked a\\ncolorable constitutional claim,17 and further noted that\\n§ 1252(e)(2) provides “some avenues of judicial review.” Id.\\nat 456–57. All that Pena says, therefore, is that § 1252(e)\\ndoes not implicate the Webster doctrine when a petitioner\\nfails to raise colorable constitutional claims. Pena never\\naddressed the Suspension Clause.\\n\\n Because neither Garcia de Rincon nor Pena addressed\\nwhether § 1252(e)(2) unlawfully suspends the writ as applied\\n\\n\\n 16\\n In case there were any doubt, Smith subsequently reserved the\\nquestion of whether, as applied to a noncitizen in expedited removal, the\\nSuspension Clause requires review beyond that provided for in\\n§ 1252(e)(2). See 741 F.3d at 1022 n.6. That reservation necessarily\\ndetermined that Garcia de Rincon had not settled the question.\\n 17\\n Pena claimed that the IJ violated due process by failing to elicit a\\nvoluntary waiver of his right to counsel, but we noted that that claim was\\ncontradicted by the record. Id. at 455–56.\\n\\f THURAISSIGIAM v. USDHS 37\\n\\nto a petitioner like Thuraissigiam, we reject the government’s\\nargument that those cases alone require us to affirm.\\n\\n B. Reach of the Suspension Clause\\n\\n At Boumediene step one, we must consider the reach of\\nthe Suspension Clause, or, in other words, whether\\nThuraissigiam is “barred from seeking the writ or invoking\\nthe protections of the Suspension Clause . . . because of [his]\\nstatus . . . .” Boumediene, 553 U.S. at 739. In Boumediene,\\nthe Court answered this question by reference to its\\nprecedents and the common law history of the writ. We\\ntherefore do the same.18\\n\\n\\n 18\\n As described above, the Court in Boumediene generated a three-\\nfactor test at step one in light of the extraterritoriality question presented.\\nThis test does not clearly fit in the present case, given that Thuraissigiam\\nwas apprehended and detained in the United States. See 553 U.S. at 764,\\n766. However, even were we to apply Boumediene’s three-factor test\\nhere, it would, as in Boumediene, support application of the Suspension\\nClause.\\n\\n The first factor, Thuraissigiam’s “citizenship and status” and “the\\nadequacy of the process through which that status determination was\\nmade,” Id. at 766, weighs in favor of applying the Suspension Clause.\\nThuraissigiam is a foreign national who contests his status – he contends\\nthat he has a credible fear of persecution and therefore qualifies as a\\nrefugee entitled to asylum. Like the CRST process at issue in\\nBoumediene, the determination as to whether a noncitizen has a credible\\nfear is not made via a “rigorous adversarial process to test the legality of\\n[his] detention.” Id. at 767. The determination is made by an asylum\\nofficer, 8 C.F.R. § 208.30(d), and although the noncitizen may consult\\nothers and even have them present a statement at the end of the interview,\\n8 C.F.R. § 208.30(d)(4), other hallmarks of the adversarial process are\\nlacking. If the noncitizen then chooses to contest an asylum officer’s\\nnegative credible fear determination, the noncitizen is entitled only to\\ncursory review by an IJ. 8 C.F.R. § 208.30(g); 8 C.F.R. § 1208.39(g)(2).\\n\\f38 THURAISSIGIAM v. USDHS\\n\\n As explained, in St. Cyr, the Court canvassed cases from\\nEngland and historical accounts to conclude that the writ was\\navailable before 1789 to “nonenemy aliens as well as to\\ncitizens.” 533 U.S. at 301; accord Boumediene, 553 U.S. at\\n747 (“We know that at common law a petitioner’s status as an\\n\\n\\nCritically, unlike in Boumediene, a noncitizen cannot seek review of the\\ncredible fear determination in an Article III court. See 8 U.S.C.\\n§ 1252(e)(2); cf. Boumediene, 553 U.S. at 767 (“And although the\\ndetainee can seek review of his status determination in the Court of\\nAppeals, that review process cannot cure all defects in the earlier\\nproceedings.”). Accordingly, the procedural protections available to\\nThuraissigiam and other noncitizens in expedited removal “fall well short\\nof the procedures and adversarial mechanisms that would eliminate the\\nneed for habeas corpus review.” Boumediene, 553 U.S. at 767 (internal\\nquotation marks omitted).\\n\\n As to the second factor, there is no question that Thuraissigiam was\\napprehended and detained within the sovereign territory of the United\\nStates. This factor weighs strongly in favor of finding Thuraissigiam has\\nrights under the Suspension Clause. See id. at 768–69. The government\\ninsists that the nature and length of a noncitizen’s detention is relevant to\\nthis factor. Not so. Boumediene only invokes these considerations under\\nstep two. The second factor (under step one) is wholly focused on the\\nlevel and duration of control exerted by the United States over the territory\\n– which is not at issue here, where the territory is the United States.\\nBoumediene, 553 U.S. at 768–69.\\n\\n As in Boumediene, the third factor is somewhat equivocal: “there are\\ncosts to holding the Suspension Clause applicable in a case of [asylum-\\nseekers in expedited removal proceedings.]” Id. at 769. But\\n“[c]ompliance with any judicial process requires some incremental\\nexpenditure of resources,” and direct review by the courts already exists\\nand functions in non-expedited removal proceedings. Id. Thus, here, as\\nin Boumediene, “[w]hile we are sensitive to [the government’s] concerns,\\nwe do not find them dispositive.” Id.\\n\\n Consequently, Boumediene’s extraterritorial step one factors, if they\\nwere relevant here, would support application of the Suspension Clause.\\n\\f THURAISSIGIAM v. USDHS 39\\n\\nalien was not a categorical bar to habeas corpus relief.”);\\nRasul, 542 U.S. at 481 (“At common law, courts exercised\\nhabeas jurisdiction over the claims of aliens detained within\\nsovereign territory of the realm. . . .”); see also Gerald L.\\nNeuman, Habeas Corpus, Executive Detention, and the\\nRemoval of Aliens, 98 Colum. L. Rev. 961, 989–90 (1998)\\n(collecting cases).\\n\\n After the adoption of the Constitution and its Suspension\\nClause, courts in the United States applied the same\\napproach. For example, in Ex parte D’Olivera, 7 F. Cas. 853\\n(C.C.D. Mass. 1813) (No. 3,967), a federal court in\\nMassachusetts permitted an arrested noncitizen seaman to\\ninvoke habeas. In later years, the Supreme Court continued\\nto hold that habeas was available to noncitizens – even\\nexcluded noncitizens stopped at the border. United States v.\\nJung Ah Lung, 124 U.S. 621, 628–32 (1888); see also\\nNeuman, 98 Colum. L. Rev. at 1006. Cases throughout the\\nfinality era, from the 1890s to the 1950s, which carry\\nsignificant weight here, held firm to this constitutional\\npremise. In Nishimura Ekiu v. United States, 142 U.S. 651\\n(1892), the Court affirmed that despite the finality law, “[a]n\\nalien immigrant, prevented from landing by any such officer\\nclaiming authority to do so under an act of congress, and\\nthereby restrained of his liberty, is doubtless entitled to a writ\\nof habeas corpus to ascertain whether the restraint is lawful.”\\nId. at 660.\\n\\n The Court continued that approach later in the finality era.\\nGegiow v. Uhl, 239 U.S. 3, 9 (1915) (“The courts are not\\nforbidden by the [finality] statute to consider whether the\\nreasons, when they are given, agree with the requirements of\\nthe act.”). In Gegiow, for example, the Court reversed the\\ngovernment’s legal conclusion that the petitioner was subject\\n\\f40 THURAISSIGIAM v. USDHS\\n\\nto exclusion as a public charge based on a lack of labor\\nopportunities in his immediate destination. Id. at 9–10; see\\nalso Shaughnessy v. United States ex rel. Mezei, 345 U.S.\\n206, 212–13 (1953) (stating that even though a noncitizen\\nwho had not entered the country lacks due process, he “may\\nby habeas corpus test the validity of his exclusion”); United\\nStates ex rel. Knauff v. Shaughnessy, 338 U.S. 537, 544\\n(1950) (addressing, but rejecting, noncitizen’s “contention\\nthat the regulations were not ‘reasonable’ as they were\\nrequired to be [under a federal statute]”). In United States ex\\nrel. Accardi v. Shaughnessy, the Court granted habeas on\\nlegal grounds to a noncitizen who entered from Canada\\n“without immigration inspection and without an immigration\\nvisa.” 347 U.S. 260, 262, 268 (1954). In Heikkila, the Court\\nexplained that the Constitution was the source for habeas\\nreview during the finality era, 345 U.S. at 234–35, and in St.\\nCyr, the Court clarified that the Suspension Clause was the\\nspecific source of such review. See 533 U.S. at 304. Indeed,\\nthe government points to no alternative reading.\\n\\n More broadly, the government offers no convincing\\nreason to discount the finality era, nor does it offer a\\ncompeting account of the common-law scope of the writ or of\\nthe finality era. The government, citing Mezei, 345 U.S. at\\n214, answers Boumediene step one by contending that\\nThuraissigiam, “as an alien apprehended immediately after\\ncrossing the border illegally, is no different from other aliens\\nat the border, and is therefore ‘assimilated to [that] status’ for\\nconstitutional purposes.” However, Mezei spoke only of such\\nassimilation for the purposes of due process, and it otherwise\\naffirmed the principle that habeas is available even when a\\npetitioner lacks due process rights. Id. at 213. And, crucially,\\nBoumediene never linked Suspension Clause rights to due\\nprocess rights. The government provides no authority from\\n\\f THURAISSIGIAM v. USDHS 41\\n\\nSuspension Clause cases to support its contention that\\nThuraissigiam lacks Suspension Clause rights. Because in\\nthe finality era the Court permitted even arriving noncitizens\\nto invoke habeas review, we conclude that Thuraissigiam,\\nwho was arrested within the United States, may invoke the\\nSuspension Clause.19\\n\\n C. Compliance with the Suspension Clause\\n\\n Having concluded that Thuraissigiam may invoke the\\nSuspension Clause, we must consider at Boumediene step two\\nwhether habeas review under § 1252(e) is so limited so as\\neffectively to suspend the writ as applied to Thuraissigiam.\\nAt a minimum, the Suspension Clause “entitles the\\n[petitioner] to a meaningful opportunity to demonstrate that\\nhe is being held pursuant to ‘the erroneous application or\\ninterpretation of relevant law.’” Boumediene, 553 U.S. at 779\\n(quoting St. Cyr, 553 U.S. at 302).\\n\\n Congress may modify the scope of habeas review so long\\nas the review “is neither inadequate nor ineffective to test the\\nlegality of a person’s detention.” Swain v. Pressley, 430 U.S.\\n372, 381 (1977); see Crater v. Galarza, 491 F.3d 1119,\\n1124–25 (9th Cir. 2007) (noting, in the context of a challenge\\nto AEDPA, that not all restrictions on habeas review\\neffectively suspend the writ). We bear in mind that “[a]t its\\nhistorical core, the writ of habeas corpus has served as a\\n\\n\\n 19\\n In so doing, we reject any argument that only noncitizens who have\\n“been lawfully admitted” may invoke the Suspension Clause. Because the\\nwrit is an indispensable separation of powers mechanism, “[t]he test for\\ndetermining the scope of this provision must not be subject to\\nmanipulation by those whose power it is designed to restrain.”\\nBoumediene, 553 U.S. at 765–66.\\n\\f42 THURAISSIGIAM v. USDHS\\n\\nmeans of reviewing the legality of Executive detention, and\\nit is in that context that its protections have been strongest.”\\nSt. Cyr, 533 U.S. at 300–01; see also Boumediene, 553 U.S.\\nat 783 (noting that in cases of executive detention, “the need\\nfor habeas corpus is more urgent”). Therefore, when\\nevaluating whether a substitute is adequate, we consider “the\\nrigor of any earlier proceedings” and “the intended duration\\nof the detention and the reasons for it.” Id. at 781, 783.\\n\\n The government urges a different approach to step two.\\nThe government contends that Thuraissigiam’s status matters\\nto the extent of review the Suspension Clause requires. The\\ngovernment even suggests we should apply the Boumediene\\nstep one extraterritorial factors to determine whether\\n§ 1252(e)(2) provides sufficient review. However, those\\nfactors have no bearing on step two; only step one considers\\nthe petitioner’s status. See Boumediene, 533 U.S. at 773–93\\n(considering, without any reference to Guantanamo\\ndetainees’ status, whether the DTA was an adequate habeas\\nstatute by assessing “the sum total of procedural protections\\nafforded to the detainee at all stages, direct and collateral”).\\nBoth logically and as applied in Boumediene, the\\ncircumstances relevant to step two – the extent of review the\\nSuspension Clause requires – are those relating to the\\ndetainer, not the detainee. We also reject the government’s\\ncontention that because, in its view, Thuraissigiam lacks due\\nprocess rights, there are no rights for the Suspension Clause\\nto protect. Boumediene foreclosed that argument by holding\\nthat, whether or not due process was satisfied, the Suspension\\nClause might require more. 553 U.S. at 785.\\n\\n As a reminder, Thuraissigiam’s petition contends that the\\ngovernment denied him a “fair procedure,” “appl[ied] an\\nincorrect legal standard” to his credible fear contentions, and\\n\\f THURAISSIGIAM v. USDHS 43\\n\\n“fail[ed] to comply with the applicable statutory and\\nregulatory requirements.” The core of his claim is that the\\ngovernment failed to follow the required procedures and\\napply the correct legal standards when evaluating his credible\\nfear claim. As Thuraissigiam’s brief states: “Petitioner’s\\nclaims merely assert his right to the meaningful credible fear\\nprocedure to which he is entitled under the immigration\\nstatute, regulations, and Constitution.” We therefore consider\\nwhether the Suspension Clause requires review of those\\nclaims.20 We conclude that the Clause requires such review\\nin Thuraissigiam’s case and that because § 1252(e)(2) fails to\\nprovide a meaningful opportunity for such review, it raises\\nserious Suspension Clause questions.\\n\\n At step two, the finality era again informs our analysis of\\nwhat the Suspension Clause requires when a removal order is\\nchallenged. Finality era precedent establishes that the Court\\nregularly reviewed on habeas “claims for statutory as well as\\nconstitutional error in deportation proceedings” and “claims\\n\\n\\n 20\\n Thuraissigiam’s petition indicates that he might be asking a federal\\ncourt to review the agency’s credible fear determination, as he contends\\nthat he “can show a significant possibility of prevailing on his claims for\\nasylum and other forms of relief.” The government accordingly contends\\nthat Thuraissigiam’s petition instead requests “ultimate application of a\\nlegal standard to factual determinations and weighing of evidence\\nunderlying the Executive’s negative credible-fear findings.” However, we\\nread Thuraissigiam’s petition to be explaining why, in his view, DHS’\\nprocedural errors matter, particularly given his express assertion that he\\nonly wants review of the procedural errors. We therefore do not consider\\nhere whether the Suspension Clause requires judicial review of DHS’\\ncredible fear determination on the merits. Cf. Ortiz-Alfaro v. Holder,\\n694 F.3d 955, 958 (9th Cir. 2012) (noting, based on St. Cyr, that\\n“depriving [petitioner] the opportunity for judicial review of a\\ndetermination that he lacks a reasonable fear of persecution could raise\\nserious constitutional concerns”).\\n\\f44 THURAISSIGIAM v. USDHS\\n\\nthat deportation hearings were conducted unfairly.” Flores-\\nMiramontes, 212 F.3d at 1143 (citing, inter alia, Fong Haw\\nTan v. Phelan, 333 U.S. 6, 8–10 (1948) (interpreting statute\\non habeas)); Kessler v. Strecker, 307 U.S. 22, 33–34 (1939)\\n(same)). In Gegiow, the Court also reviewed the executive’s\\napplication of a legal standard to undisputed facts, concluding\\nthat the government had incorrectly determined that the\\npetitioner was likely to become a public charge. 239 U.S. at\\n9–10. Similarly, we have interpreted the nature of habeas\\nreview, encompassing inquiry into “the erroneous application\\nor interpretation of statutes,” St. Cyr, 533 U.S. at 302, to\\nrequire that “mixed questions of fact and law – those\\ninvolving an application of law to undisputed fact . . . be\\nprovided meaningful judicial review.” Ramadan v. Gonzales,\\n479 F.3d 646, 652 (9th Cir. 2007) (per curiam).\\n\\n Thuraissigiam and amici point us to other examples of the\\nCourt reviewing not just pure legal questions like the one at\\nissue in St. Cyr, but also the application of a legal standard to\\nundisputed facts. See, e.g., Hansen v. Haff, 291 U.S. 559,\\n562–63 (1934) (rejecting government determination that\\n“petitioner’s entry was for the purpose” of immoral\\nrelations); Mahler v. Eby, 264 U.S. 32, 44 (1924) (holding\\nthat the government failed to comply “with all the statutory\\nrequirements”). Those cases suggest that the Suspension\\nClause requires review of legal and mixed questions of law\\nand fact related to removal orders, including expedited\\nremoval orders.\\n\\n As in Boumediene, the decision to place a noncitizen in\\nexpedited removal and the finding of whether that noncitizen\\nhas a credible fear are both executive determinations,\\nmeaning that the requirements of habeas are “more urgent.”\\nCf. Boumediene, 553 U.S. at 783. While the duration of\\n\\f THURAISSIGIAM v. USDHS 45\\n\\nThuraissigiam’s detention may seem to cut against review,\\nthe Court has recognized that an excluded person’s\\n“movements are restrained by authority of the United States.”\\nMezei, 345 U.S. at 213. Moreover, “it would be difficult to\\nsay that [Thuraissigiam] was not imprisoned, theoretically as\\nwell as practically, when to turn him back meant that he must\\nget into a vessel against his wish and be carried to [Sri\\nLanka].” Chin Yow v. United States, 208 U.S. 8, 12 (1908).\\nThe finality era cases also demonstrate that habeas is a viable\\nmeans of reviewing exclusion and removal orders.\\n\\n Most important, habeas review provides important\\noversight of whether DHS complied with the required\\ncredible fear procedures.21 Under the existing administrative\\nscheme, there are no rigorous adversarial proceedings prior\\nto a negative credible fear determination. First, the credible\\nfear interview is initiated only after the CBP officer identifies\\na noncitizen who fears persecution and refers that individual\\nto a USCIS officer. See 8 C.F.R. § 235.3(b)(4); see also\\nRefugee and Human Rights Amicus Br. 11–12. A noncitizen\\ncan consult with someone at his own expense before his\\nasylum officer interview, but only as long as such\\nconsultation does not “unreasonably delay the process and is\\nat no expense to the government.” 8 C.F.R. § 208.30(d)(4).\\nBefore the IJ hearing, a noncitizen in expedited removal may\\n\\n\\n 21\\n Section 1252(e)(2) also restricts judicial oversight of whether the\\nagency properly placed a person in expedited removal in the first place:\\n“The troubling reality of the expedited removal procedure is that a CBP\\nofficer can create the § 1182(a)(7) charge by deciding to convert the\\nperson’s status from a non-immigrant with valid papers to an intending\\nimmigrant without the proper papers, and then that same officer, free from\\nthe risk of judicial oversight, can confirm his or her suspicions of the\\nperson’s intentions and find the person guilty of that charge.” Khan,\\n608 F.3d at 329.\\n\\f46 THURAISSIGIAM v. USDHS\\n\\nagain consult with someone at his own expense, but the\\nperiod to obtain such assistance is extremely abbreviated: an\\nIJ “shall conclude the review to the maximum extent\\npracticable within 24 hours” of the supervisory officer’s\\napproval of the asylum officer’s determination. 8 C.F.R.\\n§ 1003.42(c), (e). Such review may take place “in person or\\nvia telephonic or video connection.” Jaya Ramji, Legislating\\nAway International Law: The Refugee Provisions of the\\nIllegal Immigration Reform and Immigrant Responsibility\\nAct, 37 Stan. J. Int’l L. 117, 134 –41 (2001). There is also no\\nrequirement that the IJ provide reasons for her decision.\\nIndeed, in this case, the IJ simply checked a box on a form\\nstating that the immigration officer’s decision was\\n“Affirmed.”\\n\\n These meager procedural protections are compounded by\\nthe fact that § 1252(e)(2) prevents any judicial review of\\nwhether DHS complied with the procedures in an individual\\ncase, or applied the correct legal standards.22 We think it\\nobvious that the constitutional minimum – whether\\nThuraissigiam was detained pursuant to the “erroneous\\ninterpretation or application of relevant law” – is not satisfied\\nby such a scheme.23 Our conclusion is bolstered by the Third\\n\\n\\n 22\\n One amicus brief describes reports that the agency does not always\\nfollow the required procedures. See Refugee and Human Rights Amicus\\nBr. 16–27; see also Michele R. Pistone & John J. Hoeffner, Rules Are\\nMade to Be Broken: How the Process of Expedited Removal Fails Asylum\\nSeekers, 20 Geo. Immigr. L.J. 167, 175–93 (2006) (describing procedural\\nerrors commonly committed during the expedited removal process). If\\ntrue, those reports only underscore the need for judicial review.\\n 23\\n A petitioner’s perceived ultimate desire – as Judge Hardiman put\\nit in Castro, to “alter their status in the United States in the hope of\\navoiding release to their homelands,” 835 F.3d at 450–51 (Hardiman, J.,\\n\\f THURAISSIGIAM v. USDHS 47\\n\\nCircuit’s recent decision in Osorio-Martinez. As Osorio-\\nMartinez put it, § 1252(e)(2) fails to provide “even [that]\\n‘uncontroversial’ baseline of review” required by\\nBoumediene. 893 F.3d at 177. Because the statute prevented\\nthe district court from considering whether the agency\\nlawfully applied the expedited removal statute, it a fortiori\\nprecluded review of “the erroneous application or\\ninterpretation of relevant law.” Id. (citing Boumediene,\\n553 U.S. at 779). So too here, because § 1252(e)(2) prevents\\na court from reviewing claims of procedural error relating to\\na negative credible fear determination, it precludes review of\\nthe agency’s application of relevant law and thus raises\\nserious Suspension Clause questions.24 Plenary power\\nconcerns cannot in all circumstances overwhelm the\\n“fundamental procedural protections of habeas corpus . . . , a\\nright of first importance.” Boumediene, 553 U.S. at 798.\\n\\nIV. The Canon of Constitutional Avoidance\\n\\n We further decline to interpret § 1252(e)(2) to avoid the\\nserious Suspension Clause problems engendered by the\\nstatute. The constitutional avoidance canon applies “if an\\notherwise acceptable construction of a statute would raise\\n\\n\\nconcurring dubitante) – is not relevant where a petitioner challenges the\\nfairness of specific procedures leading to an expedited removal order.\\n 24\\n Because Thuraissigiam’s petition does not present the question, we\\ndo not consider one amicus’ argument that “there is a compelling case for\\nallowing habeas courts to review factual challenges to an expedited\\nremoval order.” Scholars of Habeas Corpus Law Amicus Br. 18.\\n“Generally . . . the court will not consider arguments raised only in amicus\\nbriefs.” United States v. Wahchumwah, 710 F.3d 862, 868 (9th Cir. 2013)\\n(citation omitted); see also Russian River Watershed Prot. Comm. v. City\\nof Santa Rosa, 142 F.3d 1136, 1141 n.1 (9th Cir. 1998).\\n\\f48 THURAISSIGIAM v. USDHS\\n\\nserious constitutional problems, and where an alternative\\ninterpretation of the statute is ‘fairly possible.’” St. Cyr,\\n533 U.S. at 299–300 (citation omitted); see also Ramadan,\\n479 F.3d at 654 (“The Supreme Court has been careful to\\nconstrue statutes in light of the Suspension Clause.”).\\nHowever, for us to apply the canon, the statute in question\\nmust be “susceptible of more than one construction.” Clark\\nv. Martinez, 543 U.S. 371, 385 (2005).\\n\\n As explained at length above, we and other courts have\\nconsistently interpreted § 1252(e)(2) to foreclose review of\\nclaims like Thuraissigiam’s. Section 1252(a)(2)(A)(i)\\nprecludes review of “any other cause or claim arising from or\\nrelating to the implementation of or operation of” an\\nexpedited removal order, which clearly bars claims relating\\nto procedural error. We do not think the statute can bear a\\nreading that avoids the constitutional problems it creates.\\n\\n • ! •\\n\\n Therefore, we hold that § 1252(e)(2) violates the\\nSuspension Clause as applied to Thuraissigiam, although we\\ndo not profess to decide in this opinion what right or rights\\nThuraissigiam may vindicate via use of the writ. The district\\ncourt has jurisdiction and, on remand, should exercise that\\njurisdiction to consider Thuraissigiam’s legal challenges to\\nthe procedures leading to his expedited removal order.\\n\\n The judgment of the district court is REVERSED and the\\ncase is REMANDED for further proceedings consistent with\\nthis opinion.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1274,"cells":{"id":{"kind":"number","value":4597699,"string":"4,597,699"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Dw Aina Le'a Development, LLC v. State of Hawaii Land Use Comm"},"decision_date":{"kind":"string","value":"2019-03-07"},"docket_number":{"kind":"string","value":"17-16280"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Richard C. Tallman, Jay S. Bybee, and N. Randy Smith, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"Jay S. Bybee, Presiding Judge\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nDW AINA LE‘A No. 17-16280\\nDEVELOPMENT, LLC,\\n Plaintiff-Appellant, D.C. No.\\n 1:17-cv-00113-SOM-\\n v. RLP\\n\\nSTATE OF HAWAI‘I LAND USE\\nCOMMISSION; DOES, ORDER CERTIFYING\\nGovernmental Units, 1–10; QUESTION TO THE\\nSTATE OF HAWAI‘I, SUPREME COURT OF\\n Defendants-Appellees. HAWAI‘I\\n\\n\\n Filed March 7, 2019\\n\\n Before: Richard C. Tallman, Jay S. Bybee,\\n and N. Randy Smith, Circuit Judges.\\n\\f2 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\n SUMMARY*\\n\\n\\n Civil Rights\\n\\n The panel certified the following question to the Hawai’i\\nSupreme Court:\\n\\n What is the applicable statute of limitations\\n for a claim against the State of Hawai’i\\n alleging an unlawful taking of “[p]rivate\\n property . . . for public use without just\\n compensation,” Haw. Const. art. I, § 20?\\n\\n\\n\\n COUNSEL\\n\\nSang J. Peter Sim, Sim & Record LLP, Bayside, New York,\\nfor Plaintiff-Appellant.\\n\\nEwan Christopher Ravner, Deputy Solicitor General; Clyde\\nJ. Wadsworth, Solicitor General; Douglas S. Chin, Attorney\\nGeneral; Department of the Attorney General, Honolulu,\\nHawai‘i; for Defendants-Appellees.\\n\\n\\n\\n\\n *\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 3\\n\\n ORDER\\n\\n Pursuant to Haw. Rev. Stat. § 602-5(a)(2) and Hawai‘i\\nRule of Appellate Procedure 13, we respectfully certify the\\nfollowing question to the Hawai‘i Supreme Court:\\n\\n What is the applicable statute of limitations\\n for a claim against the State of Hawai‘i\\n alleging an unlawful taking of “[p]rivate\\n property . . . for public use without just\\n compensation,” Haw. Const. art. I, § 20?\\n\\n This significant question of law is “determinative of the\\ncause” in this court and is not answered by statute or any\\n“clear controlling precedent in the Hawai‘i judicial\\ndecisions.” Haw. R. App. P. 13(a). We therefore respectfully\\nask the Hawai‘i Supreme Court to exercise its discretion to\\naccept and decide it.\\n\\n Below we provide a “statement of facts showing the\\nnature of the cause,” a “statement of prior proceedings in the\\ncase,” the “circumstances out of which the [certified]\\nquestion arises,” and the “question of law to be answered.”\\nHaw. R. App. P. 13(b).\\n\\n I. STATEMENT OF FACTS\\n\\n This appeal arises out of a dispute over the classification\\nof approximately 1,060 acres of land in South Kohala on\\nHawai‘i Island. The facts of the underlying dispute are\\ndetailed in DW Aina Le‘a Development, LLC v. Bridge Aina\\nLe‘a, LLC, 339 P.3d 685, 689–703 (Haw. 2014). We briefly\\nreview them here along with the factual allegations contained\\nin the complaint, which we accept as true for purposes of this\\n\\f4 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nappeal. See Wadsworth v. Talmage, 911 F.3d 994, 995 (9th\\nCir. 2018).\\n\\n In 1989, the State of Hawai‘i Land Use Commission\\n(“Commission”) reclassified the land in South Kohala from\\n“agricultural” use to “urban” use in order to allow for the\\ndevelopment of a residential community. The reclassification\\nwas subject to several conditions, including a condition that\\na certain percentage of the newly constructed residential units\\nbe affordable. The Commission amended that condition over\\nthe years as ownership of the land changed hands, and by\\n2005, the Commission required the landowner, Bridge Aina\\nLe‘a, LLC (“Bridge”), to construct no fewer than\\n385 affordable units, which constituted 20% of the total units\\nto be constructed.\\n\\n In December 2008, the Commission issued a show cause\\norder requiring Bridge to explain why the land should not\\nrevert to its former agricultural use classification. The\\nCommission believed that Bridge and its predecessors-in-\\ninterest had failed “to perform according to the conditions\\nimposed and to the representations and commitments made to\\nthe [Commission] in obtaining reclassification” of the\\nproperty. In addition to responding to the show cause order,\\nBridge informed the Commission in February 2009 that it had\\nentered into a purchase agreement with DW Aina Le‘a\\nDevelopment, LLC (“DW”) that gave DW the right to\\ndevelop the residential community. DW then invested more\\nthan $28 million developing the property, constructing more\\nthan a dozen townhouses and grading the site for roads,\\nutilities, and additional townhouses.\\n\\n Nevertheless, in April 2009, the Commission voted\\nunanimously to return the land to its former agricultural use\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 5\\n\\nclassification. Bridge sought reconsideration, with DW now\\non board as a co-petitioner before the Commission, but those\\nefforts failed. Following numerous hearings, the Commission\\nissued a final, written order on April 25, 2011, reclassifying\\nthe property as agricultural. Shortly thereafter, Bridge and\\nDW sought judicial review of the Commission’s order in state\\ncourt. The Hawai‘i Supreme Court ultimately vacated the\\nCommission’s order, concluding that, because use of the\\nproperty had “substantially commenced,” the Commission\\nimproperly reclassified the property without complying with\\ncertain statutory requirements for doing so. DW Aina Le‘a,\\n339 P.3d at 711–14 (discussing the requirements set forth in\\nHaw. Rev. Stat. § 205-4).1\\n\\n II. STATEMENT OF PRIOR PROCEEDINGS\\n\\n On February 23, 2017, DW filed a complaint in Hawai‘i\\nstate court against the Commission and the State of Hawai‘i\\nasserting takings claims under both the United States\\nConstitution and the Hawai‘i Constitution. DW alleged that\\nthe Commission’s reclassification of the property as\\n\\n 1\\n After the Commission entered its written order, Bridge also filed a\\nsecond lawsuit against the Commission alleging, among other things, that\\nthe reclassification of the land constituted an unconstitutional taking\\nwithout just compensation. That lawsuit, which the State removed to\\nfederal court, was stayed pending the Hawai‘i Supreme Court’s review of\\nthe Commission’s reclassification order. See Bridge Aina Le‘a, LLC v.\\nHaw. Land Use Comm’n, 125 F. Supp. 3d 1051, 1057 (D. Haw. 2015).\\nOnce that decision came down, Bridge’s takings claims proceeded to trial,\\nwhere a “jury found that the State had taken [Bridge’s] property without\\njust compensation.” Bridge Aina Le‘a, LLC v. Haw. Land Use Comm’n,\\nNo. 1:11-cv-00414-SOM-KJM, 2018 WL 3149489, at *1 (D. Haw.\\nJune 27, 2018). Bridge was awarded only nominal damages. See id. For\\nreasons that remain largely unexplained, DW never sought to become a\\nparty in Bridge’s takings suit.\\n\\f6 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nagricultural constituted a regulatory taking of DW’s property\\nwithout just compensation. Specifically, DW alleged that the\\nreclassification increased the purchase price of the property\\nafter DW had agreed to purchase it, destroyed DW’s\\n“sophisticated funding arrangement [with] Asia” for\\ndeveloping the property, and caused DW to sustain other\\nincreased costs and losses in business opportunities. The\\nState removed the case to federal court and moved to dismiss\\nthe complaint as barred by the applicable statute of\\nlimitations.\\n\\n The district court granted the State’s motion to dismiss.2\\nThe parties agreed that DW’s takings claims accrued when\\nthe Commission issued its order on April 25, 2011.\\nRegarding the federal takings claim, the court determined\\nthat, although this claim was not (and could not be) brought\\nagainst the State under 42 U.S.C. § 1983, see Will v. Mich.\\nDep’t of State Police, 491 U.S. 58, 71 (1989), the statute of\\nlimitations that ordinarily governs § 1983 actions—the statute\\nof limitations for general personal injury torts, see Wilson v.\\nGarcia, 471 U.S. 261, 276–79 (1985)—applies. The\\n\\n 2\\n The district court also addressed several threshold issues, none of\\nwhich is at issue in this appeal. First, the court concluded that DW’s\\nfederal takings claim was not barred by the exhaustion requirement set\\nforth in Williamson County Regional Planning Commission v. Hamilton\\nBank, 473 U.S. 172, 192–97 (1985), noting that the State removed the case\\nto federal court and raised no jurisdictional or ripeness argument in doing\\nso. Second, the court concluded that the State waived its Eleventh\\nAmendment sovereign immunity from suit in federal court by voluntarily\\nremoving the case to federal court. See Lapides v. Bd. of Regents of Univ.\\nSys. of Ga., 535 U.S. 613, 618–20 (2002). And third, the court concluded\\nthat it could retain supplemental jurisdiction over DW’s state takings\\nclaim despite having dismissed the federal takings claim. See 28 U.S.C.\\n§ 1367. The parties do not challenge any of those conclusions on appeal,\\nand we accordingly express no opinion on them.\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 7\\n\\nlimitations period for personal injury torts in Hawai‘i is two\\nyears. Haw. Rev. Stat. § 657-7. The court thus concluded\\nthat DW’s federal takings claim is time-barred. The court\\nalso concluded in the alternative that DW’s federal takings\\nclaim is time-barred under Haw. Rev. Stat. § 661-5, which\\nprovides a two-year limitations period for “[e]very claim\\nagainst the State, cognizable under [Part I of Chapter 661].”\\nIn light of these two statutes, the court rejected DW’s reliance\\non the six-year “catch-all” statute of limitations, Haw. Rev.\\nStat. § 657-1(4). Finally, the court concluded that, like the\\nfederal takings claim, DW’s state takings claim under the\\nHawai‘i Constitution is time-barred under either Haw. Rev.\\nStat. §§ 657-7 or 661-5. The court thus dismissed this action\\nin its entirety, and DW filed a timely notice of appeal.\\n\\n III. LEGAL CIRCUMSTANCES\\n\\n On appeal, the parties do not dispute the relevant dates for\\nlimitations purposes—DW’s takings claims accrued on\\nApril 25, 2011, and DW filed its complaint on February 23,\\n2017, more than five years later. The only issue is identifying\\nthe applicable statute of limitations. And although the parties\\nfocus on DW’s federal takings claim, DW’s state takings\\nclaim remains at issue, and the applicable limitations period\\nmay control its federal takings claim as well.3 Accordingly,\\n\\n\\n 3\\n There is no federal statute of limitations for federal takings claims\\nbrought against a state. When, as here, there is no “controlling federal\\nlimitations period, the general rule is that a state limitations period for an\\nanalogous cause of action is borrowed and applied to the federal claim,\\nprovided that the application of the state statute would not be inconsistent\\nwith underlying federal policies.” County of Oneida v. Oneida Indian\\nNation, 470 U.S. 226, 240 (1985). In this case, the most analogous cause\\nof action would be an inverse condemnation action under state law. And\\neven if we, like the district court, were to accept the State’s argument that\\n\\f8 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nin all events, determining the correct statute of limitations for\\nDW’s state takings claim is outcome determinative for some,\\nand perhaps all, of DW’s claims.\\n\\n Under the Hawai‘i Constitution, “[p]rivate property shall\\nnot be taken or damaged for public use without just\\ncompensation.” Haw. Const. art. I, § 20. Although\\nproceedings for the condemnation of private property are\\ngoverned by statute, see Haw. Rev. Stat. § 101-1 et seq., an\\n“inverse condemnation” action—i.e., an action by a property\\nowner “to recover the value of property that has been taken\\nby the government . . . without exercising the power of\\neminent domain”—is “not specifically provided by statute,”\\nLeone v. County of Maui, 284 P.3d 956, 962 (Haw. Ct. App.\\n2012). Nor do the Hawai‘i statutes prescribe a limitations\\nperiod for bringing an inverse condemnation action. DW\\nargues that the six-year “catch-all” limitations period in Haw.\\nRev. Stat. § 657-1(4) applies, while the State argues that the\\ntwo-year limitations periods in either Haw. Rev. Stat.\\n§§ 657-7 or 661-5 applies.\\n\\n The Hawai‘i Supreme Court has not had occasion to\\ndecide this issue, and other states that have addressed it have\\ntaken various approaches in doing so. See, e.g., Klumpp v.\\nBorough of Avalon, 997 A.2d 967, 977 (N.J. 2010)\\n(describing the divergent views among states in which “there\\nis no statute of limitations on point” for inverse condemnation\\n\\n\\nthe two-year limitations period for personal injury claims, Haw. Rev. Stat.\\n§ 657-7, applies to DW’s federal takings claim based on the reasoning in\\nWilson, 471 U.S. at 275–79, and Van Strum v. Lawn, 940 F.2d 406,\\n408–10 (9th Cir. 1991), the timeliness of DW’s state takings claim would\\nremain at issue. We accordingly do not address the limitations period\\napplicable to DW’s federal takings claim at this time.\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 9\\n\\nactions); see also 27 Am. Jur. 2d Eminent Domain § 732\\n(Feb. 2019); Charles C. Marvel, Annotation, State Statute of\\nLimitations Applicable to Inverse Condemnation or Similar\\nProceedings by Landowner to Obtain Compensation for\\nDirect Appropriation of Land Without the Institution or\\nConclusion of Formal Proceedings Against Specific Owner,\\n26 A.L.R.4th 68 (1983 & Supp. 2018). Ordinarily, “[w]hen\\nthe highest court of a state has not directly spoken on a matter\\nof state law,” we “use [our] ‘own best judgment in predicting\\nhow the state’s highest court would decide the case.’”\\nT-Mobile USA Inc. v. Selective Ins. Co. of Am., 908 F.3d 581,\\n586 (9th Cir. 2018) (quoting Fiorito Bros. v. Fruehauf Corp.,\\n747 F.2d 1309, 1314 (9th Cir. 1984)). But given the sparse\\nHawai‘i case law and the variety of approaches taken in other\\njurisdictions, we are unable to conclude with any certainty\\nwhich statute of limitations the Hawai‘i Supreme Court\\nwould apply.\\n\\nA. Section 657-1(4)\\n\\n According to DW, the applicable statute of limitations is\\nHaw. Rev. Stat. § 657-1(4), which imposes a six-year\\nlimitations period on “[p]ersonal actions of any nature\\nwhatsoever not specifically covered by the laws of the State.”\\nBut for this “catch-all” provision to apply, the claim must not\\nbe “governed by a specific limitations period.” Au v. Au,\\n626 P.2d 173, 179 (Haw. 1981). In this case, the State has\\nidentified two specific statutes of limitations that it claims\\ngovern DW’s takings claim—Haw. Rev. Stat. §§ 661-5 and\\n657-7. We accordingly address each statute in turn.\\n\\f10 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nB. Section 661-5\\n\\n The State argues that the applicable limitations period is\\nHaw. Rev. Stat. § 661-5, which provides: “Every claim\\nagainst the State, cognizable under this part, shall be forever\\nbarred unless the action is commenced within two years after\\nthe claim first accrues.” Chapter 661 is titled “Actions by and\\nAgainst the State,” and Part I vests state courts with\\njurisdiction to hear “[a]ll claims against the State founded\\nupon any statute of the State; upon any rule of an executive\\ndepartment; or upon any contract, expressed or implied, with\\nthe State, and all claims which may be referred to any such\\ncourt by the legislature.” Id. § 661-1(1). Although DW’s\\ntakings claim is a claim “against the State,” the State does\\nvery little to explain how this claim, brought under the\\nHawai‘i Constitution, is “founded upon” a state “statute,”\\nexecutive “rule,” or “contract.”\\n\\n The State advances two primary arguments. First, the\\nState relies on a footnote in Maunalua Bay Beach Ohana 28\\nv. State, 222 P.3d 441, 458 n.12 (Haw. Ct. App. 2009), a\\ndecision from the Hawai‘i Intermediate Court of Appeals that\\nthe State offers as “the only Hawai‘i case addressing the\\nlimitations period for takings claims.” In that case, the\\nplaintiffs filed an “inverse-condemnation lawsuit” against the\\nState challenging a statute governing ownership of future\\naccreted land. Id. at 442–43. The plaintiffs claimed that the\\nstatute constituted a taking of their “right to future accretions”\\nwithout just compensation “and thereby violated article I,\\nsection 20 of the Hawai‘i State Constitution.” Id. at 460. In\\nthe “Procedural History” section of the decision, the court\\nnoted that the plaintiffs filed their lawsuit “one day shy of\\ntwo years from the date of [the statute’s] enactment.” Id. at\\n458 (capitalization altered). And following that observation,\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 11\\n\\nthe court inserted a footnote: “Pursuant to [§ 661-5], ‘[e]very\\nclaim against the State, cognizable under this chapter, shall be\\nforever barred unless the action is commenced within two\\nyears after the claim first accrues.’” Id. at 458 n.12 (second\\nalteration in original) (quoting Haw. Rev. Stat. § 661-5\\n(1993)).4\\n\\n We are skeptical of the State’s reliance on Maunalua Bay.\\nThe State is correct that, although we are not bound by\\n“decisions of [a] state’s intermediate appellate court” when\\ndeciding an unresolved issue of state law, we can look to such\\ndecisions “as guidance” in “predict[ing] how the highest state\\ncourt would decide the issue.” In re NCAA Student-Athlete\\nName & Likeness Licensing Litig., 724 F.3d 1268, 1278 (9th\\nCir. 2013) (citation omitted). But Maunalua Bay provides\\nvery little guidance, as the timeliness of the lawsuit was not\\nat issue in that case. The court thus had no occasion to\\nexplain how a takings claim based on the Hawai‘i\\nConstitution is “founded upon” a state “statute,” executive\\n“rule,” or “contract,” Haw. Rev. Stat. § 661-1(1), such that it\\nis “cognizable under” Part I of Chapter 661, id. § 661-5.\\n\\n In fact, a decision from the Hawai‘i Supreme Court seems\\nto reject that conclusion. In Kaho‘ohanohano v. State,\\n162 P.3d 696 (Haw. 2007), the plaintiffs argued that a statute\\nreducing contributions to a state employee retirement system\\nwas “unconstitutional as being violative of article XVI,\\nsection 2 of the Hawai‘i Constitution,” id. at 732 (emphasis\\n\\n\\n 4\\n In 2016, the Hawai‘i State Legislature amended § 661-5 by\\nreplacing the phrase “cognizable under this chapter” with the phrase\\n“cognizable under this part.” 2016 Haw. Sess. Laws 109. This technical\\namendment makes no difference here, as the only other “part” of Chapter\\n661 governs false claims actions.\\n\\f12 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nand internal alterations omitted), which provides that\\nmembership in an employee retirement system “shall be a\\ncontractual relationship, the accrued benefits of which shall\\nnot be diminished or impaired,” Haw. Const. art. XVI, § 2.\\nThe Hawai‘i Supreme Court held that the two-year limitations\\nperiod in § 661-5 did not apply to these “constitutional\\nclaims,” as they were “plainly not ‘founded upon any statute\\nof the State; or upon any regulation of an executive\\ndepartment; or upon any contract,’” and thus “not ‘cognizable\\nunder [Chapter 661].’” Kaho‘ohanohano, 162 P.3d at 732\\n(citations omitted); see also Kaleikini v. Yoshioka, 304 P.3d\\n252, 266 (Haw. 2013) (“[C]laims based on the constitution\\nare not ‘founded upon any statute of the State.’” (internal\\nalteration omitted) (quoting Kaho‘ohanohano, 162 P.3d at\\n732)). And although the State correctly observes that\\nKaho‘ohanohano did not involve a constitutional takings\\nclaim, the reasoning in that decision does not appear to be\\nbased on the type of constitutional claim but on the fact that\\nconstitutional claims are “plainly not ‘founded upon’” a state\\nstatute, rule, or contract. 162 P.3d at 732 (quoting Haw. Rev.\\nStat. § 661-1(1)).\\n\\n Second, the State argues that DW’s takings claim should\\nbe “characterized as premised upon an implied contract\\nbetween the State and the landowner.” But the State does not\\nsupport this argument with Hawai‘i law; instead, the State\\nrelies on cases from other jurisdictions that characterize\\ntakings claims as contractual in nature for statute-of-\\nlimitations purposes. See Hager v. City of Devils Lake,\\n773 N.W.2d 420, 432 (N.D. 2009); Richmeade, L.P. v. City\\nof Richmond, 594 S.E.2d 606, 608–10 (Va. 2004); State ex\\nrel. R.T.G., Inc. v. State, 780 N.E.2d 998, 1005 (Ohio 2002);\\nHunter v. City of Mobile, 13 So. 2d 656, 659 (Ala. 1943).\\nAlthough these cases might support the State’s proposed rule,\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 13\\n\\nthat rule has not been uniformly adopted. Indeed, at least one\\nstate supreme court has expressly rejected it, reasoning that\\n“the right to recover damages for property taken does not rest\\nsolely upon a contract . . . but rests primarily upon a vested\\nconstitutional right.” Hiji v. City of Garnett, 804 P.2d 950,\\n958 (Kan. 1991).\\n\\n If anything, the State’s implied-contract theory is in\\nsignificant tension with Kaho‘ohanohano, which concluded\\nthat the plaintiffs’ constitutional claims were “not ‘founded\\nupon . . . any contract,’” 162 P.3d at 732 (quoting Haw. Rev.\\nStat. § 661-1(1)), even though the constitutional provision at\\nissue in that case expressly refers to “a contractual\\nrelationship,” Haw. Const. art. XVI, § 2. Thus, although the\\nHawai‘i Supreme Court could decide—as Maunalua Bay\\nsuggests—that constitutional takings claims are governed by\\nthe two-year limitations period in § 661-5, the reasoning in\\nKaho‘ohanohano seems to counsel against that conclusion.\\n\\nC. Section 657-7\\n\\n The State argues in the alternative that the applicable\\nstatute of limitations is Haw. Rev. Stat. § 657-7, which\\nprovides a two-year limitations period for “[a]ctions for the\\nrecovery of compensation for damage or injury to persons or\\nproperty.” According to the State, the fact that the word\\n“compensation” appears in both § 657-7 and the Takings\\nClause of the Hawai‘i Constitution “alone is enough to show\\nthat the statute is applicable to claims under the takings\\nclause.”\\n\\n The State’s focus on the word “compensation” ignores the\\nremainder of § 657-7’s text. That a plaintiff is seeking “just\\ncompensation” for the taking of private property, Haw. Const.\\n\\f14 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nart. I, § 20, does not necessarily mean that the plaintiff is\\nseeking “compensation for damage or injury to persons or\\nproperty,” Haw. Rev. Stat. § 657-7 (emphasis added). The\\nrelevant question is whether an unconstitutional taking of\\nprivate property without just compensation amounts to\\n“damage or injury to . . . property.”\\n\\n The Hawai‘i Supreme Court has interpreted § 657-7 to\\napply to “claims for damages resulting from physical injury\\nto persons or physical injury to tangible interests in property,”\\nbut not to claims involving a “nonphysical injury to an\\nintangible interest of [the] plaintiff.” Au, 626 P.2d at 178\\n(emphasis added) (quoting Higa v. Mirikitani, 517 P.2d 1, 3\\n(Haw. 1973)). Thus, with respect to real property, the statute\\nhas been applied to actions alleging “trespass quare clausum\\nfregit,”5 Chin Kee v. Kaeleku Sugar Co., 30 Haw. 17, 19–21\\n(1927); Kauha v. Palolo Land & Improvement Co., 20 Haw.\\n237, 239 (1910), actions alleging physical damage to land due\\nto the negligence of a neighboring property owner, Mauian\\nHotel, Inc. v. Maui Pineapple Co., 481 P.2d 310, 312 & n.1\\n(Haw. 1971); Basque v. Yuk Lin Liau, 441 P.2d 636, 637\\n(Haw. 1968), and an action alleging physical destruction of\\ncrops due to the diversion of a water source necessary for\\nirrigation, Wong Nin v. City & Cty. of Honolulu, 33 Haw.\\n379, 380 (1935).\\n\\n DW’s takings claim, by contrast, is predicated on the\\nState’s improper reclassification of DW’s property from\\nurban use to agricultural use. The injury is the State’s\\nregulatory taking of property without just compensation, not\\n\\n 5\\n This is a tort action “to recover damages resulting from another’s\\nunlawful entry on one’s land that is visibly enclosed.” Trespass Quare\\nClausum Fregit, Black’s Law Dictionary (10th ed. 2014).\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 15\\n\\nphysical damage to the property or trespass upon it. This\\ndistinction has led courts in other jurisdictions to conclude\\nthat a “statute of limitations for ‘injuries to persons and\\nproperty’ is not applicable to inverse condemnation actions.”\\nHart v. City of Detroit, 331 N.W.2d 438, 444 (Mich. 1982);\\nsee also Marvel, Annotation, 26 A.L.R.4th 68, § 7[b]\\n(collecting cases in which “statutes of limitation pertaining\\ngenerally to trespass actions or actions for injury to property\\nwere held inapplicable to actions for compensation for the\\ntaking of property without compensation”). In light of the\\nHawai‘i Supreme Court’s insistence on a “physical injury to\\ntangible interests in property,” Au, 626 P.2d at 178 (emphasis\\nadded) (citation omitted), the Hawai‘i Supreme Court could\\nreasonably conclude that § 657-7 does not apply to a\\nregulatory takings claim brought in an inverse condemnation\\naction.\\n\\n * * *\\n\\n The State may be correct in its assertion that either\\n§ 661-5 or § 657-7 provides the applicable limitations period\\nfor DW’s state takings claim under the Hawai‘i Constitution.\\nBut we believe that the best course of action is to certify that\\nquestion to the Hawai‘i Supreme Court.6 Not only is the\\nquestion dispositive in this case and unanswered by Hawai‘i\\nauthority, see Haw. R. App. P. 13(a), but it is also a\\nsignificant question of state law that could “have far-reaching\\nconsequences for the people and businesses of Hawai‘i,”\\nApana v. TIG Ins. Co., 574 F.3d 679, 684 (9th Cir. 2009). In\\nour view, allowing the Hawai‘i Supreme Court to resolve this\\nquestion after analyzing the competing considerations as well\\n\\n 6\\n Although the parties do not request certification, we may “certify a\\nquestion sua sponte.” T-Mobile, 908 F.3d at 587 n.6.\\n\\f16 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nas the views of other jurisdictions—rather than to trying “to\\npredict uncertain [state] law” in the first instance—will foster\\nthe “cooperative judicial federalism” that the certification\\nprocedure is designed to promote. Lehman Bros. v. Schein,\\n416 U.S. 386, 391 (1974).\\n\\n IV. CERTIFIED QUESTION\\n\\n In light of the foregoing, we respectfully certify the\\nfollowing question to the Hawai‘i Supreme Court:\\n\\n What is the applicable statute of limitations\\n for a claim against the State of Hawai‘i\\n alleging an unlawful taking of “[p]rivate\\n property . . . for public use without just\\n compensation,” Haw. Const. art. I, § 20?\\n\\n We do not intend our phrasing of the question to restrict\\nthe Hawai‘i Supreme Court’s consideration of the relevant\\nissues. If the Hawai‘i Supreme Court accepts review of the\\ncertified question, “it may reformulate the [question] in light\\nof the parties’ contentions or other relevant considerations.”\\nApana, 574 F.3d at 684.\\n\\n V. ORDER\\n\\n The clerk of our court is hereby ordered to transmit a\\ncopy of this order to the Hawai‘i Supreme Court under\\nofficial seal of the United States Court of Appeals for the\\nNinth Circuit. See Haw. R. App. P. 13(c). The clerk is also\\nordered to provide “original or copies of all or any portion of\\nthe record” in this case as “[t]he Hawai‘i Supreme Court may,\\nin its discretion, require.” Id.\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 17\\n\\n Further proceedings in this court are stayed pending the\\nHawai‘i Supreme Court’s decision. This case is withdrawn\\nfrom submission and the clerk is directed to administratively\\nclose this docket, pending further order from this court. The\\nparties shall file a joint notice in this court within one week\\nof the Hawai‘i Supreme Court’s decision to accept or reject\\ncertification. If the Hawai‘i Supreme Court accepts the\\ncertified question, the parties shall file a joint status report to\\nthis court every six months after the date of acceptance, or\\nmore frequently if circumstances warrant. This panel retains\\njurisdiction over further proceedings in this court.\\n\\n SO ORDERED.\\n\\n FOR THE COURT:\\n\\n\\n Jay S. Bybee, Presiding Judge\\n U.S. Court of Appeals for the\\n Ninth Circuit\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4374952/\", \"author_raw\": \"Jay S. Bybee, Presiding Judge\"}]}"},"judge1":{"kind":"string","value":"RICHARD C TALLMAN"},"judge2":{"kind":"string","value":"JAY S BYBEE"},"judge3":{"kind":"string","value":"N RANDY SMITH"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4597699/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Dw Aina Le'a Development, LLC v. State of Hawaii Land Use Comm\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-07\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nDW AINA LE‘A No. 17-16280\nDEVELOPMENT, LLC,\n Plaintiff-Appellant, D.C. No.\n 1:17-cv-00113-SOM-\n v. RLP\n\nSTATE OF HAWAI‘I LAND USE\nCOMMISSION; DOES, ORDER CERTIFYING\nGovernmental Units, 1–10; QUESTION TO THE\nSTATE OF HAWAI‘I, SUPREME COURT OF\n Defendants-Appellees. HAWAI‘I\n\n\n Filed March 7, 2019\n\n Before: Richard C. Tallman, Jay S. Bybee,\n and N. Randy Smith, Circuit Judges.\n\f2 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\n\n SUMMARY*\n\n\n Civil Rights\n\n The panel certified the following question to the Hawai’i\nSupreme Court:\n\n What is the applicable statute of limitations\n for a claim against the State of Hawai’i\n alleging an unlawful taking of “[p]rivate\n property . . . for public use without just\n compensation,” Haw. Const. art. I, § 20?\n\n\n\n COUNSEL\n\nSang J. Peter Sim, Sim & Record LLP, Bayside, New York,\nfor Plaintiff-Appellant.\n\nEwan Christopher Ravner, Deputy Solicitor General; Clyde\nJ. Wadsworth, Solicitor General; Douglas S. Chin, Attorney\nGeneral; Department of the Attorney General, Honolulu,\nHawai‘i; for Defendants-Appellees.\n\n\n\n\n *\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 3\n\n ORDER\n\n Pursuant to Haw. Rev. Stat. § 602-5(a)(2) and Hawai‘i\nRule of Appellate Procedure 13, we respectfully certify the\nfollowing question to the Hawai‘i Supreme Court:\n\n What is the applicable statute of limitations\n for a claim against the State of Hawai‘i\n alleging an unlawful taking of “[p]rivate\n property . . . for public use without just\n compensation,” Haw. Const. art. I, § 20?\n\n This significant question of law is “determinative of the\ncause” in this court and is not answered by statute or any\n“clear controlling precedent in the Hawai‘i judicial\ndecisions.” Haw. R. App. P. 13(a). We therefore respectfully\nask the Hawai‘i Supreme Court to exercise its discretion to\naccept and decide it.\n\n Below we provide a “statement of facts showing the\nnature of the cause,” a “statement of prior proceedings in the\ncase,” the “circumstances out of which the [certified]\nquestion arises,” and the “question of law to be answered.”\nHaw. R. App. P. 13(b).\n\n I. STATEMENT OF FACTS\n\n This appeal arises out of a dispute over the classification\nof approximately 1,060 acres of land in South Kohala on\nHawai‘i Island. The facts of the underlying dispute are\ndetailed in DW Aina Le‘a Development, LLC v. Bridge Aina\nLe‘a, LLC, 339 P.3d 685, 689–703 (Haw. 2014). We briefly\nreview them here along with the factual allegations contained\nin the complaint, which we accept as true for purposes of this\n\f4 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\n\nappeal. See Wadsworth v. Talmage, 911 F.3d 994, 995 (9th\nCir. 2018).\n\n In 1989, the State of Hawai‘i Land Use Commission\n(“Commission”) reclassified the land in South Kohala from\n“agricultural” use to “urban” use in order to allow for the\ndevelopment of a residential community. The reclassification\nwas subject to several conditions, including a condition that\na certain percentage of the newly constructed residential units\nbe affordable. The Commission amended that condition over\nthe years as ownership of the land changed hands, and by\n2005, the Commission required the landowner, Bridge Aina\nLe‘a, LLC (“Bridge”), to construct no fewer than\n385 affordable units, which constituted 20% of the total units\nto be constructed.\n\n In December 2008, the Commission issued a show cause\norder requiring Bridge to explain why the land should not\nrevert to its former agricultural use classification. The\nCommission believed that Bridge and its predecessors-in-\ninterest had failed “to perform according to the conditions\nimposed and to the representations and commitments made to\nthe [Commission] in obtaining reclassification” of the\nproperty. In addition to responding to the show cause order,\nBridge informed the Commission in February 2009 that it had\nentered into a purchase agreement with DW Aina Le‘a\nDevelopment, LLC (“DW”) that gave DW the right to\ndevelop the residential community. DW then invested more\nthan $28 million developing the property, constructing more\nthan a dozen townhouses and grading the site for roads,\nutilities, and additional townhouses.\n\n Nevertheless, in April 2009, the Commission voted\nunanimously to return the land to its former agricultural use\n\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 5\n\nclassification. Bridge sought reconsideration, with DW now\non board as a co-petitioner before the Commission, but those\nefforts failed. Following numerous hearings, the Commission\nissued a final, written order on April 25, 2011, reclassifying\nthe property as agricultural. Shortly thereafter, Bridge and\nDW sought judicial review of the Commission’s order in state\ncourt. The Hawai‘i Supreme Court ultimately vacated the\nCommission’s order, concluding that, because use of the\nproperty had “substantially commenced,” the Commission\nimproperly reclassified the property without complying with\ncertain statutory requirements for doing so. DW Aina Le‘a,\n339 P.3d at 711–14 (discussing the requirements set forth in\nHaw. Rev. Stat. § 205-4).1\n\n II. STATEMENT OF PRIOR PROCEEDINGS\n\n On February 23, 2017, DW filed a complaint in Hawai‘i\nstate court against the Commission and the State of Hawai‘i\nasserting takings claims under both the United States\nConstitution and the Hawai‘i Constitution. DW alleged that\nthe Commission’s reclassification of the property as\n\n 1\n After the Commission entered its written order, Bridge also filed a\nsecond lawsuit against the Commission alleging, among other things, that\nthe reclassification of the land constituted an unconstitutional taking\nwithout just compensation. That lawsuit, which the State removed to\nfederal court, was stayed pending the Hawai‘i Supreme Court’s review of\nthe Commission’s reclassification order. See Bridge Aina Le‘a, LLC v.\nHaw. Land Use Comm’n, 125 F. Supp. 3d 1051, 1057 (D. Haw. 2015).\nOnce that decision came down, Bridge’s takings claims proceeded to trial,\nwhere a “jury found that the State had taken [Bridge’s] property without\njust compensation.” Bridge Aina Le‘a, LLC v. Haw. Land Use Comm’n,\nNo. 1:11-cv-00414-SOM-KJM, 2018 WL 3149489, at *1 (D. Haw.\nJune 27, 2018). Bridge was awarded only nominal damages. See id. For\nreasons that remain largely unexplained, DW never sought to become a\nparty in Bridge’s takings suit.\n\f6 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\n\nagricultural constituted a regulatory taking of DW’s property\nwithout just compensation. Specifically, DW alleged that the\nreclassification increased the purchase price of the property\nafter DW had agreed to purchase it, destroyed DW’s\n“sophisticated funding arrangement [with] Asia” for\ndeveloping the property, and caused DW to sustain other\nincreased costs and losses in business opportunities. The\nState removed the case to federal court and moved to dismiss\nthe complaint as barred by the applicable statute of\nlimitations.\n\n The district court granted the State’s motion to dismiss.2\nThe parties agreed that DW’s takings claims accrued when\nthe Commission issued its order on April 25, 2011.\nRegarding the federal takings claim, the court determined\nthat, although this claim was not (and could not be) brought\nagainst the State under 42 U.S.C. § 1983, see Will v. Mich.\nDep’t of State Police, 491 U.S. 58, 71 (1989), the statute of\nlimitations that ordinarily governs § 1983 actions—the statute\nof limitations for general personal injury torts, see Wilson v.\nGarcia, 471 U.S. 261, 276–79 (1985)—applies. The\n\n 2\n The district court also addressed several threshold issues, none of\nwhich is at issue in this appeal. First, the court concluded that DW’s\nfederal takings claim was not barred by the exhaustion requirement set\nforth in Williamson County Regional Planning Commission v. Hamilton\nBank, 473 U.S. 172, 192–97 (1985), noting that the State removed the case\nto federal court and raised no jurisdictional or ripeness argument in doing\nso. Second, the court concluded that the State waived its Eleventh\nAmendment sovereign immunity from suit in federal court by voluntarily\nremoving the case to federal court. See Lapides v. Bd. of Regents of Univ.\nSys. of Ga., 535 U.S. 613, 618–20 (2002). And third, the court concluded\nthat it could retain supplemental jurisdiction over DW’s state takings\nclaim despite having dismissed the federal takings claim. See 28 U.S.C.\n§ 1367. The parties do not challenge any of those conclusions on appeal,\nand we accordingly express no opinion on them.\n\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 7\n\nlimitations period for personal injury torts in Hawai‘i is two\nyears. Haw. Rev. Stat. § 657-7. The court thus concluded\nthat DW’s federal takings claim is time-barred. The court\nalso concluded in the alternative that DW’s federal takings\nclaim is time-barred under Haw. Rev. Stat. § 661-5, which\nprovides a two-year limitations period for “[e]very claim\nagainst the State, cognizable under [Part I of Chapter 661].”\nIn light of these two statutes, the court rejected DW’s reliance\non the six-year “catch-all” statute of limitations, Haw. Rev.\nStat. § 657-1(4). Finally, the court concluded that, like the\nfederal takings claim, DW’s state takings claim under the\nHawai‘i Constitution is time-barred under either Haw. Rev.\nStat. §§ 657-7 or 661-5. The court thus dismissed this action\nin its entirety, and DW filed a timely notice of appeal.\n\n III. LEGAL CIRCUMSTANCES\n\n On appeal, the parties do not dispute the relevant dates for\nlimitations purposes—DW’s takings claims accrued on\nApril 25, 2011, and DW filed its complaint on February 23,\n2017, more than five years later. The only issue is identifying\nthe applicable statute of limitations. And although the parties\nfocus on DW’s federal takings claim, DW’s state takings\nclaim remains at issue, and the applicable limitations period\nmay control its federal takings claim as well.3 Accordingly,\n\n\n 3\n There is no federal statute of limitations for federal takings claims\nbrought against a state. When, as here, there is no “controlling federal\nlimitations period, the general rule is that a state limitations period for an\nanalogous cause of action is borrowed and applied to the federal claim,\nprovided that the application of the state statute would not be inconsistent\nwith underlying federal policies.” County of Oneida v. Oneida Indian\nNation, 470 U.S. 226, 240 (1985). In this case, the most analogous cause\nof action would be an inverse condemnation action under state law. And\neven if we, like the district court, were to accept the State’s argument that\n\f8 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\n\nin all events, determining the correct statute of limitations for\nDW’s state takings claim is outcome determinative for some,\nand perhaps all, of DW’s claims.\n\n Under the Hawai‘i Constitution, “[p]rivate property shall\nnot be taken or damaged for public use without just\ncompensation.” Haw. Const. art. I, § 20. Although\nproceedings for the condemnation of private property are\ngoverned by statute, see Haw. Rev. Stat. § 101-1 et seq., an\n“inverse condemnation” action—i.e., an action by a property\nowner “to recover the value of property that has been taken\nby the government . . . without exercising the power of\neminent domain”—is “not specifically provided by statute,”\nLeone v. County of Maui, 284 P.3d 956, 962 (Haw. Ct. App.\n2012). Nor do the Hawai‘i statutes prescribe a limitations\nperiod for bringing an inverse condemnation action. DW\nargues that the six-year “catch-all” limitations period in Haw.\nRev. Stat. § 657-1(4) applies, while the State argues that the\ntwo-year limitations periods in either Haw. Rev. Stat.\n§§ 657-7 or 661-5 applies.\n\n The Hawai‘i Supreme Court has not had occasion to\ndecide this issue, and other states that have addressed it have\ntaken various approaches in doing so. See, e.g., Klumpp v.\nBorough of Avalon, 997 A.2d 967, 977 (N.J. 2010)\n(describing the divergent views among states in which “there\nis no statute of limitations on point” for inverse condemnation\n\n\nthe two-year limitations period for personal injury claims, Haw. Rev. Stat.\n§ 657-7, applies to DW’s federal takings claim based on the reasoning in\nWilson, 471 U.S. at 275–79, and Van Strum v. Lawn, 940 F.2d 406,\n408–10 (9th Cir. 1991), the timeliness of DW’s state takings claim would\nremain at issue. We accordingly do not address the limitations period\napplicable to DW’s federal takings claim at this time.\n\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 9\n\nactions); see also 27 Am. Jur. 2d Eminent Domain § 732\n(Feb. 2019); Charles C. Marvel, Annotation, State Statute of\nLimitations Applicable to Inverse Condemnation or Similar\nProceedings by Landowner to Obtain Compensation for\nDirect Appropriation of Land Without the Institution or\nConclusion of Formal Proceedings Against Specific Owner,\n26 A.L.R.4th 68 (1983 & Supp. 2018). Ordinarily, “[w]hen\nthe highest court of a state has not directly spoken on a matter\nof state law,” we “use [our] ‘own best judgment in predicting\nhow the state’s highest court would decide the case.’”\nT-Mobile USA Inc. v. Selective Ins. Co. of Am., 908 F.3d 581,\n586 (9th Cir. 2018) (quoting Fiorito Bros. v. Fruehauf Corp.,\n747 F.2d 1309, 1314 (9th Cir. 1984)). But given the sparse\nHawai‘i case law and the variety of approaches taken in other\njurisdictions, we are unable to conclude with any certainty\nwhich statute of limitations the Hawai‘i Supreme Court\nwould apply.\n\nA. Section 657-1(4)\n\n According to DW, the applicable statute of limitations is\nHaw. Rev. Stat. § 657-1(4), which imposes a six-year\nlimitations period on “[p]ersonal actions of any nature\nwhatsoever not specifically covered by the laws of the State.”\nBut for this “catch-all” provision to apply, the claim must not\nbe “governed by a specific limitations period.” Au v. Au,\n626 P.2d 173, 179 (Haw. 1981). In this case, the State has\nidentified two specific statutes of limitations that it claims\ngovern DW’s takings claim—Haw. Rev. Stat. §§ 661-5 and\n657-7. We accordingly address each statute in turn.\n\f10 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\n\nB. Section 661-5\n\n The State argues that the applicable limitations period is\nHaw. Rev. Stat. § 661-5, which provides: “Every claim\nagainst the State, cognizable under this part, shall be forever\nbarred unless the action is commenced within two years after\nthe claim first accrues.” Chapter 661 is titled “Actions by and\nAgainst the State,” and Part I vests state courts with\njurisdiction to hear “[a]ll claims against the State founded\nupon any statute of the State; upon any rule of an executive\ndepartment; or upon any contract, expressed or implied, with\nthe State, and all claims which may be referred to any such\ncourt by the legislature.” Id. § 661-1(1). Although DW’s\ntakings claim is a claim “against the State,” the State does\nvery little to explain how this claim, brought under the\nHawai‘i Constitution, is “founded upon” a state “statute,”\nexecutive “rule,” or “contract.”\n\n The State advances two primary arguments. First, the\nState relies on a footnote in Maunalua Bay Beach Ohana 28\nv. State, 222 P.3d 441, 458 n.12 (Haw. Ct. App. 2009), a\ndecision from the Hawai‘i Intermediate Court of Appeals that\nthe State offers as “the only Hawai‘i case addressing the\nlimitations period for takings claims.” In that case, the\nplaintiffs filed an “inverse-condemnation lawsuit” against the\nState challenging a statute governing ownership of future\naccreted land. Id. at 442–43. The plaintiffs claimed that the\nstatute constituted a taking of their “right to future accretions”\nwithout just compensation “and thereby violated article I,\nsection 20 of the Hawai‘i State Constitution.” Id. at 460. In\nthe “Procedural History” section of the decision, the court\nnoted that the plaintiffs filed their lawsuit “one day shy of\ntwo years from the date of [the statute’s] enactment.” Id. at\n458 (capitalization altered). And following that observation,\n\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 11\n\nthe court inserted a footnote: “Pursuant to [§ 661-5], ‘[e]very\nclaim against the State, cognizable under this chapter, shall be\nforever barred unless the action is commenced within two\nyears after the claim first accrues.’” Id. at 458 n.12 (second\nalteration in original) (quoting Haw. Rev. Stat. § 661-5\n(1993)).4\n\n We are skeptical of the State’s reliance on Maunalua Bay.\nThe State is correct that, although we are not bound by\n“decisions of [a] state’s intermediate appellate court” when\ndeciding an unresolved issue of state law, we can look to such\ndecisions “as guidance” in “predict[ing] how the highest state\ncourt would decide the issue.” In re NCAA Student-Athlete\nName & Likeness Licensing Litig., 724 F.3d 1268, 1278 (9th\nCir. 2013) (citation omitted). But Maunalua Bay provides\nvery little guidance, as the timeliness of the lawsuit was not\nat issue in that case. The court thus had no occasion to\nexplain how a takings claim based on the Hawai‘i\nConstitution is “founded upon” a state “statute,” executive\n“rule,” or “contract,” Haw. Rev. Stat. § 661-1(1), such that it\nis “cognizable under” Part I of Chapter 661, id. § 661-5.\n\n In fact, a decision from the Hawai‘i Supreme Court seems\nto reject that conclusion. In Kaho‘ohanohano v. State,\n162 P.3d 696 (Haw. 2007), the plaintiffs argued that a statute\nreducing contributions to a state employee retirement system\nwas “unconstitutional as being violative of article XVI,\nsection 2 of the Hawai‘i Constitution,” id. at 732 (emphasis\n\n\n 4\n In 2016, the Hawai‘i State Legislature amended § 661-5 by\nreplacing the phrase “cognizable under this chapter” with the phrase\n“cognizable under this part.” 2016 Haw. Sess. Laws 109. This technical\namendment makes no difference here, as the only other “part” of Chapter\n661 governs false claims actions.\n\f12 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\n\nand internal alterations omitted), which provides that\nmembership in an employee retirement system “shall be a\ncontractual relationship, the accrued benefits of which shall\nnot be diminished or impaired,” Haw. Const. art. XVI, § 2.\nThe Hawai‘i Supreme Court held that the two-year limitations\nperiod in § 661-5 did not apply to these “constitutional\nclaims,” as they were “plainly not ‘founded upon any statute\nof the State; or upon any regulation of an executive\ndepartment; or upon any contract,’” and thus “not ‘cognizable\nunder [Chapter 661].’” Kaho‘ohanohano, 162 P.3d at 732\n(citations omitted); see also Kaleikini v. Yoshioka, 304 P.3d\n252, 266 (Haw. 2013) (“[C]laims based on the constitution\nare not ‘founded upon any statute of the State.’” (internal\nalteration omitted) (quoting Kaho‘ohanohano, 162 P.3d at\n732)). And although the State correctly observes that\nKaho‘ohanohano did not involve a constitutional takings\nclaim, the reasoning in that decision does not appear to be\nbased on the type of constitutional claim but on the fact that\nconstitutional claims are “plainly not ‘founded upon’” a state\nstatute, rule, or contract. 162 P.3d at 732 (quoting Haw. Rev.\nStat. § 661-1(1)).\n\n Second, the State argues that DW’s takings claim should\nbe “characterized as premised upon an implied contract\nbetween the State and the landowner.” But the State does not\nsupport this argument with Hawai‘i law; instead, the State\nrelies on cases from other jurisdictions that characterize\ntakings claims as contractual in nature for statute-of-\nlimitations purposes. See Hager v. City of Devils Lake,\n773 N.W.2d 420, 432 (N.D. 2009); Richmeade, L.P. v. City\nof Richmond, 594 S.E.2d 606, 608–10 (Va. 2004); State ex\nrel. R.T.G., Inc. v. State, 780 N.E.2d 998, 1005 (Ohio 2002);\nHunter v. City of Mobile, 13 So. 2d 656, 659 (Ala. 1943).\nAlthough these cases might support the State’s proposed rule,\n\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 13\n\nthat rule has not been uniformly adopted. Indeed, at least one\nstate supreme court has expressly rejected it, reasoning that\n“the right to recover damages for property taken does not rest\nsolely upon a contract . . . but rests primarily upon a vested\nconstitutional right.” Hiji v. City of Garnett, 804 P.2d 950,\n958 (Kan. 1991).\n\n If anything, the State’s implied-contract theory is in\nsignificant tension with Kaho‘ohanohano, which concluded\nthat the plaintiffs’ constitutional claims were “not ‘founded\nupon . . . any contract,’” 162 P.3d at 732 (quoting Haw. Rev.\nStat. § 661-1(1)), even though the constitutional provision at\nissue in that case expressly refers to “a contractual\nrelationship,” Haw. Const. art. XVI, § 2. Thus, although the\nHawai‘i Supreme Court could decide—as Maunalua Bay\nsuggests—that constitutional takings claims are governed by\nthe two-year limitations period in § 661-5, the reasoning in\nKaho‘ohanohano seems to counsel against that conclusion.\n\nC. Section 657-7\n\n The State argues in the alternative that the applicable\nstatute of limitations is Haw. Rev. Stat. § 657-7, which\nprovides a two-year limitations period for “[a]ctions for the\nrecovery of compensation for damage or injury to persons or\nproperty.” According to the State, the fact that the word\n“compensation” appears in both § 657-7 and the Takings\nClause of the Hawai‘i Constitution “alone is enough to show\nthat the statute is applicable to claims under the takings\nclause.”\n\n The State’s focus on the word “compensation” ignores the\nremainder of § 657-7’s text. That a plaintiff is seeking “just\ncompensation” for the taking of private property, Haw. Const.\n\f14 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\n\nart. I, § 20, does not necessarily mean that the plaintiff is\nseeking “compensation for damage or injury to persons or\nproperty,” Haw. Rev. Stat. § 657-7 (emphasis added). The\nrelevant question is whether an unconstitutional taking of\nprivate property without just compensation amounts to\n“damage or injury to . . . property.”\n\n The Hawai‘i Supreme Court has interpreted § 657-7 to\napply to “claims for damages resulting from physical injury\nto persons or physical injury to tangible interests in property,”\nbut not to claims involving a “nonphysical injury to an\nintangible interest of [the] plaintiff.” Au, 626 P.2d at 178\n(emphasis added) (quoting Higa v. Mirikitani, 517 P.2d 1, 3\n(Haw. 1973)). Thus, with respect to real property, the statute\nhas been applied to actions alleging “trespass quare clausum\nfregit,”5 Chin Kee v. Kaeleku Sugar Co., 30 Haw. 17, 19–21\n(1927); Kauha v. Palolo Land & Improvement Co., 20 Haw.\n237, 239 (1910), actions alleging physical damage to land due\nto the negligence of a neighboring property owner, Mauian\nHotel, Inc. v. Maui Pineapple Co., 481 P.2d 310, 312 & n.1\n(Haw. 1971); Basque v. Yuk Lin Liau, 441 P.2d 636, 637\n(Haw. 1968), and an action alleging physical destruction of\ncrops due to the diversion of a water source necessary for\nirrigation, Wong Nin v. City & Cty. of Honolulu, 33 Haw.\n379, 380 (1935).\n\n DW’s takings claim, by contrast, is predicated on the\nState’s improper reclassification of DW’s property from\nurban use to agricultural use. The injury is the State’s\nregulatory taking of property without just compensation, not\n\n 5\n This is a tort action “to recover damages resulting from another’s\nunlawful entry on one’s land that is visibly enclosed.” Trespass Quare\nClausum Fregit, Black’s Law Dictionary (10th ed. 2014).\n\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 15\n\nphysical damage to the property or trespass upon it. This\ndistinction has led courts in other jurisdictions to conclude\nthat a “statute of limitations for ‘injuries to persons and\nproperty’ is not applicable to inverse condemnation actions.”\nHart v. City of Detroit, 331 N.W.2d 438, 444 (Mich. 1982);\nsee also Marvel, Annotation, 26 A.L.R.4th 68, § 7[b]\n(collecting cases in which “statutes of limitation pertaining\ngenerally to trespass actions or actions for injury to property\nwere held inapplicable to actions for compensation for the\ntaking of property without compensation”). In light of the\nHawai‘i Supreme Court’s insistence on a “physical injury to\ntangible interests in property,” Au, 626 P.2d at 178 (emphasis\nadded) (citation omitted), the Hawai‘i Supreme Court could\nreasonably conclude that § 657-7 does not apply to a\nregulatory takings claim brought in an inverse condemnation\naction.\n\n * * *\n\n The State may be correct in its assertion that either\n§ 661-5 or § 657-7 provides the applicable limitations period\nfor DW’s state takings claim under the Hawai‘i Constitution.\nBut we believe that the best course of action is to certify that\nquestion to the Hawai‘i Supreme Court.6 Not only is the\nquestion dispositive in this case and unanswered by Hawai‘i\nauthority, see Haw. R. App. P. 13(a), but it is also a\nsignificant question of state law that could “have far-reaching\nconsequences for the people and businesses of Hawai‘i,”\nApana v. TIG Ins. Co., 574 F.3d 679, 684 (9th Cir. 2009). In\nour view, allowing the Hawai‘i Supreme Court to resolve this\nquestion after analyzing the competing considerations as well\n\n 6\n Although the parties do not request certification, we may “certify a\nquestion sua sponte.” T-Mobile, 908 F.3d at 587 n.6.\n\f16 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\n\nas the views of other jurisdictions—rather than to trying “to\npredict uncertain [state] law” in the first instance—will foster\nthe “cooperative judicial federalism” that the certification\nprocedure is designed to promote. Lehman Bros. v. Schein,\n416 U.S. 386, 391 (1974).\n\n IV. CERTIFIED QUESTION\n\n In light of the foregoing, we respectfully certify the\nfollowing question to the Hawai‘i Supreme Court:\n\n What is the applicable statute of limitations\n for a claim against the State of Hawai‘i\n alleging an unlawful taking of “[p]rivate\n property . . . for public use without just\n compensation,” Haw. Const. art. I, § 20?\n\n We do not intend our phrasing of the question to restrict\nthe Hawai‘i Supreme Court’s consideration of the relevant\nissues. If the Hawai‘i Supreme Court accepts review of the\ncertified question, “it may reformulate the [question] in light\nof the parties’ contentions or other relevant considerations.”\nApana, 574 F.3d at 684.\n\n V. ORDER\n\n The clerk of our court is hereby ordered to transmit a\ncopy of this order to the Hawai‘i Supreme Court under\nofficial seal of the United States Court of Appeals for the\nNinth Circuit. See Haw. R. App. P. 13(c). The clerk is also\nordered to provide “original or copies of all or any portion of\nthe record” in this case as “[t]he Hawai‘i Supreme Court may,\nin its discretion, require.” Id.\n\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 17\n\n Further proceedings in this court are stayed pending the\nHawai‘i Supreme Court’s decision. This case is withdrawn\nfrom submission and the clerk is directed to administratively\nclose this docket, pending further order from this court. The\nparties shall file a joint notice in this court within one week\nof the Hawai‘i Supreme Court’s decision to accept or reject\ncertification. If the Hawai‘i Supreme Court accepts the\ncertified question, the parties shall file a joint status report to\nthis court every six months after the date of acceptance, or\nmore frequently if circumstances warrant. This panel retains\njurisdiction over further proceedings in this court.\n\n SO ORDERED.\n\n FOR THE COURT:\n\n\n Jay S. Bybee, Presiding Judge\n U.S. Court of Appeals for the\n Ninth Circuit\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Dw Aina Le'a Development, LLC v. State of Hawaii Land Use Comm\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-07\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nDW AINA LE‘A No. 17-16280\\nDEVELOPMENT, LLC,\\n Plaintiff-Appellant, D.C. No.\\n 1:17-cv-00113-SOM-\\n v. RLP\\n\\nSTATE OF HAWAI‘I LAND USE\\nCOMMISSION; DOES, ORDER CERTIFYING\\nGovernmental Units, 1–10; QUESTION TO THE\\nSTATE OF HAWAI‘I, SUPREME COURT OF\\n Defendants-Appellees. HAWAI‘I\\n\\n\\n Filed March 7, 2019\\n\\n Before: Richard C. Tallman, Jay S. Bybee,\\n and N. Randy Smith, Circuit Judges.\\n\\f2 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\n SUMMARY*\\n\\n\\n Civil Rights\\n\\n The panel certified the following question to the Hawai’i\\nSupreme Court:\\n\\n What is the applicable statute of limitations\\n for a claim against the State of Hawai’i\\n alleging an unlawful taking of “[p]rivate\\n property . . . for public use without just\\n compensation,” Haw. Const. art. I, § 20?\\n\\n\\n\\n COUNSEL\\n\\nSang J. Peter Sim, Sim & Record LLP, Bayside, New York,\\nfor Plaintiff-Appellant.\\n\\nEwan Christopher Ravner, Deputy Solicitor General; Clyde\\nJ. Wadsworth, Solicitor General; Douglas S. Chin, Attorney\\nGeneral; Department of the Attorney General, Honolulu,\\nHawai‘i; for Defendants-Appellees.\\n\\n\\n\\n\\n *\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 3\\n\\n ORDER\\n\\n Pursuant to Haw. Rev. Stat. § 602-5(a)(2) and Hawai‘i\\nRule of Appellate Procedure 13, we respectfully certify the\\nfollowing question to the Hawai‘i Supreme Court:\\n\\n What is the applicable statute of limitations\\n for a claim against the State of Hawai‘i\\n alleging an unlawful taking of “[p]rivate\\n property . . . for public use without just\\n compensation,” Haw. Const. art. I, § 20?\\n\\n This significant question of law is “determinative of the\\ncause” in this court and is not answered by statute or any\\n“clear controlling precedent in the Hawai‘i judicial\\ndecisions.” Haw. R. App. P. 13(a). We therefore respectfully\\nask the Hawai‘i Supreme Court to exercise its discretion to\\naccept and decide it.\\n\\n Below we provide a “statement of facts showing the\\nnature of the cause,” a “statement of prior proceedings in the\\ncase,” the “circumstances out of which the [certified]\\nquestion arises,” and the “question of law to be answered.”\\nHaw. R. App. P. 13(b).\\n\\n I. STATEMENT OF FACTS\\n\\n This appeal arises out of a dispute over the classification\\nof approximately 1,060 acres of land in South Kohala on\\nHawai‘i Island. The facts of the underlying dispute are\\ndetailed in DW Aina Le‘a Development, LLC v. Bridge Aina\\nLe‘a, LLC, 339 P.3d 685, 689–703 (Haw. 2014). We briefly\\nreview them here along with the factual allegations contained\\nin the complaint, which we accept as true for purposes of this\\n\\f4 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nappeal. See Wadsworth v. Talmage, 911 F.3d 994, 995 (9th\\nCir. 2018).\\n\\n In 1989, the State of Hawai‘i Land Use Commission\\n(“Commission”) reclassified the land in South Kohala from\\n“agricultural” use to “urban” use in order to allow for the\\ndevelopment of a residential community. The reclassification\\nwas subject to several conditions, including a condition that\\na certain percentage of the newly constructed residential units\\nbe affordable. The Commission amended that condition over\\nthe years as ownership of the land changed hands, and by\\n2005, the Commission required the landowner, Bridge Aina\\nLe‘a, LLC (“Bridge”), to construct no fewer than\\n385 affordable units, which constituted 20% of the total units\\nto be constructed.\\n\\n In December 2008, the Commission issued a show cause\\norder requiring Bridge to explain why the land should not\\nrevert to its former agricultural use classification. The\\nCommission believed that Bridge and its predecessors-in-\\ninterest had failed “to perform according to the conditions\\nimposed and to the representations and commitments made to\\nthe [Commission] in obtaining reclassification” of the\\nproperty. In addition to responding to the show cause order,\\nBridge informed the Commission in February 2009 that it had\\nentered into a purchase agreement with DW Aina Le‘a\\nDevelopment, LLC (“DW”) that gave DW the right to\\ndevelop the residential community. DW then invested more\\nthan $28 million developing the property, constructing more\\nthan a dozen townhouses and grading the site for roads,\\nutilities, and additional townhouses.\\n\\n Nevertheless, in April 2009, the Commission voted\\nunanimously to return the land to its former agricultural use\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 5\\n\\nclassification. Bridge sought reconsideration, with DW now\\non board as a co-petitioner before the Commission, but those\\nefforts failed. Following numerous hearings, the Commission\\nissued a final, written order on April 25, 2011, reclassifying\\nthe property as agricultural. Shortly thereafter, Bridge and\\nDW sought judicial review of the Commission’s order in state\\ncourt. The Hawai‘i Supreme Court ultimately vacated the\\nCommission’s order, concluding that, because use of the\\nproperty had “substantially commenced,” the Commission\\nimproperly reclassified the property without complying with\\ncertain statutory requirements for doing so. DW Aina Le‘a,\\n339 P.3d at 711–14 (discussing the requirements set forth in\\nHaw. Rev. Stat. § 205-4).1\\n\\n II. STATEMENT OF PRIOR PROCEEDINGS\\n\\n On February 23, 2017, DW filed a complaint in Hawai‘i\\nstate court against the Commission and the State of Hawai‘i\\nasserting takings claims under both the United States\\nConstitution and the Hawai‘i Constitution. DW alleged that\\nthe Commission’s reclassification of the property as\\n\\n 1\\n After the Commission entered its written order, Bridge also filed a\\nsecond lawsuit against the Commission alleging, among other things, that\\nthe reclassification of the land constituted an unconstitutional taking\\nwithout just compensation. That lawsuit, which the State removed to\\nfederal court, was stayed pending the Hawai‘i Supreme Court’s review of\\nthe Commission’s reclassification order. See Bridge Aina Le‘a, LLC v.\\nHaw. Land Use Comm’n, 125 F. Supp. 3d 1051, 1057 (D. Haw. 2015).\\nOnce that decision came down, Bridge’s takings claims proceeded to trial,\\nwhere a “jury found that the State had taken [Bridge’s] property without\\njust compensation.” Bridge Aina Le‘a, LLC v. Haw. Land Use Comm’n,\\nNo. 1:11-cv-00414-SOM-KJM, 2018 WL 3149489, at *1 (D. Haw.\\nJune 27, 2018). Bridge was awarded only nominal damages. See id. For\\nreasons that remain largely unexplained, DW never sought to become a\\nparty in Bridge’s takings suit.\\n\\f6 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nagricultural constituted a regulatory taking of DW’s property\\nwithout just compensation. Specifically, DW alleged that the\\nreclassification increased the purchase price of the property\\nafter DW had agreed to purchase it, destroyed DW’s\\n“sophisticated funding arrangement [with] Asia” for\\ndeveloping the property, and caused DW to sustain other\\nincreased costs and losses in business opportunities. The\\nState removed the case to federal court and moved to dismiss\\nthe complaint as barred by the applicable statute of\\nlimitations.\\n\\n The district court granted the State’s motion to dismiss.2\\nThe parties agreed that DW’s takings claims accrued when\\nthe Commission issued its order on April 25, 2011.\\nRegarding the federal takings claim, the court determined\\nthat, although this claim was not (and could not be) brought\\nagainst the State under 42 U.S.C. § 1983, see Will v. Mich.\\nDep’t of State Police, 491 U.S. 58, 71 (1989), the statute of\\nlimitations that ordinarily governs § 1983 actions—the statute\\nof limitations for general personal injury torts, see Wilson v.\\nGarcia, 471 U.S. 261, 276–79 (1985)—applies. The\\n\\n 2\\n The district court also addressed several threshold issues, none of\\nwhich is at issue in this appeal. First, the court concluded that DW’s\\nfederal takings claim was not barred by the exhaustion requirement set\\nforth in Williamson County Regional Planning Commission v. Hamilton\\nBank, 473 U.S. 172, 192–97 (1985), noting that the State removed the case\\nto federal court and raised no jurisdictional or ripeness argument in doing\\nso. Second, the court concluded that the State waived its Eleventh\\nAmendment sovereign immunity from suit in federal court by voluntarily\\nremoving the case to federal court. See Lapides v. Bd. of Regents of Univ.\\nSys. of Ga., 535 U.S. 613, 618–20 (2002). And third, the court concluded\\nthat it could retain supplemental jurisdiction over DW’s state takings\\nclaim despite having dismissed the federal takings claim. See 28 U.S.C.\\n§ 1367. The parties do not challenge any of those conclusions on appeal,\\nand we accordingly express no opinion on them.\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 7\\n\\nlimitations period for personal injury torts in Hawai‘i is two\\nyears. Haw. Rev. Stat. § 657-7. The court thus concluded\\nthat DW’s federal takings claim is time-barred. The court\\nalso concluded in the alternative that DW’s federal takings\\nclaim is time-barred under Haw. Rev. Stat. § 661-5, which\\nprovides a two-year limitations period for “[e]very claim\\nagainst the State, cognizable under [Part I of Chapter 661].”\\nIn light of these two statutes, the court rejected DW’s reliance\\non the six-year “catch-all” statute of limitations, Haw. Rev.\\nStat. § 657-1(4). Finally, the court concluded that, like the\\nfederal takings claim, DW’s state takings claim under the\\nHawai‘i Constitution is time-barred under either Haw. Rev.\\nStat. §§ 657-7 or 661-5. The court thus dismissed this action\\nin its entirety, and DW filed a timely notice of appeal.\\n\\n III. LEGAL CIRCUMSTANCES\\n\\n On appeal, the parties do not dispute the relevant dates for\\nlimitations purposes—DW’s takings claims accrued on\\nApril 25, 2011, and DW filed its complaint on February 23,\\n2017, more than five years later. The only issue is identifying\\nthe applicable statute of limitations. And although the parties\\nfocus on DW’s federal takings claim, DW’s state takings\\nclaim remains at issue, and the applicable limitations period\\nmay control its federal takings claim as well.3 Accordingly,\\n\\n\\n 3\\n There is no federal statute of limitations for federal takings claims\\nbrought against a state. When, as here, there is no “controlling federal\\nlimitations period, the general rule is that a state limitations period for an\\nanalogous cause of action is borrowed and applied to the federal claim,\\nprovided that the application of the state statute would not be inconsistent\\nwith underlying federal policies.” County of Oneida v. Oneida Indian\\nNation, 470 U.S. 226, 240 (1985). In this case, the most analogous cause\\nof action would be an inverse condemnation action under state law. And\\neven if we, like the district court, were to accept the State’s argument that\\n\\f8 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nin all events, determining the correct statute of limitations for\\nDW’s state takings claim is outcome determinative for some,\\nand perhaps all, of DW’s claims.\\n\\n Under the Hawai‘i Constitution, “[p]rivate property shall\\nnot be taken or damaged for public use without just\\ncompensation.” Haw. Const. art. I, § 20. Although\\nproceedings for the condemnation of private property are\\ngoverned by statute, see Haw. Rev. Stat. § 101-1 et seq., an\\n“inverse condemnation” action—i.e., an action by a property\\nowner “to recover the value of property that has been taken\\nby the government . . . without exercising the power of\\neminent domain”—is “not specifically provided by statute,”\\nLeone v. County of Maui, 284 P.3d 956, 962 (Haw. Ct. App.\\n2012). Nor do the Hawai‘i statutes prescribe a limitations\\nperiod for bringing an inverse condemnation action. DW\\nargues that the six-year “catch-all” limitations period in Haw.\\nRev. Stat. § 657-1(4) applies, while the State argues that the\\ntwo-year limitations periods in either Haw. Rev. Stat.\\n§§ 657-7 or 661-5 applies.\\n\\n The Hawai‘i Supreme Court has not had occasion to\\ndecide this issue, and other states that have addressed it have\\ntaken various approaches in doing so. See, e.g., Klumpp v.\\nBorough of Avalon, 997 A.2d 967, 977 (N.J. 2010)\\n(describing the divergent views among states in which “there\\nis no statute of limitations on point” for inverse condemnation\\n\\n\\nthe two-year limitations period for personal injury claims, Haw. Rev. Stat.\\n§ 657-7, applies to DW’s federal takings claim based on the reasoning in\\nWilson, 471 U.S. at 275–79, and Van Strum v. Lawn, 940 F.2d 406,\\n408–10 (9th Cir. 1991), the timeliness of DW’s state takings claim would\\nremain at issue. We accordingly do not address the limitations period\\napplicable to DW’s federal takings claim at this time.\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 9\\n\\nactions); see also 27 Am. Jur. 2d Eminent Domain § 732\\n(Feb. 2019); Charles C. Marvel, Annotation, State Statute of\\nLimitations Applicable to Inverse Condemnation or Similar\\nProceedings by Landowner to Obtain Compensation for\\nDirect Appropriation of Land Without the Institution or\\nConclusion of Formal Proceedings Against Specific Owner,\\n26 A.L.R.4th 68 (1983 & Supp. 2018). Ordinarily, “[w]hen\\nthe highest court of a state has not directly spoken on a matter\\nof state law,” we “use [our] ‘own best judgment in predicting\\nhow the state’s highest court would decide the case.’”\\nT-Mobile USA Inc. v. Selective Ins. Co. of Am., 908 F.3d 581,\\n586 (9th Cir. 2018) (quoting Fiorito Bros. v. Fruehauf Corp.,\\n747 F.2d 1309, 1314 (9th Cir. 1984)). But given the sparse\\nHawai‘i case law and the variety of approaches taken in other\\njurisdictions, we are unable to conclude with any certainty\\nwhich statute of limitations the Hawai‘i Supreme Court\\nwould apply.\\n\\nA. Section 657-1(4)\\n\\n According to DW, the applicable statute of limitations is\\nHaw. Rev. Stat. § 657-1(4), which imposes a six-year\\nlimitations period on “[p]ersonal actions of any nature\\nwhatsoever not specifically covered by the laws of the State.”\\nBut for this “catch-all” provision to apply, the claim must not\\nbe “governed by a specific limitations period.” Au v. Au,\\n626 P.2d 173, 179 (Haw. 1981). In this case, the State has\\nidentified two specific statutes of limitations that it claims\\ngovern DW’s takings claim—Haw. Rev. Stat. §§ 661-5 and\\n657-7. We accordingly address each statute in turn.\\n\\f10 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nB. Section 661-5\\n\\n The State argues that the applicable limitations period is\\nHaw. Rev. Stat. § 661-5, which provides: “Every claim\\nagainst the State, cognizable under this part, shall be forever\\nbarred unless the action is commenced within two years after\\nthe claim first accrues.” Chapter 661 is titled “Actions by and\\nAgainst the State,” and Part I vests state courts with\\njurisdiction to hear “[a]ll claims against the State founded\\nupon any statute of the State; upon any rule of an executive\\ndepartment; or upon any contract, expressed or implied, with\\nthe State, and all claims which may be referred to any such\\ncourt by the legislature.” Id. § 661-1(1). Although DW’s\\ntakings claim is a claim “against the State,” the State does\\nvery little to explain how this claim, brought under the\\nHawai‘i Constitution, is “founded upon” a state “statute,”\\nexecutive “rule,” or “contract.”\\n\\n The State advances two primary arguments. First, the\\nState relies on a footnote in Maunalua Bay Beach Ohana 28\\nv. State, 222 P.3d 441, 458 n.12 (Haw. Ct. App. 2009), a\\ndecision from the Hawai‘i Intermediate Court of Appeals that\\nthe State offers as “the only Hawai‘i case addressing the\\nlimitations period for takings claims.” In that case, the\\nplaintiffs filed an “inverse-condemnation lawsuit” against the\\nState challenging a statute governing ownership of future\\naccreted land. Id. at 442–43. The plaintiffs claimed that the\\nstatute constituted a taking of their “right to future accretions”\\nwithout just compensation “and thereby violated article I,\\nsection 20 of the Hawai‘i State Constitution.” Id. at 460. In\\nthe “Procedural History” section of the decision, the court\\nnoted that the plaintiffs filed their lawsuit “one day shy of\\ntwo years from the date of [the statute’s] enactment.” Id. at\\n458 (capitalization altered). And following that observation,\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 11\\n\\nthe court inserted a footnote: “Pursuant to [§ 661-5], ‘[e]very\\nclaim against the State, cognizable under this chapter, shall be\\nforever barred unless the action is commenced within two\\nyears after the claim first accrues.’” Id. at 458 n.12 (second\\nalteration in original) (quoting Haw. Rev. Stat. § 661-5\\n(1993)).4\\n\\n We are skeptical of the State’s reliance on Maunalua Bay.\\nThe State is correct that, although we are not bound by\\n“decisions of [a] state’s intermediate appellate court” when\\ndeciding an unresolved issue of state law, we can look to such\\ndecisions “as guidance” in “predict[ing] how the highest state\\ncourt would decide the issue.” In re NCAA Student-Athlete\\nName & Likeness Licensing Litig., 724 F.3d 1268, 1278 (9th\\nCir. 2013) (citation omitted). But Maunalua Bay provides\\nvery little guidance, as the timeliness of the lawsuit was not\\nat issue in that case. The court thus had no occasion to\\nexplain how a takings claim based on the Hawai‘i\\nConstitution is “founded upon” a state “statute,” executive\\n“rule,” or “contract,” Haw. Rev. Stat. § 661-1(1), such that it\\nis “cognizable under” Part I of Chapter 661, id. § 661-5.\\n\\n In fact, a decision from the Hawai‘i Supreme Court seems\\nto reject that conclusion. In Kaho‘ohanohano v. State,\\n162 P.3d 696 (Haw. 2007), the plaintiffs argued that a statute\\nreducing contributions to a state employee retirement system\\nwas “unconstitutional as being violative of article XVI,\\nsection 2 of the Hawai‘i Constitution,” id. at 732 (emphasis\\n\\n\\n 4\\n In 2016, the Hawai‘i State Legislature amended § 661-5 by\\nreplacing the phrase “cognizable under this chapter” with the phrase\\n“cognizable under this part.” 2016 Haw. Sess. Laws 109. This technical\\namendment makes no difference here, as the only other “part” of Chapter\\n661 governs false claims actions.\\n\\f12 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nand internal alterations omitted), which provides that\\nmembership in an employee retirement system “shall be a\\ncontractual relationship, the accrued benefits of which shall\\nnot be diminished or impaired,” Haw. Const. art. XVI, § 2.\\nThe Hawai‘i Supreme Court held that the two-year limitations\\nperiod in § 661-5 did not apply to these “constitutional\\nclaims,” as they were “plainly not ‘founded upon any statute\\nof the State; or upon any regulation of an executive\\ndepartment; or upon any contract,’” and thus “not ‘cognizable\\nunder [Chapter 661].’” Kaho‘ohanohano, 162 P.3d at 732\\n(citations omitted); see also Kaleikini v. Yoshioka, 304 P.3d\\n252, 266 (Haw. 2013) (“[C]laims based on the constitution\\nare not ‘founded upon any statute of the State.’” (internal\\nalteration omitted) (quoting Kaho‘ohanohano, 162 P.3d at\\n732)). And although the State correctly observes that\\nKaho‘ohanohano did not involve a constitutional takings\\nclaim, the reasoning in that decision does not appear to be\\nbased on the type of constitutional claim but on the fact that\\nconstitutional claims are “plainly not ‘founded upon’” a state\\nstatute, rule, or contract. 162 P.3d at 732 (quoting Haw. Rev.\\nStat. § 661-1(1)).\\n\\n Second, the State argues that DW’s takings claim should\\nbe “characterized as premised upon an implied contract\\nbetween the State and the landowner.” But the State does not\\nsupport this argument with Hawai‘i law; instead, the State\\nrelies on cases from other jurisdictions that characterize\\ntakings claims as contractual in nature for statute-of-\\nlimitations purposes. See Hager v. City of Devils Lake,\\n773 N.W.2d 420, 432 (N.D. 2009); Richmeade, L.P. v. City\\nof Richmond, 594 S.E.2d 606, 608–10 (Va. 2004); State ex\\nrel. R.T.G., Inc. v. State, 780 N.E.2d 998, 1005 (Ohio 2002);\\nHunter v. City of Mobile, 13 So. 2d 656, 659 (Ala. 1943).\\nAlthough these cases might support the State’s proposed rule,\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 13\\n\\nthat rule has not been uniformly adopted. Indeed, at least one\\nstate supreme court has expressly rejected it, reasoning that\\n“the right to recover damages for property taken does not rest\\nsolely upon a contract . . . but rests primarily upon a vested\\nconstitutional right.” Hiji v. City of Garnett, 804 P.2d 950,\\n958 (Kan. 1991).\\n\\n If anything, the State’s implied-contract theory is in\\nsignificant tension with Kaho‘ohanohano, which concluded\\nthat the plaintiffs’ constitutional claims were “not ‘founded\\nupon . . . any contract,’” 162 P.3d at 732 (quoting Haw. Rev.\\nStat. § 661-1(1)), even though the constitutional provision at\\nissue in that case expressly refers to “a contractual\\nrelationship,” Haw. Const. art. XVI, § 2. Thus, although the\\nHawai‘i Supreme Court could decide—as Maunalua Bay\\nsuggests—that constitutional takings claims are governed by\\nthe two-year limitations period in § 661-5, the reasoning in\\nKaho‘ohanohano seems to counsel against that conclusion.\\n\\nC. Section 657-7\\n\\n The State argues in the alternative that the applicable\\nstatute of limitations is Haw. Rev. Stat. § 657-7, which\\nprovides a two-year limitations period for “[a]ctions for the\\nrecovery of compensation for damage or injury to persons or\\nproperty.” According to the State, the fact that the word\\n“compensation” appears in both § 657-7 and the Takings\\nClause of the Hawai‘i Constitution “alone is enough to show\\nthat the statute is applicable to claims under the takings\\nclause.”\\n\\n The State’s focus on the word “compensation” ignores the\\nremainder of § 657-7’s text. That a plaintiff is seeking “just\\ncompensation” for the taking of private property, Haw. Const.\\n\\f14 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nart. I, § 20, does not necessarily mean that the plaintiff is\\nseeking “compensation for damage or injury to persons or\\nproperty,” Haw. Rev. Stat. § 657-7 (emphasis added). The\\nrelevant question is whether an unconstitutional taking of\\nprivate property without just compensation amounts to\\n“damage or injury to . . . property.”\\n\\n The Hawai‘i Supreme Court has interpreted § 657-7 to\\napply to “claims for damages resulting from physical injury\\nto persons or physical injury to tangible interests in property,”\\nbut not to claims involving a “nonphysical injury to an\\nintangible interest of [the] plaintiff.” Au, 626 P.2d at 178\\n(emphasis added) (quoting Higa v. Mirikitani, 517 P.2d 1, 3\\n(Haw. 1973)). Thus, with respect to real property, the statute\\nhas been applied to actions alleging “trespass quare clausum\\nfregit,”5 Chin Kee v. Kaeleku Sugar Co., 30 Haw. 17, 19–21\\n(1927); Kauha v. Palolo Land & Improvement Co., 20 Haw.\\n237, 239 (1910), actions alleging physical damage to land due\\nto the negligence of a neighboring property owner, Mauian\\nHotel, Inc. v. Maui Pineapple Co., 481 P.2d 310, 312 & n.1\\n(Haw. 1971); Basque v. Yuk Lin Liau, 441 P.2d 636, 637\\n(Haw. 1968), and an action alleging physical destruction of\\ncrops due to the diversion of a water source necessary for\\nirrigation, Wong Nin v. City & Cty. of Honolulu, 33 Haw.\\n379, 380 (1935).\\n\\n DW’s takings claim, by contrast, is predicated on the\\nState’s improper reclassification of DW’s property from\\nurban use to agricultural use. The injury is the State’s\\nregulatory taking of property without just compensation, not\\n\\n 5\\n This is a tort action “to recover damages resulting from another’s\\nunlawful entry on one’s land that is visibly enclosed.” Trespass Quare\\nClausum Fregit, Black’s Law Dictionary (10th ed. 2014).\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 15\\n\\nphysical damage to the property or trespass upon it. This\\ndistinction has led courts in other jurisdictions to conclude\\nthat a “statute of limitations for ‘injuries to persons and\\nproperty’ is not applicable to inverse condemnation actions.”\\nHart v. City of Detroit, 331 N.W.2d 438, 444 (Mich. 1982);\\nsee also Marvel, Annotation, 26 A.L.R.4th 68, § 7[b]\\n(collecting cases in which “statutes of limitation pertaining\\ngenerally to trespass actions or actions for injury to property\\nwere held inapplicable to actions for compensation for the\\ntaking of property without compensation”). In light of the\\nHawai‘i Supreme Court’s insistence on a “physical injury to\\ntangible interests in property,” Au, 626 P.2d at 178 (emphasis\\nadded) (citation omitted), the Hawai‘i Supreme Court could\\nreasonably conclude that § 657-7 does not apply to a\\nregulatory takings claim brought in an inverse condemnation\\naction.\\n\\n * * *\\n\\n The State may be correct in its assertion that either\\n§ 661-5 or § 657-7 provides the applicable limitations period\\nfor DW’s state takings claim under the Hawai‘i Constitution.\\nBut we believe that the best course of action is to certify that\\nquestion to the Hawai‘i Supreme Court.6 Not only is the\\nquestion dispositive in this case and unanswered by Hawai‘i\\nauthority, see Haw. R. App. P. 13(a), but it is also a\\nsignificant question of state law that could “have far-reaching\\nconsequences for the people and businesses of Hawai‘i,”\\nApana v. TIG Ins. Co., 574 F.3d 679, 684 (9th Cir. 2009). In\\nour view, allowing the Hawai‘i Supreme Court to resolve this\\nquestion after analyzing the competing considerations as well\\n\\n 6\\n Although the parties do not request certification, we may “certify a\\nquestion sua sponte.” T-Mobile, 908 F.3d at 587 n.6.\\n\\f16 DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N\\n\\nas the views of other jurisdictions—rather than to trying “to\\npredict uncertain [state] law” in the first instance—will foster\\nthe “cooperative judicial federalism” that the certification\\nprocedure is designed to promote. Lehman Bros. v. Schein,\\n416 U.S. 386, 391 (1974).\\n\\n IV. CERTIFIED QUESTION\\n\\n In light of the foregoing, we respectfully certify the\\nfollowing question to the Hawai‘i Supreme Court:\\n\\n What is the applicable statute of limitations\\n for a claim against the State of Hawai‘i\\n alleging an unlawful taking of “[p]rivate\\n property . . . for public use without just\\n compensation,” Haw. Const. art. I, § 20?\\n\\n We do not intend our phrasing of the question to restrict\\nthe Hawai‘i Supreme Court’s consideration of the relevant\\nissues. If the Hawai‘i Supreme Court accepts review of the\\ncertified question, “it may reformulate the [question] in light\\nof the parties’ contentions or other relevant considerations.”\\nApana, 574 F.3d at 684.\\n\\n V. ORDER\\n\\n The clerk of our court is hereby ordered to transmit a\\ncopy of this order to the Hawai‘i Supreme Court under\\nofficial seal of the United States Court of Appeals for the\\nNinth Circuit. See Haw. R. App. P. 13(c). The clerk is also\\nordered to provide “original or copies of all or any portion of\\nthe record” in this case as “[t]he Hawai‘i Supreme Court may,\\nin its discretion, require.” Id.\\n\\f DW AINA LE‘A DEV. V. STATE OF HAWAI‘I LAND USE COMM’N 17\\n\\n Further proceedings in this court are stayed pending the\\nHawai‘i Supreme Court’s decision. This case is withdrawn\\nfrom submission and the clerk is directed to administratively\\nclose this docket, pending further order from this court. The\\nparties shall file a joint notice in this court within one week\\nof the Hawai‘i Supreme Court’s decision to accept or reject\\ncertification. If the Hawai‘i Supreme Court accepts the\\ncertified question, the parties shall file a joint status report to\\nthis court every six months after the date of acceptance, or\\nmore frequently if circumstances warrant. This panel retains\\njurisdiction over further proceedings in this court.\\n\\n SO ORDERED.\\n\\n FOR THE COURT:\\n\\n\\n Jay S. Bybee, Presiding Judge\\n U.S. Court of Appeals for the\\n Ninth Circuit\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1275,"cells":{"id":{"kind":"number","value":4597700,"string":"4,597,700"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Danilo Mairena v. William Barr"},"decision_date":{"kind":"string","value":"2019-03-07"},"docket_number":{"kind":"string","value":"15-72833"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Ronald M. Gould and Mary H. Murguia, Circuit Judges, and Carol Bagley Amon, ** District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PER CURIAM\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n DANILO ALBERTO MAIRENA, AKA No. 15-72833\\n Danilo Alberto Mairewa, AKA\\n Danilo Alberto Mariena, AKA Agency No.\\n Danilo Marieno, A027-142-897\\n Petitioner,\\n\\n v. OPINION\\n\\n WILLIAM P. BARR, Attorney General,\\n Respondent.\\n\\n\\n\\n On Petition for Review of an Order of the\\n Board of Immigration Appeals\\n\\n Submitted November 14, 2018 *\\n Pasadena, California\\n\\n Filed March 7, 2019\\n\\n\\n\\n\\n *\\n The panel unanimously concludes this case is suitable for decision\\nwithout oral argument. See Fed. R. App. P. 34(a)(2).\\n\\f2 MAIRENA V. BARR\\n\\n Before: Ronald M. Gould and Mary H. Murguia, Circuit\\n Judges, and Carol Bagley Amon, ** District Judge.\\n\\n Per Curiam Opinion\\n\\n\\n SUMMARY ***\\n\\n\\n Immigration\\n\\n Denying Danilo Mairena’s petition for review of a\\ndecision of the Board of Immigration Appeals that upheld an\\nimmigration judge’s denial of withholding of removal,\\nprotection under the Convention Against Torture (“CAT”),\\nand related relief, the panel held that it is appropriate for the\\nBIA to consider sentencing enhancements when it\\ndetermines that a petitioner was convicted of a per se\\nparticularly serious crime.\\n\\n Mairena was convicted of willful infliction of corporal\\ninjury upon the mother of his child with a prior conviction,\\nin violation of California Penal Code § 273.5(e)(1), and was\\nsentenced to five years of imprisonment: four years for the\\noffense, plus a one-year enhancement, pursuant to California\\nPenal Code § 12022.5(b)(1), for using a weapon during the\\ncommission of the offense.\\n\\n\\n\\n **\\n The Honorable Carol Bagley Amon, United States District Judge\\nfor the Eastern District of New York, sitting by designation.\\n ***\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f MAIRENA V. BARR 3\\n\\n In removal proceedings, the BIA concluded that (1)\\nMariena was statutorily ineligible for withholding of\\nremoval because he was sentenced to an aggregate term of\\nfive years of imprisonment for his corporal injury\\nconviction, factoring in the one-year enhancement; and (2)\\nthe IJ did not clearly err in determining that Mairena failed\\nto establish that he would more likely than not be tortured if\\nhe returned to Nicaragua.\\n\\n Under 8 U.S.C. § 1231(b)(3)(B)(ii), withholding of\\nremoval is not available “if the Attorney General decides\\nthat . . . the alien, having been convicted by a final judgment\\nof a particularly serious crime is a danger to the community\\nof the United States . . . .” The provision further explains:\\n“For purposes of clause (ii), an alien who has been convicted\\nof an aggravated felony (or felonies) for which the alien has\\nbeen sentenced to an aggregate term of imprisonment of at\\nleast 5 years shall be considered to have committed a\\nparticularly serious crime.”\\n\\n Mairena did not dispute that his conviction constituted\\nan aggravated felony, but argued that the BIA erred by\\nconsidering the one-year sentencing enhancement in\\ndeciding that he was sentenced to an aggregate term of\\nimprisonment of five years—and consequently convicted of\\na per se particularly serious crime—because the\\nenhancement was not an element of the offense and because\\nthe statutory maximum for his offense was four years.\\n\\n The panel noted that aggravated felonies with resulting\\nsentences of at least five years are per se particularly serious\\nand, by contrast, aggravated felonies resulting in sentences\\nfewer than five years are not per se particularly serious and\\nrequire a case-by-case analysis. The panel also observed that\\nthis court has already held that it is appropriate for the BIA\\n\\f4 MAIRENA V. BARR\\n\\nto consider sentencing enhancements when it determines\\nthat a petitioner was convicted of a particularly serious crime\\non a case-by-case basis.\\n\\n The panel now clarified that it is also appropriate for the\\nBIA to consider sentencing enhancements when it\\ndetermines that a petitioner was convicted of a per se\\nparticularly serious crime. The panel explained that\\nMairena’s contention that the BIA could not consider his\\nsentencing enhancement was foreclosed by the plain\\nlanguage of § 1231(b)(3)(B), in that the five-year\\nrequirement is keyed to the “aggregate term of\\nimprisonment” for the actual sentence imposed—not to the\\nstatutory maximum. The panel also observed that nothing in\\nthe text of the statute prohibits the BIA from considering\\nsentencing enhancements in computing the aggregate term\\nof imprisonment.\\n\\n Moreover, the panel noted that, even if it accepted\\nMairena’s contention that it should look to the statutory\\nmaximum in interpreting § 1231(b)(3)(B), the statutory\\nmaximum for Mairena’s offense of conviction was in fact\\nfive years, not four years.\\n\\n The panel also concluded that substantial evidence\\nsupported the BIA’s conclusion that Mairena was ineligible\\nfor CAT relief. The panel concluded that the IJ was not\\nrequired to conduct a separate credibility analysis in\\nadjudicating the CAT claim, explaining that the IJ was\\nentitled to rely on the adverse credibility determination in\\ndenying CAT relief, provided that the IJ considered other\\nevidence in the record on country conditions in Nicaragua.\\nThe panel concluded that the IJ did so in this case. The panel\\nalso concluded that the record did not compel the conclusion\\n\\f MAIRENA V. BARR 5\\n\\nthat Mairena would more likely than not be tortured upon\\nreturn to Nicaragua.\\n\\n\\n COUNSEL\\n\\nTania Pham, Law Offices of Tania T. Pham, Woodland\\nHills, California, for Petitioner.\\n\\nBenjamin J. Zeitlin, Attorney; Carl McIntyre, Assistant\\nDirector; Office of Immigration Litigation, Civil Division,\\nUnited States Department of Justice, Washington, D.C.; for\\nRespondent.\\n\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n Petitioner Danilo Mairena, a native and citizen of\\nNicaragua, petitions for review of a final order of the Board\\nof Immigration Appeals (“BIA”) dismissing his appeal of the\\nImmigration Judge’s (“IJ”) denial of his applications for\\nwithholding of removal, protection under the Convention\\nAgainst Torture (“CAT”), and related relief. We have\\njurisdiction under 8 U.S.C. § 1252. We deny his petition.\\n\\n BACKGROUND\\n\\n Mairena, a native and citizen of Nicaragua born in 1979,\\nentered the United States on a visitor visa in 1984 and\\nattained lawful permanent resident status in 1988. Mairena’s\\nwife, two daughters, and parents reside in the United States,\\nand he has no family left in Nicaragua.\\n\\f6 MAIRENA V. BARR\\n\\n On August 9, 2010, Mairena was convicted of willful\\ninfliction of corporal injury upon the mother of his child with\\na prior conviction, in violation of California Penal Code\\n§ 273.5(e)(1). 1 He had been previously convicted of\\ncorporal injury to a spouse in 2008. Mairena was sentenced\\nto five years of imprisonment: four years for the offense,\\nplus a one-year enhancement, pursuant to California Penal\\nCode § 12022.5(b)(1), 2 for using a weapon during the\\ncommission of the offense. That same day, Mairena was\\nalso convicted of dissuading a witness, in violation of\\nCalifornia Penal Code § 136.1(c)(1), for which he was\\nsentenced to three years of imprisonment.\\n\\n On October 10, 2013, the Department of Homeland\\nSecurity served Mairena with a Notice to Appear and\\ncharged him as removable based on those two felony\\nconvictions under § 237(a)(2)(A)(iii) of the Immigration and\\nNationality Act. See 8 U.S.C. § 1227(a)(2)(A)(iii). On\\n\\n\\n 1\\n California Penal Code § 273.5(e)(1) has been amended since\\nMairena’s conviction. Mairena was convicted under what is now\\nCalifornia Penal Code § 273.5(f)(1). Compare Cal. Penal Code\\n§ 273.5(e)(1) (2010), with id. § 273.5(f)(1) (2018).\\n\\n 2\\n Although the judgment of conviction lists California Penal Code\\n§ 12022.5(b)(1) as the source of the one-year enhancement, it\\npresumably should have listed California Penal Code § 12022(b)(1).\\nThe former authorizes an additional term of five, six, or ten years for the\\nuse of an assault weapon; the latter authorizes an additional term of one\\nyear for the use of a deadly or dangerous weapon in the commission of\\na felony. Compare Cal. Penal Code § 12022.5(b)(1) (2010), with id.\\n§ 12022(b)(1) (2010). The First Amended Information charges Mairena\\nunder the latter provision for using a pipe during the offense. In any\\nevent, the particular enhancement provision is irrelevant to our\\nresolution of the issues presented in this case.\\n\\f MAIRENA V. BARR 7\\n\\nMarch 20, 2014, the IJ (Walsh, I.J.) concluded that Mairena\\nwas removable as charged.\\n\\n On May 15, 2014, Mariena applied for asylum,\\nwithholding of removal, and CAT protection. In his\\napplication, Mairena alleged that he feared that the\\nSandinistas would kill him because his family was\\npreviously persecuted by the Sandinistas and was currently\\nfighting with President Daniel Ortega to recover seized\\nfamily property. 3 On October 15, 2014, Mairena applied for\\nadjustment of status and a waiver of inadmissibility.\\n\\n On April 1, 2015, the IJ denied all relief. As relevant to\\nthis petition, the IJ concluded as follows: (1) Mairena was\\nstatutorily ineligible for withholding of removal because he\\nwas sentenced to an aggregate term of eight years of\\nimprisonment for his two felony convictions, and therefore\\nconvicted of a per se particularly serious crime; and\\n(2) Mariena failed to carry his burden of proving that he\\nwould more likely than not be tortured if he returned to\\nNicaragua, and thus CAT protection was not warranted.\\n\\n On August 27, 2015, the BIA affirmed the IJ’s decision\\nand dismissed Mairena’s appeal. As relevant to this petition,\\nthe BIA concluded as follows: (1) Mariena was statutorily\\nineligible for withholding of removal because he was\\nsentenced to an aggregate term of five years of imprisonment\\nfor his corporal injury conviction, factoring in the one-year\\nenhancement; and (2) the IJ did not clearly err in determining\\nthat Mairena failed to establish that he would more likely\\n\\n 3\\n According to the United States Department of State Nicaragua\\n2013 Human Rights Report, contained in the record below, the\\nSandinista National Liberation Front is a political party that has\\nincreasingly concentrated political power in Nicaragua.\\n\\f8 MAIRENA V. BARR\\n\\nthan not be tortured if he returned to Nicaragua. The BIA\\ndid not rely on Mairena’s conviction for dissuading a\\nwitness, for which he was sentenced to three years of\\nimprisonment.\\n\\n On September 14, 2015, Mairena timely petitioned this\\nCourt for review.\\n\\n JURISDICTION AND STANDARD OF REVIEW\\n\\n We lack jurisdiction to review “any final order of\\nremoval against an alien who is removable” because he\\ncommitted an aggravated felony, see 8 U.S.C.\\n§§ 1252(a)(2)(C), 1227(a)(2)(A)(iii), but “we retain\\njurisdiction to decide our own jurisdiction and to resolve\\nquestions of law,” Bolanos v. Holder, 734 F.3d 875, 876 (9th\\nCir. 2013). “Although we cannot reweigh evidence to\\ndetermine if the crime was indeed particularly serious, [we]\\ncan determine whether the BIA applied the correct legal\\nstandard.” Konou v. Holder, 750 F.3d 1120, 1127 (9th Cir.\\n2014) (alteration in original) (quoting Blandino-Medina v.\\nHolder, 712 F.3d 1338, 1343 (9th Cir. 2013)).\\n\\n We also have jurisdiction to review the BIA’s denial of\\nCAT protection where, as here, “the IJ did not rely on\\n[petitioner’s] conviction . . . but instead denied relief on the\\nmerits.” Alphonsus v. Holder, 705 F.3d 1031, 1036–37 (9th\\nCir. 2013), abrogated on other grounds as recognized in\\nGuerrero v. Whitaker, 908 F.3d 541 (9th Cir. 2018).\\n\\n Where “the BIA conducts its own review of the evidence\\nand law, rather than adopting the IJ’s decision, our review is\\nlimited to the BIA’s decision, except to the extent the IJ’s\\nopinion is expressly adopted.” Zumel v. Lynch, 803 F.3d\\n463, 471 (9th Cir. 2015) (quoting Rodriguez v. Holder,\\n683 F.3d 1164, 1169 (9th Cir. 2012)). “[W]e treat the\\n\\f MAIRENA V. BARR 9\\n\\nincorporated parts of the IJ’s decision as the BIA’s.” Rivera\\nv. Mukasey, 508 F.3d 1271, 1275 (9th Cir. 2007).\\n\\n We review legal questions de novo and factual findings,\\nincluding adverse credibility determinations, for substantial\\nevidence. Vilchez v. Holder, 682 F.3d 1195, 1198–99 (9th\\nCir. 2012); Shrestha v. Holder, 590 F.3d 1034, 1039 (9th Cir.\\n2010). Under the substantial evidence standard,\\n“administrative findings of fact are conclusive unless any\\nreasonable adjudicator would be compelled to conclude to\\nthe contrary . . . .” 8 U.S.C. § 1252(b)(4)(B); see also\\nArteaga v. Mukasey, 511 F.3d 940, 944 (9th Cir. 2007)\\n(“Under the substantial evidence standard, the court upholds\\nthe BIA’s determination unless the evidence in the record\\ncompels a contrary conclusion.”).\\n\\n DISCUSSION\\n\\nI. The Particularly Serious Crime Determination\\n\\n In general, an alien is entitled to withholding of removal\\nif “the alien’s life or freedom would be threatened in that\\ncountry because of the alien’s race, religion, nationality,\\nmembership in a particular social group, or political\\nopinion.” 8 U.S.C. § 1231(b)(3)(A). But withholding of\\nremoval is not available “if the Attorney General decides\\nthat . . . the alien, having been convicted by a final judgment\\nof a particularly serious crime is a danger to the community\\nof the United States . . . .” Id. § 1231(b)(3)(B)(ii). The\\nprovision further explains:\\n\\n For purposes of clause (ii), an alien who has\\n been convicted of an aggravated felony (or\\n felonies) for which the alien has been\\n sentenced to an aggregate term of\\n imprisonment of at least 5 years shall be\\n\\f10 MAIRENA V. BARR\\n\\n considered to have committed a particularly\\n serious crime. The previous sentence shall\\n not preclude the Attorney General from\\n determining that, notwithstanding the length\\n of sentence imposed, an alien has been\\n convicted of a particularly serious crime.\\n\\nId. § 1231(b)(3)(B).\\n\\n Under the statute, the appropriate analytical lens depends\\non the length of the sentence imposed. “[A]ggravated\\nfelonies with resulting sentences of at least five years are per\\nse particularly serious . . . .” Guerrero, 908 F.3d at 545. By\\ncontrast, “aggravated felonies resulting in sentences fewer\\nthan five years are not per se particularly serious and still\\nrequire a case-by-case analysis . . . .” Blandino-Medina,\\n712 F.3d at 1347 (quoting Afridi v. Gonzales, 442 F.3d 1212,\\n1220 n.4 (9th Cir. 2006)). 4\\n\\n Mairena does not dispute that his conviction for corporal\\ninjury constitutes an aggravated felony. 5 He argues,\\n\\n\\n 4\\n Under the case-by-case analysis, this Court considers the\\nconviction in light of the Frentescu factors: “the nature of the conviction,\\nthe circumstances and underlying facts of the conviction, the type of\\nsentence imposed, and, most importantly, whether the type and\\ncircumstances of the crime indicate that the alien will be a danger to the\\ncommunity.” Konou, 750 F.3d at 1127 (quoting Matter of Frentescu,\\n18 I & N. Dec. 244, 247 (B.I.A. 1982)). We review for abuse of\\ndiscretion the BIA’s case-by-case determination that an individual was\\nconvicted of a particularly serious crime. Id.\\n\\n 5\\n “As used in immigration law, ‘aggravated felony’ is a term of art\\nreferring to the offenses enumerated under [8 U.S.C.] § 1101(a)(43).”\\nDelgado v. Holder, 648 F.3d 1095, 1101 (9th Cir. 2011) (en banc).\\nAmong the enumerated offenses is “a crime of violence . . . for which\\n\\f MAIRENA V. BARR 11\\n\\nhowever, that the BIA erred by considering the one-year\\nsentencing enhancement in deciding that he was sentenced\\nto an aggregate term of imprisonment of five years—and\\nconsequently convicted of a per se particularly serious\\ncrime—because the enhancement was not an element of the\\noffense and because the statutory maximum for his offense\\nis four years. Instead, he contends, the BIA should have\\nengaged in a case-by-case analysis.\\n\\n We have already held that it is appropriate for the BIA to\\nconsider sentencing enhancements when it determines that a\\npetitioner is convicted of a particularly serious crime on a\\ncase-by-case basis. Konou, 750 F.3d at 1128. As this Court\\nexplained, the case-by-case analysis calls for analyzing the\\n“type of sentence imposed,” and “[a]n enhanced sentence by\\nits plain language can be considered a type of sentence.”\\nKonou, 750 F.3d at 1128. We thus rejected the argument\\nthat “the BIA cannot consider a sentencing enhancement\\nwhen it determines whether a crime is particularly serious.”\\nId.\\n\\n We now clarify that it is also appropriate for the BIA to\\nconsider sentencing enhancements when it determines that a\\npetitioner was convicted of a per se particularly serious\\ncrime. See Garcia v. Lynch, 652 F. App’x 591, 593 (9th Cir.\\n2016) (applying Konou to a per se particularly serious crime\\ndetermination). Mairena’s contention that the BIA could not\\nconsider his sentencing enhancement is foreclosed by the\\nplain language of the statute. Section 1231(b)(3)(B) asks\\nwhether the individual has been convicted of an aggravated\\nfelony for which he “has been sentenced to an aggregate\\nterm of imprisonment of at least 5 years.” 8 U.S.C.\\n\\nthe term of imprisonment [is] at least one year.” 8 U.S.C.\\n§ 1101(a)(43)(F).\\n\\f12 MAIRENA V. BARR\\n\\n§ 1231(b)(3)(B). Thus, the five-year requirement is keyed\\nto the “aggregate term of imprisonment” for the actual\\nsentence imposed—not to the statutory maximum. Cf.\\nUnited States v. Corona-Sanchez, 291 F.3d 1201, 1208–09\\n(9th Cir. 2002) (en banc) (in determining whether a\\nconviction categorically qualifies as an aggravated felony\\nbecause it is an offense “for which the term of imprisonment\\n[is] at least one year,” 8 U.S.C. § 1101(a)(43)(G), a court\\n“must consider the sentence available for the crime itself,\\nwithout considering separate recidivist sentencing\\nenhancements”), superseded by statute on other grounds as\\nexplained in United States v. Gomez-Mendez, 486 F.3d 599,\\n604–05 (9th Cir. 2007); Rusz v. Ashcroft, 376 F.3d 1182,\\n1184–85 (9th Cir. 2004) (applying Corona-Sanchez to a\\nprovision of the Immigration and Nationality Act that\\nprovides for the removal of an alien convicted of, inter alia,\\n“a crime for which a sentence of one year or longer may be\\nimposed” (quoting 8 U.S.C. § 1227(a)(2)(A)(i)(II))).\\nNothing in the text of § 1231(b)(3)(B) prohibits the BIA\\nfrom considering sentencing enhancements in computing the\\naggregate term of imprisonment.\\n\\n Moreover, even if we accepted Mairena’s contention that\\nwe should look to the statutory maximum in interpreting\\n§ 1231(b)(3)(B), the statutory maximum for Mairena’s\\noffense of conviction was in fact five years, not four years.\\nAlthough the statutory maximum for a corporal injury\\nconviction under California Penal Code § 273.5(a) was\\nindeed four years, Mairena was convicted under California\\nPenal Code § 273.5(e)(1), which provided for an enhanced\\nstatutory maximum of five years for an individual with a\\nprior corporal injury conviction within the last seven years.\\nCompare Cal. Penal Code § 273.5(a) (2010), with id.\\n§ 273.5(e)(1) (2010). When he was convicted in 2010,\\nMairena had such a prior conviction from 2008.\\n\\f MAIRENA V. BARR 13\\n\\n Accordingly, the BIA applied the correct legal standard\\nwhen it determined that Mairena was convicted of a per se\\nparticularly serious crime and was therefore ineligible for\\nwithholding of removal.\\n\\nII. The Denial of CAT Protection\\n\\n An individual who is ineligible for withholding of\\nremoval is nevertheless eligible for “deferral of removal to\\nthe country where he or she is more likely than not to be\\ntortured.” 8 C.F.R. § 1208.17(a). “Torture is an extreme\\nform of cruel and inhuman treatment that either (1) is not\\nlawfully sanctioned by that country or (2) is lawfully\\nsanctioned by that country, but defeats the object and\\npurpose of CAT.” Konou, 750 F.3d at 1124 (quoting\\nGarcia-Milian v. Holder, 755 F.3d 1026, 1033 (9th Cir.\\n2014)); see also 8 C.F.R. § 1208.18(a)(1) (defining torture).\\nIn addition, the torture must be “inflicted by or at the\\ninstigation of or with the consent or acquiescence of a public\\nofficial or other person acting in an official capacity.”\\n8 C.F.R. § 1208.18(a)(1); see also Garcia-Milian, 755 F.3d\\nat 1033.\\n\\n The applicant bears the burden of proving that he is\\neligible for deferral of removal under CAT. 8 C.F.R.\\n§ 1208.17(d)(3). “Evidence of past torture inflicted upon the\\napplicant” is relevant in assessing whether torture is more\\nlikely than not. 8 C.F.R. § 1208.16(c)(3); see also\\nKamalthas v. I.N.S., 251 F.3d 1279, 1282 (9th Cir. 2001). A\\nUnited States Department of State report “can carry an\\napplicant’s burden of establishing a probability of torture,”\\nbut it “can also serve to outweigh an applicant’s evidence of\\na probability of torture.” Konou, 750 F.3d at 1125. In\\naddition, the “testimony of the applicant, if credible, may be\\nsufficient to sustain the burden of proof without\\n\\f14 MAIRENA V. BARR\\n\\ncorroboration.” 8 C.F.R. § 1208.16(c)(2); see also Konou,\\n750 F.3d at 1124.\\n\\n Substantial evidence supports the BIA’s conclusion that\\nMairena was ineligible for CAT relief. First, the IJ was not\\nrequired to conduct a separate credibility analysis in\\nadjudicating Mairena’s CAT claim. Contrary to Mairena’s\\ncontention, the IJ was entitled to rely on the adverse\\ncredibility determination in denying CAT relief, see Singh v.\\nLynch, 802 F.3d 972, 977 (9th Cir. 2015), provided that the\\nIJ considered other evidence in the record on country\\nconditions in Nicaragua, Kamalthas, 251 F.3d at 1282–84.\\nThe IJ did so in this case. Second, the record does not\\ncompel the conclusion that Mairena would more likely than\\nnot be tortured upon return to Nicaragua. Although both\\nMairena and his family expressed fear that he would be\\ntortured because of his family’s pro-Contra association,\\n(1) the persecution for which his family received political\\nasylum transpired thirty years ago, (2) Mairena himself was\\nnever tortured, and (3) there was no evidence that Mairena\\nor anyone in his family had received threats while attempting\\nto regain family land from the government. Mairena has not\\ncontested the accuracy of these findings on appeal; instead,\\nhe essentially seems to disagree with the agency’s\\nconclusion. But “our task ‘is to determine whether there is\\nsubstantial evidence to support the BIA’s finding, not to\\nsubstitute an analysis of which side in the factual dispute we\\nfind more persuasive.’” Singh, 802 F.3d at 974–75 (quoting\\nMolina-Morales v. I.N.S., 237 F.3d 1048, 1050 (9th Cir.\\n2001)). The BIA also concluded that, while the Department\\nof State report indicates that some former Contras were\\nlikely killed by the Nicaraguan government, those\\nindividuals—unlike Mairena—were well known and\\ninvolved in armed confrontations. The “BIA’s interpretation\\nof the Report is entitled to deference.” Konou, 750 F.3d at\\n\\f MAIRENA V. BARR 15\\n\\n1125; see also Dhital v. Mukasey, 532 F.3d 1044, 1051–52\\n(9th Cir. 2008) (per curiam) (denying CAT relief where the\\nDepartment of State reports indicate that torture has\\noccurred, but “do not indicate that [petitioner] would face\\nany particular threat of torture beyond that of which all\\ncitizens of Nepal are at risk”). Finally, Mairena has not\\nsuggested on appeal that either the IJ or the BIA failed to\\nconsider relevant evidence in the record.\\n\\n Accordingly, substantial evidence supports the BIA’s\\ndetermination that Mairena failed to establish that he would\\nmore likely than not be tortured in Nicaragua.\\n\\n Petition DENIED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4374953/\", \"author_raw\": \"PER CURIAM\"}]}"},"judge1":{"kind":"string","value":"RONALD M GOULD"},"judge2":{"kind":"string","value":"MARY H MURGUIA"},"judge3":{"kind":"string","value":"CAROL BAGLEY AMON"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4597700/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Danilo Mairena v. William Barr\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-07\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n DANILO ALBERTO MAIRENA, AKA No. 15-72833\n Danilo Alberto Mairewa, AKA\n Danilo Alberto Mariena, AKA Agency No.\n Danilo Marieno, A027-142-897\n Petitioner,\n\n v. OPINION\n\n WILLIAM P. BARR, Attorney General,\n Respondent.\n\n\n\n On Petition for Review of an Order of the\n Board of Immigration Appeals\n\n Submitted November 14, 2018 *\n Pasadena, California\n\n Filed March 7, 2019\n\n\n\n\n *\n The panel unanimously concludes this case is suitable for decision\nwithout oral argument. See Fed. R. App. P. 34(a)(2).\n\f2 MAIRENA V. BARR\n\n Before: Ronald M. Gould and Mary H. Murguia, Circuit\n Judges, and Carol Bagley Amon, ** District Judge.\n\n Per Curiam Opinion\n\n\n SUMMARY ***\n\n\n Immigration\n\n Denying Danilo Mairena’s petition for review of a\ndecision of the Board of Immigration Appeals that upheld an\nimmigration judge’s denial of withholding of removal,\nprotection under the Convention Against Torture (“CAT”),\nand related relief, the panel held that it is appropriate for the\nBIA to consider sentencing enhancements when it\ndetermines that a petitioner was convicted of a per se\nparticularly serious crime.\n\n Mairena was convicted of willful infliction of corporal\ninjury upon the mother of his child with a prior conviction,\nin violation of California Penal Code § 273.5(e)(1), and was\nsentenced to five years of imprisonment: four years for the\noffense, plus a one-year enhancement, pursuant to California\nPenal Code § 12022.5(b)(1), for using a weapon during the\ncommission of the offense.\n\n\n\n **\n The Honorable Carol Bagley Amon, United States District Judge\nfor the Eastern District of New York, sitting by designation.\n ***\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f MAIRENA V. BARR 3\n\n In removal proceedings, the BIA concluded that (1)\nMariena was statutorily ineligible for withholding of\nremoval because he was sentenced to an aggregate term of\nfive years of imprisonment for his corporal injury\nconviction, factoring in the one-year enhancement; and (2)\nthe IJ did not clearly err in determining that Mairena failed\nto establish that he would more likely than not be tortured if\nhe returned to Nicaragua.\n\n Under 8 U.S.C. § 1231(b)(3)(B)(ii), withholding of\nremoval is not available “if the Attorney General decides\nthat . . . the alien, having been convicted by a final judgment\nof a particularly serious crime is a danger to the community\nof the United States . . . .” The provision further explains:\n“For purposes of clause (ii), an alien who has been convicted\nof an aggravated felony (or felonies) for which the alien has\nbeen sentenced to an aggregate term of imprisonment of at\nleast 5 years shall be considered to have committed a\nparticularly serious crime.”\n\n Mairena did not dispute that his conviction constituted\nan aggravated felony, but argued that the BIA erred by\nconsidering the one-year sentencing enhancement in\ndeciding that he was sentenced to an aggregate term of\nimprisonment of five years—and consequently convicted of\na per se particularly serious crime—because the\nenhancement was not an element of the offense and because\nthe statutory maximum for his offense was four years.\n\n The panel noted that aggravated felonies with resulting\nsentences of at least five years are per se particularly serious\nand, by contrast, aggravated felonies resulting in sentences\nfewer than five years are not per se particularly serious and\nrequire a case-by-case analysis. The panel also observed that\nthis court has already held that it is appropriate for the BIA\n\f4 MAIRENA V. BARR\n\nto consider sentencing enhancements when it determines\nthat a petitioner was convicted of a particularly serious crime\non a case-by-case basis.\n\n The panel now clarified that it is also appropriate for the\nBIA to consider sentencing enhancements when it\ndetermines that a petitioner was convicted of a per se\nparticularly serious crime. The panel explained that\nMairena’s contention that the BIA could not consider his\nsentencing enhancement was foreclosed by the plain\nlanguage of § 1231(b)(3)(B), in that the five-year\nrequirement is keyed to the “aggregate term of\nimprisonment” for the actual sentence imposed—not to the\nstatutory maximum. The panel also observed that nothing in\nthe text of the statute prohibits the BIA from considering\nsentencing enhancements in computing the aggregate term\nof imprisonment.\n\n Moreover, the panel noted that, even if it accepted\nMairena’s contention that it should look to the statutory\nmaximum in interpreting § 1231(b)(3)(B), the statutory\nmaximum for Mairena’s offense of conviction was in fact\nfive years, not four years.\n\n The panel also concluded that substantial evidence\nsupported the BIA’s conclusion that Mairena was ineligible\nfor CAT relief. The panel concluded that the IJ was not\nrequired to conduct a separate credibility analysis in\nadjudicating the CAT claim, explaining that the IJ was\nentitled to rely on the adverse credibility determination in\ndenying CAT relief, provided that the IJ considered other\nevidence in the record on country conditions in Nicaragua.\nThe panel concluded that the IJ did so in this case. The panel\nalso concluded that the record did not compel the conclusion\n\f MAIRENA V. BARR 5\n\nthat Mairena would more likely than not be tortured upon\nreturn to Nicaragua.\n\n\n COUNSEL\n\nTania Pham, Law Offices of Tania T. Pham, Woodland\nHills, California, for Petitioner.\n\nBenjamin J. Zeitlin, Attorney; Carl McIntyre, Assistant\nDirector; Office of Immigration Litigation, Civil Division,\nUnited States Department of Justice, Washington, D.C.; for\nRespondent.\n\n\n OPINION\n\nPER CURIAM:\n\n Petitioner Danilo Mairena, a native and citizen of\nNicaragua, petitions for review of a final order of the Board\nof Immigration Appeals (“BIA”) dismissing his appeal of the\nImmigration Judge’s (“IJ”) denial of his applications for\nwithholding of removal, protection under the Convention\nAgainst Torture (“CAT”), and related relief. We have\njurisdiction under 8 U.S.C. § 1252. We deny his petition.\n\n BACKGROUND\n\n Mairena, a native and citizen of Nicaragua born in 1979,\nentered the United States on a visitor visa in 1984 and\nattained lawful permanent resident status in 1988. Mairena’s\nwife, two daughters, and parents reside in the United States,\nand he has no family left in Nicaragua.\n\f6 MAIRENA V. BARR\n\n On August 9, 2010, Mairena was convicted of willful\ninfliction of corporal injury upon the mother of his child with\na prior conviction, in violation of California Penal Code\n§ 273.5(e)(1). 1 He had been previously convicted of\ncorporal injury to a spouse in 2008. Mairena was sentenced\nto five years of imprisonment: four years for the offense,\nplus a one-year enhancement, pursuant to California Penal\nCode § 12022.5(b)(1), 2 for using a weapon during the\ncommission of the offense. That same day, Mairena was\nalso convicted of dissuading a witness, in violation of\nCalifornia Penal Code § 136.1(c)(1), for which he was\nsentenced to three years of imprisonment.\n\n On October 10, 2013, the Department of Homeland\nSecurity served Mairena with a Notice to Appear and\ncharged him as removable based on those two felony\nconvictions under § 237(a)(2)(A)(iii) of the Immigration and\nNationality Act. See 8 U.S.C. § 1227(a)(2)(A)(iii). On\n\n\n 1\n California Penal Code § 273.5(e)(1) has been amended since\nMairena’s conviction. Mairena was convicted under what is now\nCalifornia Penal Code § 273.5(f)(1). Compare Cal. Penal Code\n§ 273.5(e)(1) (2010), with id. § 273.5(f)(1) (2018).\n\n 2\n Although the judgment of conviction lists California Penal Code\n§ 12022.5(b)(1) as the source of the one-year enhancement, it\npresumably should have listed California Penal Code § 12022(b)(1).\nThe former authorizes an additional term of five, six, or ten years for the\nuse of an assault weapon; the latter authorizes an additional term of one\nyear for the use of a deadly or dangerous weapon in the commission of\na felony. Compare Cal. Penal Code § 12022.5(b)(1) (2010), with id.\n§ 12022(b)(1) (2010). The First Amended Information charges Mairena\nunder the latter provision for using a pipe during the offense. In any\nevent, the particular enhancement provision is irrelevant to our\nresolution of the issues presented in this case.\n\f MAIRENA V. BARR 7\n\nMarch 20, 2014, the IJ (Walsh, I.J.) concluded that Mairena\nwas removable as charged.\n\n On May 15, 2014, Mariena applied for asylum,\nwithholding of removal, and CAT protection. In his\napplication, Mairena alleged that he feared that the\nSandinistas would kill him because his family was\npreviously persecuted by the Sandinistas and was currently\nfighting with President Daniel Ortega to recover seized\nfamily property. 3 On October 15, 2014, Mairena applied for\nadjustment of status and a waiver of inadmissibility.\n\n On April 1, 2015, the IJ denied all relief. As relevant to\nthis petition, the IJ concluded as follows: (1) Mairena was\nstatutorily ineligible for withholding of removal because he\nwas sentenced to an aggregate term of eight years of\nimprisonment for his two felony convictions, and therefore\nconvicted of a per se particularly serious crime; and\n(2) Mariena failed to carry his burden of proving that he\nwould more likely than not be tortured if he returned to\nNicaragua, and thus CAT protection was not warranted.\n\n On August 27, 2015, the BIA affirmed the IJ’s decision\nand dismissed Mairena’s appeal. As relevant to this petition,\nthe BIA concluded as follows: (1) Mariena was statutorily\nineligible for withholding of removal because he was\nsentenced to an aggregate term of five years of imprisonment\nfor his corporal injury conviction, factoring in the one-year\nenhancement; and (2) the IJ did not clearly err in determining\nthat Mairena failed to establish that he would more likely\n\n 3\n According to the United States Department of State Nicaragua\n2013 Human Rights Report, contained in the record below, the\nSandinista National Liberation Front is a political party that has\nincreasingly concentrated political power in Nicaragua.\n\f8 MAIRENA V. BARR\n\nthan not be tortured if he returned to Nicaragua. The BIA\ndid not rely on Mairena’s conviction for dissuading a\nwitness, for which he was sentenced to three years of\nimprisonment.\n\n On September 14, 2015, Mairena timely petitioned this\nCourt for review.\n\n JURISDICTION AND STANDARD OF REVIEW\n\n We lack jurisdiction to review “any final order of\nremoval against an alien who is removable” because he\ncommitted an aggravated felony, see 8 U.S.C.\n§§ 1252(a)(2)(C), 1227(a)(2)(A)(iii), but “we retain\njurisdiction to decide our own jurisdiction and to resolve\nquestions of law,” Bolanos v. Holder, 734 F.3d 875, 876 (9th\nCir. 2013). “Although we cannot reweigh evidence to\ndetermine if the crime was indeed particularly serious, [we]\ncan determine whether the BIA applied the correct legal\nstandard.” Konou v. Holder, 750 F.3d 1120, 1127 (9th Cir.\n2014) (alteration in original) (quoting Blandino-Medina v.\nHolder, 712 F.3d 1338, 1343 (9th Cir. 2013)).\n\n We also have jurisdiction to review the BIA’s denial of\nCAT protection where, as here, “the IJ did not rely on\n[petitioner’s] conviction . . . but instead denied relief on the\nmerits.” Alphonsus v. Holder, 705 F.3d 1031, 1036–37 (9th\nCir. 2013), abrogated on other grounds as recognized in\nGuerrero v. Whitaker, 908 F.3d 541 (9th Cir. 2018).\n\n Where “the BIA conducts its own review of the evidence\nand law, rather than adopting the IJ’s decision, our review is\nlimited to the BIA’s decision, except to the extent the IJ’s\nopinion is expressly adopted.” Zumel v. Lynch, 803 F.3d\n463, 471 (9th Cir. 2015) (quoting Rodriguez v. Holder,\n683 F.3d 1164, 1169 (9th Cir. 2012)). “[W]e treat the\n\f MAIRENA V. BARR 9\n\nincorporated parts of the IJ’s decision as the BIA’s.” Rivera\nv. Mukasey, 508 F.3d 1271, 1275 (9th Cir. 2007).\n\n We review legal questions de novo and factual findings,\nincluding adverse credibility determinations, for substantial\nevidence. Vilchez v. Holder, 682 F.3d 1195, 1198–99 (9th\nCir. 2012); Shrestha v. Holder, 590 F.3d 1034, 1039 (9th Cir.\n2010). Under the substantial evidence standard,\n“administrative findings of fact are conclusive unless any\nreasonable adjudicator would be compelled to conclude to\nthe contrary . . . .” 8 U.S.C. § 1252(b)(4)(B); see also\nArteaga v. Mukasey, 511 F.3d 940, 944 (9th Cir. 2007)\n(“Under the substantial evidence standard, the court upholds\nthe BIA’s determination unless the evidence in the record\ncompels a contrary conclusion.”).\n\n DISCUSSION\n\nI. The Particularly Serious Crime Determination\n\n In general, an alien is entitled to withholding of removal\nif “the alien’s life or freedom would be threatened in that\ncountry because of the alien’s race, religion, nationality,\nmembership in a particular social group, or political\nopinion.” 8 U.S.C. § 1231(b)(3)(A). But withholding of\nremoval is not available “if the Attorney General decides\nthat . . . the alien, having been convicted by a final judgment\nof a particularly serious crime is a danger to the community\nof the United States . . . .” Id. § 1231(b)(3)(B)(ii). The\nprovision further explains:\n\n For purposes of clause (ii), an alien who has\n been convicted of an aggravated felony (or\n felonies) for which the alien has been\n sentenced to an aggregate term of\n imprisonment of at least 5 years shall be\n\f10 MAIRENA V. BARR\n\n considered to have committed a particularly\n serious crime. The previous sentence shall\n not preclude the Attorney General from\n determining that, notwithstanding the length\n of sentence imposed, an alien has been\n convicted of a particularly serious crime.\n\nId. § 1231(b)(3)(B).\n\n Under the statute, the appropriate analytical lens depends\non the length of the sentence imposed. “[A]ggravated\nfelonies with resulting sentences of at least five years are per\nse particularly serious . . . .” Guerrero, 908 F.3d at 545. By\ncontrast, “aggravated felonies resulting in sentences fewer\nthan five years are not per se particularly serious and still\nrequire a case-by-case analysis . . . .” Blandino-Medina,\n712 F.3d at 1347 (quoting Afridi v. Gonzales, 442 F.3d 1212,\n1220 n.4 (9th Cir. 2006)). 4\n\n Mairena does not dispute that his conviction for corporal\ninjury constitutes an aggravated felony. 5 He argues,\n\n\n 4\n Under the case-by-case analysis, this Court considers the\nconviction in light of the Frentescu factors: “the nature of the conviction,\nthe circumstances and underlying facts of the conviction, the type of\nsentence imposed, and, most importantly, whether the type and\ncircumstances of the crime indicate that the alien will be a danger to the\ncommunity.” Konou, 750 F.3d at 1127 (quoting Matter of Frentescu,\n18 I & N. Dec. 244, 247 (B.I.A. 1982)). We review for abuse of\ndiscretion the BIA’s case-by-case determination that an individual was\nconvicted of a particularly serious crime. Id.\n\n 5\n “As used in immigration law, ‘aggravated felony’ is a term of art\nreferring to the offenses enumerated under [8 U.S.C.] § 1101(a)(43).”\nDelgado v. Holder, 648 F.3d 1095, 1101 (9th Cir. 2011) (en banc).\nAmong the enumerated offenses is “a crime of violence . . . for which\n\f MAIRENA V. BARR 11\n\nhowever, that the BIA erred by considering the one-year\nsentencing enhancement in deciding that he was sentenced\nto an aggregate term of imprisonment of five years—and\nconsequently convicted of a per se particularly serious\ncrime—because the enhancement was not an element of the\noffense and because the statutory maximum for his offense\nis four years. Instead, he contends, the BIA should have\nengaged in a case-by-case analysis.\n\n We have already held that it is appropriate for the BIA to\nconsider sentencing enhancements when it determines that a\npetitioner is convicted of a particularly serious crime on a\ncase-by-case basis. Konou, 750 F.3d at 1128. As this Court\nexplained, the case-by-case analysis calls for analyzing the\n“type of sentence imposed,” and “[a]n enhanced sentence by\nits plain language can be considered a type of sentence.”\nKonou, 750 F.3d at 1128. We thus rejected the argument\nthat “the BIA cannot consider a sentencing enhancement\nwhen it determines whether a crime is particularly serious.”\nId.\n\n We now clarify that it is also appropriate for the BIA to\nconsider sentencing enhancements when it determines that a\npetitioner was convicted of a per se particularly serious\ncrime. See Garcia v. Lynch, 652 F. App’x 591, 593 (9th Cir.\n2016) (applying Konou to a per se particularly serious crime\ndetermination). Mairena’s contention that the BIA could not\nconsider his sentencing enhancement is foreclosed by the\nplain language of the statute. Section 1231(b)(3)(B) asks\nwhether the individual has been convicted of an aggravated\nfelony for which he “has been sentenced to an aggregate\nterm of imprisonment of at least 5 years.” 8 U.S.C.\n\nthe term of imprisonment [is] at least one year.” 8 U.S.C.\n§ 1101(a)(43)(F).\n\f12 MAIRENA V. BARR\n\n§ 1231(b)(3)(B). Thus, the five-year requirement is keyed\nto the “aggregate term of imprisonment” for the actual\nsentence imposed—not to the statutory maximum. Cf.\nUnited States v. Corona-Sanchez, 291 F.3d 1201, 1208–09\n(9th Cir. 2002) (en banc) (in determining whether a\nconviction categorically qualifies as an aggravated felony\nbecause it is an offense “for which the term of imprisonment\n[is] at least one year,” 8 U.S.C. § 1101(a)(43)(G), a court\n“must consider the sentence available for the crime itself,\nwithout considering separate recidivist sentencing\nenhancements”), superseded by statute on other grounds as\nexplained in United States v. Gomez-Mendez, 486 F.3d 599,\n604–05 (9th Cir. 2007); Rusz v. Ashcroft, 376 F.3d 1182,\n1184–85 (9th Cir. 2004) (applying Corona-Sanchez to a\nprovision of the Immigration and Nationality Act that\nprovides for the removal of an alien convicted of, inter alia,\n“a crime for which a sentence of one year or longer may be\nimposed” (quoting 8 U.S.C. § 1227(a)(2)(A)(i)(II))).\nNothing in the text of § 1231(b)(3)(B) prohibits the BIA\nfrom considering sentencing enhancements in computing the\naggregate term of imprisonment.\n\n Moreover, even if we accepted Mairena’s contention that\nwe should look to the statutory maximum in interpreting\n§ 1231(b)(3)(B), the statutory maximum for Mairena’s\noffense of conviction was in fact five years, not four years.\nAlthough the statutory maximum for a corporal injury\nconviction under California Penal Code § 273.5(a) was\nindeed four years, Mairena was convicted under California\nPenal Code § 273.5(e)(1), which provided for an enhanced\nstatutory maximum of five years for an individual with a\nprior corporal injury conviction within the last seven years.\nCompare Cal. Penal Code § 273.5(a) (2010), with id.\n§ 273.5(e)(1) (2010). When he was convicted in 2010,\nMairena had such a prior conviction from 2008.\n\f MAIRENA V. BARR 13\n\n Accordingly, the BIA applied the correct legal standard\nwhen it determined that Mairena was convicted of a per se\nparticularly serious crime and was therefore ineligible for\nwithholding of removal.\n\nII. The Denial of CAT Protection\n\n An individual who is ineligible for withholding of\nremoval is nevertheless eligible for “deferral of removal to\nthe country where he or she is more likely than not to be\ntortured.” 8 C.F.R. § 1208.17(a). “Torture is an extreme\nform of cruel and inhuman treatment that either (1) is not\nlawfully sanctioned by that country or (2) is lawfully\nsanctioned by that country, but defeats the object and\npurpose of CAT.” Konou, 750 F.3d at 1124 (quoting\nGarcia-Milian v. Holder, 755 F.3d 1026, 1033 (9th Cir.\n2014)); see also 8 C.F.R. § 1208.18(a)(1) (defining torture).\nIn addition, the torture must be “inflicted by or at the\ninstigation of or with the consent or acquiescence of a public\nofficial or other person acting in an official capacity.”\n8 C.F.R. § 1208.18(a)(1); see also Garcia-Milian, 755 F.3d\nat 1033.\n\n The applicant bears the burden of proving that he is\neligible for deferral of removal under CAT. 8 C.F.R.\n§ 1208.17(d)(3). “Evidence of past torture inflicted upon the\napplicant” is relevant in assessing whether torture is more\nlikely than not. 8 C.F.R. § 1208.16(c)(3); see also\nKamalthas v. I.N.S., 251 F.3d 1279, 1282 (9th Cir. 2001). A\nUnited States Department of State report “can carry an\napplicant’s burden of establishing a probability of torture,”\nbut it “can also serve to outweigh an applicant’s evidence of\na probability of torture.” Konou, 750 F.3d at 1125. In\naddition, the “testimony of the applicant, if credible, may be\nsufficient to sustain the burden of proof without\n\f14 MAIRENA V. BARR\n\ncorroboration.” 8 C.F.R. § 1208.16(c)(2); see also Konou,\n750 F.3d at 1124.\n\n Substantial evidence supports the BIA’s conclusion that\nMairena was ineligible for CAT relief. First, the IJ was not\nrequired to conduct a separate credibility analysis in\nadjudicating Mairena’s CAT claim. Contrary to Mairena’s\ncontention, the IJ was entitled to rely on the adverse\ncredibility determination in denying CAT relief, see Singh v.\nLynch, 802 F.3d 972, 977 (9th Cir. 2015), provided that the\nIJ considered other evidence in the record on country\nconditions in Nicaragua, Kamalthas, 251 F.3d at 1282–84.\nThe IJ did so in this case. Second, the record does not\ncompel the conclusion that Mairena would more likely than\nnot be tortured upon return to Nicaragua. Although both\nMairena and his family expressed fear that he would be\ntortured because of his family’s pro-Contra association,\n(1) the persecution for which his family received political\nasylum transpired thirty years ago, (2) Mairena himself was\nnever tortured, and (3) there was no evidence that Mairena\nor anyone in his family had received threats while attempting\nto regain family land from the government. Mairena has not\ncontested the accuracy of these findings on appeal; instead,\nhe essentially seems to disagree with the agency’s\nconclusion. But “our task ‘is to determine whether there is\nsubstantial evidence to support the BIA’s finding, not to\nsubstitute an analysis of which side in the factual dispute we\nfind more persuasive.’” Singh, 802 F.3d at 974–75 (quoting\nMolina-Morales v. I.N.S., 237 F.3d 1048, 1050 (9th Cir.\n2001)). The BIA also concluded that, while the Department\nof State report indicates that some former Contras were\nlikely killed by the Nicaraguan government, those\nindividuals—unlike Mairena—were well known and\ninvolved in armed confrontations. The “BIA’s interpretation\nof the Report is entitled to deference.” Konou, 750 F.3d at\n\f MAIRENA V. BARR 15\n\n1125; see also Dhital v. Mukasey, 532 F.3d 1044, 1051–52\n(9th Cir. 2008) (per curiam) (denying CAT relief where the\nDepartment of State reports indicate that torture has\noccurred, but “do not indicate that [petitioner] would face\nany particular threat of torture beyond that of which all\ncitizens of Nepal are at risk”). Finally, Mairena has not\nsuggested on appeal that either the IJ or the BIA failed to\nconsider relevant evidence in the record.\n\n Accordingly, substantial evidence supports the BIA’s\ndetermination that Mairena failed to establish that he would\nmore likely than not be tortured in Nicaragua.\n\n Petition DENIED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Danilo Mairena v. William Barr\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-07\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n DANILO ALBERTO MAIRENA, AKA No. 15-72833\\n Danilo Alberto Mairewa, AKA\\n Danilo Alberto Mariena, AKA Agency No.\\n Danilo Marieno, A027-142-897\\n Petitioner,\\n\\n v. OPINION\\n\\n WILLIAM P. BARR, Attorney General,\\n Respondent.\\n\\n\\n\\n On Petition for Review of an Order of the\\n Board of Immigration Appeals\\n\\n Submitted November 14, 2018 *\\n Pasadena, California\\n\\n Filed March 7, 2019\\n\\n\\n\\n\\n *\\n The panel unanimously concludes this case is suitable for decision\\nwithout oral argument. See Fed. R. App. P. 34(a)(2).\\n\\f2 MAIRENA V. BARR\\n\\n Before: Ronald M. Gould and Mary H. Murguia, Circuit\\n Judges, and Carol Bagley Amon, ** District Judge.\\n\\n Per Curiam Opinion\\n\\n\\n SUMMARY ***\\n\\n\\n Immigration\\n\\n Denying Danilo Mairena’s petition for review of a\\ndecision of the Board of Immigration Appeals that upheld an\\nimmigration judge’s denial of withholding of removal,\\nprotection under the Convention Against Torture (“CAT”),\\nand related relief, the panel held that it is appropriate for the\\nBIA to consider sentencing enhancements when it\\ndetermines that a petitioner was convicted of a per se\\nparticularly serious crime.\\n\\n Mairena was convicted of willful infliction of corporal\\ninjury upon the mother of his child with a prior conviction,\\nin violation of California Penal Code § 273.5(e)(1), and was\\nsentenced to five years of imprisonment: four years for the\\noffense, plus a one-year enhancement, pursuant to California\\nPenal Code § 12022.5(b)(1), for using a weapon during the\\ncommission of the offense.\\n\\n\\n\\n **\\n The Honorable Carol Bagley Amon, United States District Judge\\nfor the Eastern District of New York, sitting by designation.\\n ***\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f MAIRENA V. BARR 3\\n\\n In removal proceedings, the BIA concluded that (1)\\nMariena was statutorily ineligible for withholding of\\nremoval because he was sentenced to an aggregate term of\\nfive years of imprisonment for his corporal injury\\nconviction, factoring in the one-year enhancement; and (2)\\nthe IJ did not clearly err in determining that Mairena failed\\nto establish that he would more likely than not be tortured if\\nhe returned to Nicaragua.\\n\\n Under 8 U.S.C. § 1231(b)(3)(B)(ii), withholding of\\nremoval is not available “if the Attorney General decides\\nthat . . . the alien, having been convicted by a final judgment\\nof a particularly serious crime is a danger to the community\\nof the United States . . . .” The provision further explains:\\n“For purposes of clause (ii), an alien who has been convicted\\nof an aggravated felony (or felonies) for which the alien has\\nbeen sentenced to an aggregate term of imprisonment of at\\nleast 5 years shall be considered to have committed a\\nparticularly serious crime.”\\n\\n Mairena did not dispute that his conviction constituted\\nan aggravated felony, but argued that the BIA erred by\\nconsidering the one-year sentencing enhancement in\\ndeciding that he was sentenced to an aggregate term of\\nimprisonment of five years—and consequently convicted of\\na per se particularly serious crime—because the\\nenhancement was not an element of the offense and because\\nthe statutory maximum for his offense was four years.\\n\\n The panel noted that aggravated felonies with resulting\\nsentences of at least five years are per se particularly serious\\nand, by contrast, aggravated felonies resulting in sentences\\nfewer than five years are not per se particularly serious and\\nrequire a case-by-case analysis. The panel also observed that\\nthis court has already held that it is appropriate for the BIA\\n\\f4 MAIRENA V. BARR\\n\\nto consider sentencing enhancements when it determines\\nthat a petitioner was convicted of a particularly serious crime\\non a case-by-case basis.\\n\\n The panel now clarified that it is also appropriate for the\\nBIA to consider sentencing enhancements when it\\ndetermines that a petitioner was convicted of a per se\\nparticularly serious crime. The panel explained that\\nMairena’s contention that the BIA could not consider his\\nsentencing enhancement was foreclosed by the plain\\nlanguage of § 1231(b)(3)(B), in that the five-year\\nrequirement is keyed to the “aggregate term of\\nimprisonment” for the actual sentence imposed—not to the\\nstatutory maximum. The panel also observed that nothing in\\nthe text of the statute prohibits the BIA from considering\\nsentencing enhancements in computing the aggregate term\\nof imprisonment.\\n\\n Moreover, the panel noted that, even if it accepted\\nMairena’s contention that it should look to the statutory\\nmaximum in interpreting § 1231(b)(3)(B), the statutory\\nmaximum for Mairena’s offense of conviction was in fact\\nfive years, not four years.\\n\\n The panel also concluded that substantial evidence\\nsupported the BIA’s conclusion that Mairena was ineligible\\nfor CAT relief. The panel concluded that the IJ was not\\nrequired to conduct a separate credibility analysis in\\nadjudicating the CAT claim, explaining that the IJ was\\nentitled to rely on the adverse credibility determination in\\ndenying CAT relief, provided that the IJ considered other\\nevidence in the record on country conditions in Nicaragua.\\nThe panel concluded that the IJ did so in this case. The panel\\nalso concluded that the record did not compel the conclusion\\n\\f MAIRENA V. BARR 5\\n\\nthat Mairena would more likely than not be tortured upon\\nreturn to Nicaragua.\\n\\n\\n COUNSEL\\n\\nTania Pham, Law Offices of Tania T. Pham, Woodland\\nHills, California, for Petitioner.\\n\\nBenjamin J. Zeitlin, Attorney; Carl McIntyre, Assistant\\nDirector; Office of Immigration Litigation, Civil Division,\\nUnited States Department of Justice, Washington, D.C.; for\\nRespondent.\\n\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n Petitioner Danilo Mairena, a native and citizen of\\nNicaragua, petitions for review of a final order of the Board\\nof Immigration Appeals (“BIA”) dismissing his appeal of the\\nImmigration Judge’s (“IJ”) denial of his applications for\\nwithholding of removal, protection under the Convention\\nAgainst Torture (“CAT”), and related relief. We have\\njurisdiction under 8 U.S.C. § 1252. We deny his petition.\\n\\n BACKGROUND\\n\\n Mairena, a native and citizen of Nicaragua born in 1979,\\nentered the United States on a visitor visa in 1984 and\\nattained lawful permanent resident status in 1988. Mairena’s\\nwife, two daughters, and parents reside in the United States,\\nand he has no family left in Nicaragua.\\n\\f6 MAIRENA V. BARR\\n\\n On August 9, 2010, Mairena was convicted of willful\\ninfliction of corporal injury upon the mother of his child with\\na prior conviction, in violation of California Penal Code\\n§ 273.5(e)(1). 1 He had been previously convicted of\\ncorporal injury to a spouse in 2008. Mairena was sentenced\\nto five years of imprisonment: four years for the offense,\\nplus a one-year enhancement, pursuant to California Penal\\nCode § 12022.5(b)(1), 2 for using a weapon during the\\ncommission of the offense. That same day, Mairena was\\nalso convicted of dissuading a witness, in violation of\\nCalifornia Penal Code § 136.1(c)(1), for which he was\\nsentenced to three years of imprisonment.\\n\\n On October 10, 2013, the Department of Homeland\\nSecurity served Mairena with a Notice to Appear and\\ncharged him as removable based on those two felony\\nconvictions under § 237(a)(2)(A)(iii) of the Immigration and\\nNationality Act. See 8 U.S.C. § 1227(a)(2)(A)(iii). On\\n\\n\\n 1\\n California Penal Code § 273.5(e)(1) has been amended since\\nMairena’s conviction. Mairena was convicted under what is now\\nCalifornia Penal Code § 273.5(f)(1). Compare Cal. Penal Code\\n§ 273.5(e)(1) (2010), with id. § 273.5(f)(1) (2018).\\n\\n 2\\n Although the judgment of conviction lists California Penal Code\\n§ 12022.5(b)(1) as the source of the one-year enhancement, it\\npresumably should have listed California Penal Code § 12022(b)(1).\\nThe former authorizes an additional term of five, six, or ten years for the\\nuse of an assault weapon; the latter authorizes an additional term of one\\nyear for the use of a deadly or dangerous weapon in the commission of\\na felony. Compare Cal. Penal Code § 12022.5(b)(1) (2010), with id.\\n§ 12022(b)(1) (2010). The First Amended Information charges Mairena\\nunder the latter provision for using a pipe during the offense. In any\\nevent, the particular enhancement provision is irrelevant to our\\nresolution of the issues presented in this case.\\n\\f MAIRENA V. BARR 7\\n\\nMarch 20, 2014, the IJ (Walsh, I.J.) concluded that Mairena\\nwas removable as charged.\\n\\n On May 15, 2014, Mariena applied for asylum,\\nwithholding of removal, and CAT protection. In his\\napplication, Mairena alleged that he feared that the\\nSandinistas would kill him because his family was\\npreviously persecuted by the Sandinistas and was currently\\nfighting with President Daniel Ortega to recover seized\\nfamily property. 3 On October 15, 2014, Mairena applied for\\nadjustment of status and a waiver of inadmissibility.\\n\\n On April 1, 2015, the IJ denied all relief. As relevant to\\nthis petition, the IJ concluded as follows: (1) Mairena was\\nstatutorily ineligible for withholding of removal because he\\nwas sentenced to an aggregate term of eight years of\\nimprisonment for his two felony convictions, and therefore\\nconvicted of a per se particularly serious crime; and\\n(2) Mariena failed to carry his burden of proving that he\\nwould more likely than not be tortured if he returned to\\nNicaragua, and thus CAT protection was not warranted.\\n\\n On August 27, 2015, the BIA affirmed the IJ’s decision\\nand dismissed Mairena’s appeal. As relevant to this petition,\\nthe BIA concluded as follows: (1) Mariena was statutorily\\nineligible for withholding of removal because he was\\nsentenced to an aggregate term of five years of imprisonment\\nfor his corporal injury conviction, factoring in the one-year\\nenhancement; and (2) the IJ did not clearly err in determining\\nthat Mairena failed to establish that he would more likely\\n\\n 3\\n According to the United States Department of State Nicaragua\\n2013 Human Rights Report, contained in the record below, the\\nSandinista National Liberation Front is a political party that has\\nincreasingly concentrated political power in Nicaragua.\\n\\f8 MAIRENA V. BARR\\n\\nthan not be tortured if he returned to Nicaragua. The BIA\\ndid not rely on Mairena’s conviction for dissuading a\\nwitness, for which he was sentenced to three years of\\nimprisonment.\\n\\n On September 14, 2015, Mairena timely petitioned this\\nCourt for review.\\n\\n JURISDICTION AND STANDARD OF REVIEW\\n\\n We lack jurisdiction to review “any final order of\\nremoval against an alien who is removable” because he\\ncommitted an aggravated felony, see 8 U.S.C.\\n§§ 1252(a)(2)(C), 1227(a)(2)(A)(iii), but “we retain\\njurisdiction to decide our own jurisdiction and to resolve\\nquestions of law,” Bolanos v. Holder, 734 F.3d 875, 876 (9th\\nCir. 2013). “Although we cannot reweigh evidence to\\ndetermine if the crime was indeed particularly serious, [we]\\ncan determine whether the BIA applied the correct legal\\nstandard.” Konou v. Holder, 750 F.3d 1120, 1127 (9th Cir.\\n2014) (alteration in original) (quoting Blandino-Medina v.\\nHolder, 712 F.3d 1338, 1343 (9th Cir. 2013)).\\n\\n We also have jurisdiction to review the BIA’s denial of\\nCAT protection where, as here, “the IJ did not rely on\\n[petitioner’s] conviction . . . but instead denied relief on the\\nmerits.” Alphonsus v. Holder, 705 F.3d 1031, 1036–37 (9th\\nCir. 2013), abrogated on other grounds as recognized in\\nGuerrero v. Whitaker, 908 F.3d 541 (9th Cir. 2018).\\n\\n Where “the BIA conducts its own review of the evidence\\nand law, rather than adopting the IJ’s decision, our review is\\nlimited to the BIA’s decision, except to the extent the IJ’s\\nopinion is expressly adopted.” Zumel v. Lynch, 803 F.3d\\n463, 471 (9th Cir. 2015) (quoting Rodriguez v. Holder,\\n683 F.3d 1164, 1169 (9th Cir. 2012)). “[W]e treat the\\n\\f MAIRENA V. BARR 9\\n\\nincorporated parts of the IJ’s decision as the BIA’s.” Rivera\\nv. Mukasey, 508 F.3d 1271, 1275 (9th Cir. 2007).\\n\\n We review legal questions de novo and factual findings,\\nincluding adverse credibility determinations, for substantial\\nevidence. Vilchez v. Holder, 682 F.3d 1195, 1198–99 (9th\\nCir. 2012); Shrestha v. Holder, 590 F.3d 1034, 1039 (9th Cir.\\n2010). Under the substantial evidence standard,\\n“administrative findings of fact are conclusive unless any\\nreasonable adjudicator would be compelled to conclude to\\nthe contrary . . . .” 8 U.S.C. § 1252(b)(4)(B); see also\\nArteaga v. Mukasey, 511 F.3d 940, 944 (9th Cir. 2007)\\n(“Under the substantial evidence standard, the court upholds\\nthe BIA’s determination unless the evidence in the record\\ncompels a contrary conclusion.”).\\n\\n DISCUSSION\\n\\nI. The Particularly Serious Crime Determination\\n\\n In general, an alien is entitled to withholding of removal\\nif “the alien’s life or freedom would be threatened in that\\ncountry because of the alien’s race, religion, nationality,\\nmembership in a particular social group, or political\\nopinion.” 8 U.S.C. § 1231(b)(3)(A). But withholding of\\nremoval is not available “if the Attorney General decides\\nthat . . . the alien, having been convicted by a final judgment\\nof a particularly serious crime is a danger to the community\\nof the United States . . . .” Id. § 1231(b)(3)(B)(ii). The\\nprovision further explains:\\n\\n For purposes of clause (ii), an alien who has\\n been convicted of an aggravated felony (or\\n felonies) for which the alien has been\\n sentenced to an aggregate term of\\n imprisonment of at least 5 years shall be\\n\\f10 MAIRENA V. BARR\\n\\n considered to have committed a particularly\\n serious crime. The previous sentence shall\\n not preclude the Attorney General from\\n determining that, notwithstanding the length\\n of sentence imposed, an alien has been\\n convicted of a particularly serious crime.\\n\\nId. § 1231(b)(3)(B).\\n\\n Under the statute, the appropriate analytical lens depends\\non the length of the sentence imposed. “[A]ggravated\\nfelonies with resulting sentences of at least five years are per\\nse particularly serious . . . .” Guerrero, 908 F.3d at 545. By\\ncontrast, “aggravated felonies resulting in sentences fewer\\nthan five years are not per se particularly serious and still\\nrequire a case-by-case analysis . . . .” Blandino-Medina,\\n712 F.3d at 1347 (quoting Afridi v. Gonzales, 442 F.3d 1212,\\n1220 n.4 (9th Cir. 2006)). 4\\n\\n Mairena does not dispute that his conviction for corporal\\ninjury constitutes an aggravated felony. 5 He argues,\\n\\n\\n 4\\n Under the case-by-case analysis, this Court considers the\\nconviction in light of the Frentescu factors: “the nature of the conviction,\\nthe circumstances and underlying facts of the conviction, the type of\\nsentence imposed, and, most importantly, whether the type and\\ncircumstances of the crime indicate that the alien will be a danger to the\\ncommunity.” Konou, 750 F.3d at 1127 (quoting Matter of Frentescu,\\n18 I & N. Dec. 244, 247 (B.I.A. 1982)). We review for abuse of\\ndiscretion the BIA’s case-by-case determination that an individual was\\nconvicted of a particularly serious crime. Id.\\n\\n 5\\n “As used in immigration law, ‘aggravated felony’ is a term of art\\nreferring to the offenses enumerated under [8 U.S.C.] § 1101(a)(43).”\\nDelgado v. Holder, 648 F.3d 1095, 1101 (9th Cir. 2011) (en banc).\\nAmong the enumerated offenses is “a crime of violence . . . for which\\n\\f MAIRENA V. BARR 11\\n\\nhowever, that the BIA erred by considering the one-year\\nsentencing enhancement in deciding that he was sentenced\\nto an aggregate term of imprisonment of five years—and\\nconsequently convicted of a per se particularly serious\\ncrime—because the enhancement was not an element of the\\noffense and because the statutory maximum for his offense\\nis four years. Instead, he contends, the BIA should have\\nengaged in a case-by-case analysis.\\n\\n We have already held that it is appropriate for the BIA to\\nconsider sentencing enhancements when it determines that a\\npetitioner is convicted of a particularly serious crime on a\\ncase-by-case basis. Konou, 750 F.3d at 1128. As this Court\\nexplained, the case-by-case analysis calls for analyzing the\\n“type of sentence imposed,” and “[a]n enhanced sentence by\\nits plain language can be considered a type of sentence.”\\nKonou, 750 F.3d at 1128. We thus rejected the argument\\nthat “the BIA cannot consider a sentencing enhancement\\nwhen it determines whether a crime is particularly serious.”\\nId.\\n\\n We now clarify that it is also appropriate for the BIA to\\nconsider sentencing enhancements when it determines that a\\npetitioner was convicted of a per se particularly serious\\ncrime. See Garcia v. Lynch, 652 F. App’x 591, 593 (9th Cir.\\n2016) (applying Konou to a per se particularly serious crime\\ndetermination). Mairena’s contention that the BIA could not\\nconsider his sentencing enhancement is foreclosed by the\\nplain language of the statute. Section 1231(b)(3)(B) asks\\nwhether the individual has been convicted of an aggravated\\nfelony for which he “has been sentenced to an aggregate\\nterm of imprisonment of at least 5 years.” 8 U.S.C.\\n\\nthe term of imprisonment [is] at least one year.” 8 U.S.C.\\n§ 1101(a)(43)(F).\\n\\f12 MAIRENA V. BARR\\n\\n§ 1231(b)(3)(B). Thus, the five-year requirement is keyed\\nto the “aggregate term of imprisonment” for the actual\\nsentence imposed—not to the statutory maximum. Cf.\\nUnited States v. Corona-Sanchez, 291 F.3d 1201, 1208–09\\n(9th Cir. 2002) (en banc) (in determining whether a\\nconviction categorically qualifies as an aggravated felony\\nbecause it is an offense “for which the term of imprisonment\\n[is] at least one year,” 8 U.S.C. § 1101(a)(43)(G), a court\\n“must consider the sentence available for the crime itself,\\nwithout considering separate recidivist sentencing\\nenhancements”), superseded by statute on other grounds as\\nexplained in United States v. Gomez-Mendez, 486 F.3d 599,\\n604–05 (9th Cir. 2007); Rusz v. Ashcroft, 376 F.3d 1182,\\n1184–85 (9th Cir. 2004) (applying Corona-Sanchez to a\\nprovision of the Immigration and Nationality Act that\\nprovides for the removal of an alien convicted of, inter alia,\\n“a crime for which a sentence of one year or longer may be\\nimposed” (quoting 8 U.S.C. § 1227(a)(2)(A)(i)(II))).\\nNothing in the text of § 1231(b)(3)(B) prohibits the BIA\\nfrom considering sentencing enhancements in computing the\\naggregate term of imprisonment.\\n\\n Moreover, even if we accepted Mairena’s contention that\\nwe should look to the statutory maximum in interpreting\\n§ 1231(b)(3)(B), the statutory maximum for Mairena’s\\noffense of conviction was in fact five years, not four years.\\nAlthough the statutory maximum for a corporal injury\\nconviction under California Penal Code § 273.5(a) was\\nindeed four years, Mairena was convicted under California\\nPenal Code § 273.5(e)(1), which provided for an enhanced\\nstatutory maximum of five years for an individual with a\\nprior corporal injury conviction within the last seven years.\\nCompare Cal. Penal Code § 273.5(a) (2010), with id.\\n§ 273.5(e)(1) (2010). When he was convicted in 2010,\\nMairena had such a prior conviction from 2008.\\n\\f MAIRENA V. BARR 13\\n\\n Accordingly, the BIA applied the correct legal standard\\nwhen it determined that Mairena was convicted of a per se\\nparticularly serious crime and was therefore ineligible for\\nwithholding of removal.\\n\\nII. The Denial of CAT Protection\\n\\n An individual who is ineligible for withholding of\\nremoval is nevertheless eligible for “deferral of removal to\\nthe country where he or she is more likely than not to be\\ntortured.” 8 C.F.R. § 1208.17(a). “Torture is an extreme\\nform of cruel and inhuman treatment that either (1) is not\\nlawfully sanctioned by that country or (2) is lawfully\\nsanctioned by that country, but defeats the object and\\npurpose of CAT.” Konou, 750 F.3d at 1124 (quoting\\nGarcia-Milian v. Holder, 755 F.3d 1026, 1033 (9th Cir.\\n2014)); see also 8 C.F.R. § 1208.18(a)(1) (defining torture).\\nIn addition, the torture must be “inflicted by or at the\\ninstigation of or with the consent or acquiescence of a public\\nofficial or other person acting in an official capacity.”\\n8 C.F.R. § 1208.18(a)(1); see also Garcia-Milian, 755 F.3d\\nat 1033.\\n\\n The applicant bears the burden of proving that he is\\neligible for deferral of removal under CAT. 8 C.F.R.\\n§ 1208.17(d)(3). “Evidence of past torture inflicted upon the\\napplicant” is relevant in assessing whether torture is more\\nlikely than not. 8 C.F.R. § 1208.16(c)(3); see also\\nKamalthas v. I.N.S., 251 F.3d 1279, 1282 (9th Cir. 2001). A\\nUnited States Department of State report “can carry an\\napplicant’s burden of establishing a probability of torture,”\\nbut it “can also serve to outweigh an applicant’s evidence of\\na probability of torture.” Konou, 750 F.3d at 1125. In\\naddition, the “testimony of the applicant, if credible, may be\\nsufficient to sustain the burden of proof without\\n\\f14 MAIRENA V. BARR\\n\\ncorroboration.” 8 C.F.R. § 1208.16(c)(2); see also Konou,\\n750 F.3d at 1124.\\n\\n Substantial evidence supports the BIA’s conclusion that\\nMairena was ineligible for CAT relief. First, the IJ was not\\nrequired to conduct a separate credibility analysis in\\nadjudicating Mairena’s CAT claim. Contrary to Mairena’s\\ncontention, the IJ was entitled to rely on the adverse\\ncredibility determination in denying CAT relief, see Singh v.\\nLynch, 802 F.3d 972, 977 (9th Cir. 2015), provided that the\\nIJ considered other evidence in the record on country\\nconditions in Nicaragua, Kamalthas, 251 F.3d at 1282–84.\\nThe IJ did so in this case. Second, the record does not\\ncompel the conclusion that Mairena would more likely than\\nnot be tortured upon return to Nicaragua. Although both\\nMairena and his family expressed fear that he would be\\ntortured because of his family’s pro-Contra association,\\n(1) the persecution for which his family received political\\nasylum transpired thirty years ago, (2) Mairena himself was\\nnever tortured, and (3) there was no evidence that Mairena\\nor anyone in his family had received threats while attempting\\nto regain family land from the government. Mairena has not\\ncontested the accuracy of these findings on appeal; instead,\\nhe essentially seems to disagree with the agency’s\\nconclusion. But “our task ‘is to determine whether there is\\nsubstantial evidence to support the BIA’s finding, not to\\nsubstitute an analysis of which side in the factual dispute we\\nfind more persuasive.’” Singh, 802 F.3d at 974–75 (quoting\\nMolina-Morales v. I.N.S., 237 F.3d 1048, 1050 (9th Cir.\\n2001)). The BIA also concluded that, while the Department\\nof State report indicates that some former Contras were\\nlikely killed by the Nicaraguan government, those\\nindividuals—unlike Mairena—were well known and\\ninvolved in armed confrontations. The “BIA’s interpretation\\nof the Report is entitled to deference.” Konou, 750 F.3d at\\n\\f MAIRENA V. BARR 15\\n\\n1125; see also Dhital v. Mukasey, 532 F.3d 1044, 1051–52\\n(9th Cir. 2008) (per curiam) (denying CAT relief where the\\nDepartment of State reports indicate that torture has\\noccurred, but “do not indicate that [petitioner] would face\\nany particular threat of torture beyond that of which all\\ncitizens of Nepal are at risk”). Finally, Mairena has not\\nsuggested on appeal that either the IJ or the BIA failed to\\nconsider relevant evidence in the record.\\n\\n Accordingly, substantial evidence supports the BIA’s\\ndetermination that Mairena failed to establish that he would\\nmore likely than not be tortured in Nicaragua.\\n\\n Petition DENIED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1276,"cells":{"id":{"kind":"number","value":4598128,"string":"4,598,128"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Thomas Anderson v. State Farm Mutual Auto. Ins."},"decision_date":{"kind":"string","value":"2019-03-08"},"docket_number":{"kind":"string","value":"15-35981"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: M. Margaret McKeown and Michelle T. Friedland, Circuit Judges, and Susan R. Bolton, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"McKEOWN, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n PATRICIA ANDERSON; THOMAS No. 15-35981\\n ANDERSON,\\n Plaintiffs-Appellants, D.C. No.\\n 3:15-cv-05159-RBL\\n v.\\n\\n STATE FARM MUTUAL OPINION\\n AUTOMOBILE INSURANCE\\n COMPANY,\\n Defendant-Appellee.\\n\\n\\n\\n Appeal from the United States District Court\\n for the Western District of Washington\\n Ronald B. Leighton, District Judge, Presiding\\n\\n Argued and Submitted November 7, 2018\\n Seattle, Washington\\n\\n Filed March 8, 2019\\n\\nBefore: M. Margaret McKeown and Michelle T. Friedland,\\n Circuit Judges, and Susan R. Bolton, * District Judge.\\n\\n Opinion by Judge McKeown\\n\\n *\\n The Honorable Susan R. Bolton, United States District Judge for\\nthe District of Arizona, sitting by designation.\\n\\f2 ANDERSON V. STATE FARM\\n\\n SUMMARY **\\n\\n\\n Removal\\n\\n The panel affirmed the district court’s denial of\\nplaintiffs’ motion to remand to state court, and held that the\\ndefendant insurer timely removed an insurance coverage\\ncase to federal court.\\n\\n The panel held that receipt of an initial pleading by a\\nstatutorily designated agent did not begin the thirty-day\\nremoval clock under 28 U.S.C. § 1446(b)(1), and it was\\nactual receipt by the insurer that started the removal clock.\\nApplying this rule, the panel concluded that the insurer\\ntimely removed the case.\\n\\n In a contemporaneously filed memorandum disposition,\\nthe panel affirmed in part and vacated in part the remainder\\nof the judgment, and remanded for further proceedings.\\n\\n\\n COUNSEL\\n\\nMark D. Herman (argued), Covington & Burling LLP,\\nWashington, D.C., for Plaintiffs-Appellants.\\n\\nGregory Scott Worden (argued) and Donna M. Chamberlin,\\nLewis Brisbois Bisgaard & Smith LLP, Seattle, Washington,\\nfor Defendant-Appellee.\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f ANDERSON V. STATE FARM 3\\n\\n OPINION\\n\\nMcKEOWN, Circuit Judge:\\n\\n Patricia Anderson and Thomas Anderson’s single-car\\naccident in 1998 began a saga of litigation culminating in\\nthis appeal. Little of the tragic history involving nine\\ndifferent trial and appellate proceedings is relevant to this\\nopinion, in which we determine only whether State Farm\\ntimely removed the case to federal court. 1 Addressing an\\nissue of first impression in this circuit, we join the Fourth\\nCircuit in holding that receipt of an initial pleading by a\\nstatutorily designated agent does not begin the thirty-day\\nremoval clock under 28 U.S.C. § 1446(b)(1), and that it was\\ninstead actual receipt by State Farm that started the removal\\nclock. See Elliott v. Am. States Ins. Co., 883 F.3d 384, 394\\n(4th Cir. 2018). Applying this rule, we conclude that State\\nFarm timely removed the case.\\n\\n The Andersons sued State Farm in Washington state\\ncourt. Because State Farm is an out-of-state (or “foreign”)\\ninsurer, state law designates Washington’s Insurance\\nCommissioner as State Farm’s statutory agent.\\nRCW 48.05.200(1). To serve legal process on State Farm,\\nthe Andersons served the Commissioner, who forwarded the\\ncomplaint to State Farm’s designated recipient.\\nRCW 4.28.080(7)(a), 48.05.200(1)–(2). The Andersons\\nserved the Commissioner on February 9, 2015, and State\\nFarm received the forwarded complaint four days later, on\\nFebruary 13. On March 16, State Farm removed the case to\\nfederal court.\\n\\n\\n 1\\n We address the remaining issues in a memorandum disposition\\nfiled contemporaneously with this opinion.\\n\\f4 ANDERSON V. STATE FARM\\n\\n The parties agree on how to count to thirty, but they\\ndisagree over when to begin counting. According to the\\nAndersons, removal was untimely because the thirty-day\\nremoval clock began on February 9, when the Commissioner\\nreceived the complaint. State Farm contends removal was\\ntimely as calculated from February 13, when the forwarded\\ncopy of the complaint reached State Farm’s designated\\nrecipient. 2 The district court agreed with State Farm and\\ndenied the Andersons’ motion to remand. We have\\njurisdiction under 28 U.S.C. § 1291 and review de novo the\\ndistrict court’s ruling. Patterson v. Int’l Bhd. of Teamsters,\\nLocal 959, 121 F.3d 1345, 1348 (9th Cir. 1997).\\n\\n We begin our analysis with the statutory text, but it does\\nnot take us far. A notice of removal must be filed “within\\n30 days after the receipt by the defendant, through service or\\notherwise, of a copy of the initial pleading.” 28 U.S.C.\\n§ 1446(b)(1). One reading suggests removal was timely\\nbecause the actual defendant—State Farm—did not receive\\nnotice until receiving the mailing from the Commissioner.\\nThat reading, however, does not account for the effect of\\nservice on a statutory agent. “[T]he statute itself says\\nnothing about service on a statutory agent,” Elliott, 883 F.3d\\nat 392, and an agent designated by the state legislature to\\nreceive service fundamentally differs from a defendant’s\\nagent-in-fact, because the defendant has no meaningful say\\nin or control over the former. See 14C Charles Alan Wright,\\net al., Federal Practice & Procedure § 3731 (Rev. 4th ed.\\n2018) (“Realistically speaking, of course, statutory agents\\n\\n\\n\\n 2\\n Thirty days from February 13, 2015, was Sunday, March 15, 2015,\\nso the notice of removal would have been timely filed on Monday,\\nMarch 16, 2015. See Fed. R. Civ. P. 6(a)(1)(C).\\n\\f ANDERSON V. STATE FARM 5\\n\\nare not true agents but merely are a medium for transmitting\\nthe relevant papers.”).\\n\\n Washington law, however, lends some force to the\\nposition that service on a statutorily designated agent should\\nqualify as “receipt by the defendant” for purposes of\\n28 U.S.C. § 1446(b)(1). As a practical matter, a company\\ncannot receive anything except through its agents, and\\nWashington equates service upon the Commissioner with\\nservice upon the insurer. RCW 48.05.200(1). 3 Indeed,\\nplaintiffs must serve foreign insurers via the Commissioner.\\nId.; RCW 4.28.080(7)(a). But the Supreme Court has\\ncautioned against interpreting the federal removal statute\\nwith reference to state law. See Chi., R.I. & P.R. Co. v.\\nStude, 346 U.S. 574, 580 (1954) (“For the purpose of\\nremoval, the federal law determines who is plaintiff and who\\nis defendant. It is a question of the construction of the\\nfederal statute on removal, and not the state statute.”). So\\nwe decline to place much weight on Washington law and\\nconclude that the federal removal statute is, at most,\\nambiguous on this point. Cf. Elliott, 883 F.3d at 392 (“[A]\\nstraightforward reading . . . supports holding that the 30-day\\nperiod . . . is not triggered by service on a statutory agent\\n. . . .”).\\n\\n\\n\\n 3\\n RCW 48.05.200(1) provides: “Each authorized foreign or alien\\ninsurer must appoint the commissioner as its attorney to receive service\\nof, and upon whom must be served, all legal process issued against it in\\nthis state upon causes of action arising within this state. Service upon\\nthe commissioner as attorney constitutes service upon the insurer.\\nService of legal process against the insurer can be had only by service\\nupon the commissioner, except actions upon contractor bonds pursuant\\nto RCW 18.27.040, where service may be upon the department of labor\\nand industries.”\\n\\f6 ANDERSON V. STATE FARM\\n\\n The legislative history, on the other hand, is\\nunambiguous—it clearly demonstrates Congress’ intent to\\navoid disparate application of the removal statute due to\\ndifferences in state law. See Murphy Bros., Inc. v. Michetti\\nPipe Stringing, Inc., 526 U.S. 344, 351–53 (1999)\\n(discussing congressional intent at length). Before Congress\\nenacted § 1446(b), a defendant could remove at “any time\\nbefore the defendant [wa]s required by the laws of the State\\n. . . to answer or plead to the declaration or complaint.”\\n28 U.S.C. § 72 (1940). The practical effects of this regime\\nvaried widely among the states. Seeking a replacement that\\nwould “give adequate time and operate uniformly\\nthroughout the Federal jurisdiction,” Congress enacted\\n§ 1446(b). H.R. Rep. No. 80-308, at A135 (1947).\\n\\n As initially enacted, however, § 1446(b) suffered the\\nsame infirmity: the removal deadline depended on\\nprocedural rules that varied by state. See Murphy Bros.,\\n526 U.S. at 351. In other words, the removal clock began\\n“after commencement of the action or service of process,\\nwhichever [wa]s later.” Act of June 25, 1948, Pub. L. No.\\n80-773, § 1446(b), 62 Stat. 869, 939. In states like New\\nYork, which allowed a plaintiff to commence an action by\\nserving the summons without the complaint, the removal\\ndeadline could expire before the defendant could access the\\ncomplaint. H.R. Rep. No. 81-352, at 14 (1949); Murphy\\nBros., 526 U.S. at 351.\\n\\n Recognizing this disparity, the next Congress amended\\nthe statute to provide each defendant with “20 days after he\\nhas received (or it has been made available to him) a copy of\\nthe initial pleading.” S. Rep. No. 81-303, at 6 (1949). In all\\nrelevant respects, the 1949 version of § 1446 mirrors today’s\\nversion: “The petition for removal of a civil action or\\nproceeding shall be filed within twenty days after the receipt\\n\\f ANDERSON V. STATE FARM 7\\n\\nby the defendant, through service or otherwise, of a copy of\\nthe initial pleading . . . .” 4 Act of May 24, 1949, Pub. L. No.\\n81-72, § 83, 63 Stat. 89, 101. In both versions of the statute,\\nthe key point was to peg the time calculation to receipt by\\nthe defendant.\\n\\n Intertwining the removal statute with state-specific\\nidiosyncrasies would thwart Congress’ twin aims of\\nadequate time to respond and uniform application of the rule.\\nMurphy Bros., 526 U.S. at 351. Indeed, if delivery to a\\nstatutorily designated agent began the removal clock, the\\neffective time a defendant had to remove would depend not\\nonly on differences in state law, but also on the efficiency of\\nstate agencies in each instance. The interpretation the\\nAndersons advance would compound disfavored interstate\\ndifferences by adding case-by-case, intrastate discrepancies.\\nThat arrangement cannot be reconciled with Congress’\\nunambiguous intent to provide each defendant with a fixed\\nand adequate amount of time, after obtaining access to the\\ncomplaint, to decide whether to remove. See Elliott,\\n883 F.3d at 393 (“Serving a statutory agent does not\\nguarantee that the defendant is provided with actual notice\\nof the complaint or adequate time to decide whether to\\nremove a case.”).\\n\\n As the Andersons point out, the Court held in Murphy\\nBrothers that “a named defendant’s time to remove is\\ntriggered by simultaneous service of the summons and\\ncomplaint.” 526 U.S. at 347–48. But this holding addressed\\nwhat must be received, not who must receive it. Id. The\\nCourt did not consider the implications of serving a statutory\\n\\n 4\\n Congress later extended the removal window from twenty days to\\nthirty days. Act of Sept. 29, 1965, Pub. L. No. 89-215, 79 Stat. 887, 887;\\n28 U.S.C. § 1446(b)(1).\\n\\f8 ANDERSON V. STATE FARM\\n\\nagent before actual receipt by the defendant. And although\\nit addressed a different question, the Court announced a\\nprinciple of fairness that is pertinent here:\\n\\n We read Congress’ provisions for removal in\\n light of a bedrock principle: An individual or\\n entity named as a defendant is not obliged to\\n engage in litigation unless notified of the\\n action, and brought under a court’s authority,\\n by formal process.\\n\\nId. at 347. That “bedrock principle” regarding the\\nimportance of actual notice confirms that serving the\\nCommissioner did not provide the necessary notice to State\\nFarm of the suit—that occurred only when State Farm’s\\ndesignated recipient received the complaint.\\n\\n We are also unconvinced by the Andersons’ invocation\\nof the interpretive canon counseling strict construction of\\nremoval statutes. See Gaus v. Miles, Inc., 980 F.2d 564, 566\\n(9th Cir. 1992). The canon is instructive, but it is not\\nabsolute and cannot overcome a “clear [contrary] command\\nfrom both Congress and the Supreme Court.” Durham v.\\nLockheed Martin Corp., 445 F.3d 1247, 1252 (9th Cir.\\n2006). The legislative history and Murphy Brothers provide\\njust such a clear command.\\n\\n Nor are we persuaded by the Andersons’ suggestion that\\nour ruling invites collateral litigation over “who qualifies” as\\na recipient. Our decision cleaves to “settled [district court]\\nlaw,” 14C Charles Alan Wright, et al., Federal Practice &\\nProcedure § 3731 & n.20 (4th ed.), and the current regime\\nhas hardly spawned uncertainty on this point.\\n\\n We join the Fourth Circuit and hold that the thirty-day\\nremoval clock under 28 U.S.C. § 1446(b)(1) does not begin\\n\\f ANDERSON V. STATE FARM 9\\n\\nupon service on and receipt by a statutorily designated agent,\\nand began in this case only when State Farm actually\\nreceived the Andersons’ complaint. 5 Elliott, 883 F.3d\\nat 391–94. State Farm’s removal was timely, and we affirm\\nthe district court’s denial of the Andersons’ motion to\\nremand. For reasons explained in our contemporaneously\\nfiled memorandum disposition, we affirm in part and vacate\\nin part the remainder of the judgment and remand for further\\nproceedings.\\n\\n AFFIRMED WITH RESPECT TO THE DENIAL\\nOF THE MOTION TO REMAND.\\n\\n Each party shall bear its own costs on appeal.\\n\\n\\n\\n\\n 5\\n The Washington law at issue here, RCW 48.05.200(1), and the\\nNorth Carolina statute at issue in Elliott, 883 F.3d at 390–91 (addressing\\nN.C. Gen. Stat. § 58-16-5), designate a statutory agent that foreign\\ninsurers must authorize to accept for service of process. Other states,\\nsuch as California, require foreign insurers to designate an agent, but do\\nnot designate who that agent is. See Cal. Ins. Code § 1600 (“[E]very\\nforeign insurer [must designate] an agent for service of process. The\\nagent designated may be any person residing in this state, including, but\\nnot limited to, any corporate officer of the insurer.”). Such a requirement\\nis not before us, and we take no position regarding its interplay with the\\nfederal removal statute.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4375381/\", \"author_raw\": \"McKEOWN, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"M MARGARET MCKEOWN"},"judge2":{"kind":"string","value":"MICHELLE T FRIEDLAND"},"judge3":{"kind":"string","value":"SUSAN R BOLTON"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4598128/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Thomas Anderson v. State Farm Mutual Auto. Ins.\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-08\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n PATRICIA ANDERSON; THOMAS No. 15-35981\n ANDERSON,\n Plaintiffs-Appellants, D.C. No.\n 3:15-cv-05159-RBL\n v.\n\n STATE FARM MUTUAL OPINION\n AUTOMOBILE INSURANCE\n COMPANY,\n Defendant-Appellee.\n\n\n\n Appeal from the United States District Court\n for the Western District of Washington\n Ronald B. Leighton, District Judge, Presiding\n\n Argued and Submitted November 7, 2018\n Seattle, Washington\n\n Filed March 8, 2019\n\nBefore: M. Margaret McKeown and Michelle T. Friedland,\n Circuit Judges, and Susan R. Bolton, * District Judge.\n\n Opinion by Judge McKeown\n\n *\n The Honorable Susan R. Bolton, United States District Judge for\nthe District of Arizona, sitting by designation.\n\f2 ANDERSON V. STATE FARM\n\n SUMMARY **\n\n\n Removal\n\n The panel affirmed the district court’s denial of\nplaintiffs’ motion to remand to state court, and held that the\ndefendant insurer timely removed an insurance coverage\ncase to federal court.\n\n The panel held that receipt of an initial pleading by a\nstatutorily designated agent did not begin the thirty-day\nremoval clock under 28 U.S.C. § 1446(b)(1), and it was\nactual receipt by the insurer that started the removal clock.\nApplying this rule, the panel concluded that the insurer\ntimely removed the case.\n\n In a contemporaneously filed memorandum disposition,\nthe panel affirmed in part and vacated in part the remainder\nof the judgment, and remanded for further proceedings.\n\n\n COUNSEL\n\nMark D. Herman (argued), Covington & Burling LLP,\nWashington, D.C., for Plaintiffs-Appellants.\n\nGregory Scott Worden (argued) and Donna M. Chamberlin,\nLewis Brisbois Bisgaard & Smith LLP, Seattle, Washington,\nfor Defendant-Appellee.\n\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f ANDERSON V. STATE FARM 3\n\n OPINION\n\nMcKEOWN, Circuit Judge:\n\n Patricia Anderson and Thomas Anderson’s single-car\naccident in 1998 began a saga of litigation culminating in\nthis appeal. Little of the tragic history involving nine\ndifferent trial and appellate proceedings is relevant to this\nopinion, in which we determine only whether State Farm\ntimely removed the case to federal court. 1 Addressing an\nissue of first impression in this circuit, we join the Fourth\nCircuit in holding that receipt of an initial pleading by a\nstatutorily designated agent does not begin the thirty-day\nremoval clock under 28 U.S.C. § 1446(b)(1), and that it was\ninstead actual receipt by State Farm that started the removal\nclock. See Elliott v. Am. States Ins. Co., 883 F.3d 384, 394\n(4th Cir. 2018). Applying this rule, we conclude that State\nFarm timely removed the case.\n\n The Andersons sued State Farm in Washington state\ncourt. Because State Farm is an out-of-state (or “foreign”)\ninsurer, state law designates Washington’s Insurance\nCommissioner as State Farm’s statutory agent.\nRCW 48.05.200(1). To serve legal process on State Farm,\nthe Andersons served the Commissioner, who forwarded the\ncomplaint to State Farm’s designated recipient.\nRCW 4.28.080(7)(a), 48.05.200(1)–(2). The Andersons\nserved the Commissioner on February 9, 2015, and State\nFarm received the forwarded complaint four days later, on\nFebruary 13. On March 16, State Farm removed the case to\nfederal court.\n\n\n 1\n We address the remaining issues in a memorandum disposition\nfiled contemporaneously with this opinion.\n\f4 ANDERSON V. STATE FARM\n\n The parties agree on how to count to thirty, but they\ndisagree over when to begin counting. According to the\nAndersons, removal was untimely because the thirty-day\nremoval clock began on February 9, when the Commissioner\nreceived the complaint. State Farm contends removal was\ntimely as calculated from February 13, when the forwarded\ncopy of the complaint reached State Farm’s designated\nrecipient. 2 The district court agreed with State Farm and\ndenied the Andersons’ motion to remand. We have\njurisdiction under 28 U.S.C. § 1291 and review de novo the\ndistrict court’s ruling. Patterson v. Int’l Bhd. of Teamsters,\nLocal 959, 121 F.3d 1345, 1348 (9th Cir. 1997).\n\n We begin our analysis with the statutory text, but it does\nnot take us far. A notice of removal must be filed “within\n30 days after the receipt by the defendant, through service or\notherwise, of a copy of the initial pleading.” 28 U.S.C.\n§ 1446(b)(1). One reading suggests removal was timely\nbecause the actual defendant—State Farm—did not receive\nnotice until receiving the mailing from the Commissioner.\nThat reading, however, does not account for the effect of\nservice on a statutory agent. “[T]he statute itself says\nnothing about service on a statutory agent,” Elliott, 883 F.3d\nat 392, and an agent designated by the state legislature to\nreceive service fundamentally differs from a defendant’s\nagent-in-fact, because the defendant has no meaningful say\nin or control over the former. See 14C Charles Alan Wright,\net al., Federal Practice & Procedure § 3731 (Rev. 4th ed.\n2018) (“Realistically speaking, of course, statutory agents\n\n\n\n 2\n Thirty days from February 13, 2015, was Sunday, March 15, 2015,\nso the notice of removal would have been timely filed on Monday,\nMarch 16, 2015. See Fed. R. Civ. P. 6(a)(1)(C).\n\f ANDERSON V. STATE FARM 5\n\nare not true agents but merely are a medium for transmitting\nthe relevant papers.”).\n\n Washington law, however, lends some force to the\nposition that service on a statutorily designated agent should\nqualify as “receipt by the defendant” for purposes of\n28 U.S.C. § 1446(b)(1). As a practical matter, a company\ncannot receive anything except through its agents, and\nWashington equates service upon the Commissioner with\nservice upon the insurer. RCW 48.05.200(1). 3 Indeed,\nplaintiffs must serve foreign insurers via the Commissioner.\nId.; RCW 4.28.080(7)(a). But the Supreme Court has\ncautioned against interpreting the federal removal statute\nwith reference to state law. See Chi., R.I. & P.R. Co. v.\nStude, 346 U.S. 574, 580 (1954) (“For the purpose of\nremoval, the federal law determines who is plaintiff and who\nis defendant. It is a question of the construction of the\nfederal statute on removal, and not the state statute.”). So\nwe decline to place much weight on Washington law and\nconclude that the federal removal statute is, at most,\nambiguous on this point. Cf. Elliott, 883 F.3d at 392 (“[A]\nstraightforward reading . . . supports holding that the 30-day\nperiod . . . is not triggered by service on a statutory agent\n. . . .”).\n\n\n\n 3\n RCW 48.05.200(1) provides: “Each authorized foreign or alien\ninsurer must appoint the commissioner as its attorney to receive service\nof, and upon whom must be served, all legal process issued against it in\nthis state upon causes of action arising within this state. Service upon\nthe commissioner as attorney constitutes service upon the insurer.\nService of legal process against the insurer can be had only by service\nupon the commissioner, except actions upon contractor bonds pursuant\nto RCW 18.27.040, where service may be upon the department of labor\nand industries.”\n\f6 ANDERSON V. STATE FARM\n\n The legislative history, on the other hand, is\nunambiguous—it clearly demonstrates Congress’ intent to\navoid disparate application of the removal statute due to\ndifferences in state law. See Murphy Bros., Inc. v. Michetti\nPipe Stringing, Inc., 526 U.S. 344, 351–53 (1999)\n(discussing congressional intent at length). Before Congress\nenacted § 1446(b), a defendant could remove at “any time\nbefore the defendant [wa]s required by the laws of the State\n. . . to answer or plead to the declaration or complaint.”\n28 U.S.C. § 72 (1940). The practical effects of this regime\nvaried widely among the states. Seeking a replacement that\nwould “give adequate time and operate uniformly\nthroughout the Federal jurisdiction,” Congress enacted\n§ 1446(b). H.R. Rep. No. 80-308, at A135 (1947).\n\n As initially enacted, however, § 1446(b) suffered the\nsame infirmity: the removal deadline depended on\nprocedural rules that varied by state. See Murphy Bros.,\n526 U.S. at 351. In other words, the removal clock began\n“after commencement of the action or service of process,\nwhichever [wa]s later.” Act of June 25, 1948, Pub. L. No.\n80-773, § 1446(b), 62 Stat. 869, 939. In states like New\nYork, which allowed a plaintiff to commence an action by\nserving the summons without the complaint, the removal\ndeadline could expire before the defendant could access the\ncomplaint. H.R. Rep. No. 81-352, at 14 (1949); Murphy\nBros., 526 U.S. at 351.\n\n Recognizing this disparity, the next Congress amended\nthe statute to provide each defendant with “20 days after he\nhas received (or it has been made available to him) a copy of\nthe initial pleading.” S. Rep. No. 81-303, at 6 (1949). In all\nrelevant respects, the 1949 version of § 1446 mirrors today’s\nversion: “The petition for removal of a civil action or\nproceeding shall be filed within twenty days after the receipt\n\f ANDERSON V. STATE FARM 7\n\nby the defendant, through service or otherwise, of a copy of\nthe initial pleading . . . .” 4 Act of May 24, 1949, Pub. L. No.\n81-72, § 83, 63 Stat. 89, 101. In both versions of the statute,\nthe key point was to peg the time calculation to receipt by\nthe defendant.\n\n Intertwining the removal statute with state-specific\nidiosyncrasies would thwart Congress’ twin aims of\nadequate time to respond and uniform application of the rule.\nMurphy Bros., 526 U.S. at 351. Indeed, if delivery to a\nstatutorily designated agent began the removal clock, the\neffective time a defendant had to remove would depend not\nonly on differences in state law, but also on the efficiency of\nstate agencies in each instance. The interpretation the\nAndersons advance would compound disfavored interstate\ndifferences by adding case-by-case, intrastate discrepancies.\nThat arrangement cannot be reconciled with Congress’\nunambiguous intent to provide each defendant with a fixed\nand adequate amount of time, after obtaining access to the\ncomplaint, to decide whether to remove. See Elliott,\n883 F.3d at 393 (“Serving a statutory agent does not\nguarantee that the defendant is provided with actual notice\nof the complaint or adequate time to decide whether to\nremove a case.”).\n\n As the Andersons point out, the Court held in Murphy\nBrothers that “a named defendant’s time to remove is\ntriggered by simultaneous service of the summons and\ncomplaint.” 526 U.S. at 347–48. But this holding addressed\nwhat must be received, not who must receive it. Id. The\nCourt did not consider the implications of serving a statutory\n\n 4\n Congress later extended the removal window from twenty days to\nthirty days. Act of Sept. 29, 1965, Pub. L. No. 89-215, 79 Stat. 887, 887;\n28 U.S.C. § 1446(b)(1).\n\f8 ANDERSON V. STATE FARM\n\nagent before actual receipt by the defendant. And although\nit addressed a different question, the Court announced a\nprinciple of fairness that is pertinent here:\n\n We read Congress’ provisions for removal in\n light of a bedrock principle: An individual or\n entity named as a defendant is not obliged to\n engage in litigation unless notified of the\n action, and brought under a court’s authority,\n by formal process.\n\nId. at 347. That “bedrock principle” regarding the\nimportance of actual notice confirms that serving the\nCommissioner did not provide the necessary notice to State\nFarm of the suit—that occurred only when State Farm’s\ndesignated recipient received the complaint.\n\n We are also unconvinced by the Andersons’ invocation\nof the interpretive canon counseling strict construction of\nremoval statutes. See Gaus v. Miles, Inc., 980 F.2d 564, 566\n(9th Cir. 1992). The canon is instructive, but it is not\nabsolute and cannot overcome a “clear [contrary] command\nfrom both Congress and the Supreme Court.” Durham v.\nLockheed Martin Corp., 445 F.3d 1247, 1252 (9th Cir.\n2006). The legislative history and Murphy Brothers provide\njust such a clear command.\n\n Nor are we persuaded by the Andersons’ suggestion that\nour ruling invites collateral litigation over “who qualifies” as\na recipient. Our decision cleaves to “settled [district court]\nlaw,” 14C Charles Alan Wright, et al., Federal Practice &\nProcedure § 3731 & n.20 (4th ed.), and the current regime\nhas hardly spawned uncertainty on this point.\n\n We join the Fourth Circuit and hold that the thirty-day\nremoval clock under 28 U.S.C. § 1446(b)(1) does not begin\n\f ANDERSON V. STATE FARM 9\n\nupon service on and receipt by a statutorily designated agent,\nand began in this case only when State Farm actually\nreceived the Andersons’ complaint. 5 Elliott, 883 F.3d\nat 391–94. State Farm’s removal was timely, and we affirm\nthe district court’s denial of the Andersons’ motion to\nremand. For reasons explained in our contemporaneously\nfiled memorandum disposition, we affirm in part and vacate\nin part the remainder of the judgment and remand for further\nproceedings.\n\n AFFIRMED WITH RESPECT TO THE DENIAL\nOF THE MOTION TO REMAND.\n\n Each party shall bear its own costs on appeal.\n\n\n\n\n 5\n The Washington law at issue here, RCW 48.05.200(1), and the\nNorth Carolina statute at issue in Elliott, 883 F.3d at 390–91 (addressing\nN.C. Gen. Stat. § 58-16-5), designate a statutory agent that foreign\ninsurers must authorize to accept for service of process. Other states,\nsuch as California, require foreign insurers to designate an agent, but do\nnot designate who that agent is. See Cal. Ins. Code § 1600 (“[E]very\nforeign insurer [must designate] an agent for service of process. The\nagent designated may be any person residing in this state, including, but\nnot limited to, any corporate officer of the insurer.”). Such a requirement\nis not before us, and we take no position regarding its interplay with the\nfederal removal statute.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Thomas Anderson v. State Farm Mutual Auto. Ins.\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-08\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n PATRICIA ANDERSON; THOMAS No. 15-35981\\n ANDERSON,\\n Plaintiffs-Appellants, D.C. No.\\n 3:15-cv-05159-RBL\\n v.\\n\\n STATE FARM MUTUAL OPINION\\n AUTOMOBILE INSURANCE\\n COMPANY,\\n Defendant-Appellee.\\n\\n\\n\\n Appeal from the United States District Court\\n for the Western District of Washington\\n Ronald B. Leighton, District Judge, Presiding\\n\\n Argued and Submitted November 7, 2018\\n Seattle, Washington\\n\\n Filed March 8, 2019\\n\\nBefore: M. Margaret McKeown and Michelle T. Friedland,\\n Circuit Judges, and Susan R. Bolton, * District Judge.\\n\\n Opinion by Judge McKeown\\n\\n *\\n The Honorable Susan R. Bolton, United States District Judge for\\nthe District of Arizona, sitting by designation.\\n\\f2 ANDERSON V. STATE FARM\\n\\n SUMMARY **\\n\\n\\n Removal\\n\\n The panel affirmed the district court’s denial of\\nplaintiffs’ motion to remand to state court, and held that the\\ndefendant insurer timely removed an insurance coverage\\ncase to federal court.\\n\\n The panel held that receipt of an initial pleading by a\\nstatutorily designated agent did not begin the thirty-day\\nremoval clock under 28 U.S.C. § 1446(b)(1), and it was\\nactual receipt by the insurer that started the removal clock.\\nApplying this rule, the panel concluded that the insurer\\ntimely removed the case.\\n\\n In a contemporaneously filed memorandum disposition,\\nthe panel affirmed in part and vacated in part the remainder\\nof the judgment, and remanded for further proceedings.\\n\\n\\n COUNSEL\\n\\nMark D. Herman (argued), Covington & Burling LLP,\\nWashington, D.C., for Plaintiffs-Appellants.\\n\\nGregory Scott Worden (argued) and Donna M. Chamberlin,\\nLewis Brisbois Bisgaard & Smith LLP, Seattle, Washington,\\nfor Defendant-Appellee.\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f ANDERSON V. STATE FARM 3\\n\\n OPINION\\n\\nMcKEOWN, Circuit Judge:\\n\\n Patricia Anderson and Thomas Anderson’s single-car\\naccident in 1998 began a saga of litigation culminating in\\nthis appeal. Little of the tragic history involving nine\\ndifferent trial and appellate proceedings is relevant to this\\nopinion, in which we determine only whether State Farm\\ntimely removed the case to federal court. 1 Addressing an\\nissue of first impression in this circuit, we join the Fourth\\nCircuit in holding that receipt of an initial pleading by a\\nstatutorily designated agent does not begin the thirty-day\\nremoval clock under 28 U.S.C. § 1446(b)(1), and that it was\\ninstead actual receipt by State Farm that started the removal\\nclock. See Elliott v. Am. States Ins. Co., 883 F.3d 384, 394\\n(4th Cir. 2018). Applying this rule, we conclude that State\\nFarm timely removed the case.\\n\\n The Andersons sued State Farm in Washington state\\ncourt. Because State Farm is an out-of-state (or “foreign”)\\ninsurer, state law designates Washington’s Insurance\\nCommissioner as State Farm’s statutory agent.\\nRCW 48.05.200(1). To serve legal process on State Farm,\\nthe Andersons served the Commissioner, who forwarded the\\ncomplaint to State Farm’s designated recipient.\\nRCW 4.28.080(7)(a), 48.05.200(1)–(2). The Andersons\\nserved the Commissioner on February 9, 2015, and State\\nFarm received the forwarded complaint four days later, on\\nFebruary 13. On March 16, State Farm removed the case to\\nfederal court.\\n\\n\\n 1\\n We address the remaining issues in a memorandum disposition\\nfiled contemporaneously with this opinion.\\n\\f4 ANDERSON V. STATE FARM\\n\\n The parties agree on how to count to thirty, but they\\ndisagree over when to begin counting. According to the\\nAndersons, removal was untimely because the thirty-day\\nremoval clock began on February 9, when the Commissioner\\nreceived the complaint. State Farm contends removal was\\ntimely as calculated from February 13, when the forwarded\\ncopy of the complaint reached State Farm’s designated\\nrecipient. 2 The district court agreed with State Farm and\\ndenied the Andersons’ motion to remand. We have\\njurisdiction under 28 U.S.C. § 1291 and review de novo the\\ndistrict court’s ruling. Patterson v. Int’l Bhd. of Teamsters,\\nLocal 959, 121 F.3d 1345, 1348 (9th Cir. 1997).\\n\\n We begin our analysis with the statutory text, but it does\\nnot take us far. A notice of removal must be filed “within\\n30 days after the receipt by the defendant, through service or\\notherwise, of a copy of the initial pleading.” 28 U.S.C.\\n§ 1446(b)(1). One reading suggests removal was timely\\nbecause the actual defendant—State Farm—did not receive\\nnotice until receiving the mailing from the Commissioner.\\nThat reading, however, does not account for the effect of\\nservice on a statutory agent. “[T]he statute itself says\\nnothing about service on a statutory agent,” Elliott, 883 F.3d\\nat 392, and an agent designated by the state legislature to\\nreceive service fundamentally differs from a defendant’s\\nagent-in-fact, because the defendant has no meaningful say\\nin or control over the former. See 14C Charles Alan Wright,\\net al., Federal Practice & Procedure § 3731 (Rev. 4th ed.\\n2018) (“Realistically speaking, of course, statutory agents\\n\\n\\n\\n 2\\n Thirty days from February 13, 2015, was Sunday, March 15, 2015,\\nso the notice of removal would have been timely filed on Monday,\\nMarch 16, 2015. See Fed. R. Civ. P. 6(a)(1)(C).\\n\\f ANDERSON V. STATE FARM 5\\n\\nare not true agents but merely are a medium for transmitting\\nthe relevant papers.”).\\n\\n Washington law, however, lends some force to the\\nposition that service on a statutorily designated agent should\\nqualify as “receipt by the defendant” for purposes of\\n28 U.S.C. § 1446(b)(1). As a practical matter, a company\\ncannot receive anything except through its agents, and\\nWashington equates service upon the Commissioner with\\nservice upon the insurer. RCW 48.05.200(1). 3 Indeed,\\nplaintiffs must serve foreign insurers via the Commissioner.\\nId.; RCW 4.28.080(7)(a). But the Supreme Court has\\ncautioned against interpreting the federal removal statute\\nwith reference to state law. See Chi., R.I. & P.R. Co. v.\\nStude, 346 U.S. 574, 580 (1954) (“For the purpose of\\nremoval, the federal law determines who is plaintiff and who\\nis defendant. It is a question of the construction of the\\nfederal statute on removal, and not the state statute.”). So\\nwe decline to place much weight on Washington law and\\nconclude that the federal removal statute is, at most,\\nambiguous on this point. Cf. Elliott, 883 F.3d at 392 (“[A]\\nstraightforward reading . . . supports holding that the 30-day\\nperiod . . . is not triggered by service on a statutory agent\\n. . . .”).\\n\\n\\n\\n 3\\n RCW 48.05.200(1) provides: “Each authorized foreign or alien\\ninsurer must appoint the commissioner as its attorney to receive service\\nof, and upon whom must be served, all legal process issued against it in\\nthis state upon causes of action arising within this state. Service upon\\nthe commissioner as attorney constitutes service upon the insurer.\\nService of legal process against the insurer can be had only by service\\nupon the commissioner, except actions upon contractor bonds pursuant\\nto RCW 18.27.040, where service may be upon the department of labor\\nand industries.”\\n\\f6 ANDERSON V. STATE FARM\\n\\n The legislative history, on the other hand, is\\nunambiguous—it clearly demonstrates Congress’ intent to\\navoid disparate application of the removal statute due to\\ndifferences in state law. See Murphy Bros., Inc. v. Michetti\\nPipe Stringing, Inc., 526 U.S. 344, 351–53 (1999)\\n(discussing congressional intent at length). Before Congress\\nenacted § 1446(b), a defendant could remove at “any time\\nbefore the defendant [wa]s required by the laws of the State\\n. . . to answer or plead to the declaration or complaint.”\\n28 U.S.C. § 72 (1940). The practical effects of this regime\\nvaried widely among the states. Seeking a replacement that\\nwould “give adequate time and operate uniformly\\nthroughout the Federal jurisdiction,” Congress enacted\\n§ 1446(b). H.R. Rep. No. 80-308, at A135 (1947).\\n\\n As initially enacted, however, § 1446(b) suffered the\\nsame infirmity: the removal deadline depended on\\nprocedural rules that varied by state. See Murphy Bros.,\\n526 U.S. at 351. In other words, the removal clock began\\n“after commencement of the action or service of process,\\nwhichever [wa]s later.” Act of June 25, 1948, Pub. L. No.\\n80-773, § 1446(b), 62 Stat. 869, 939. In states like New\\nYork, which allowed a plaintiff to commence an action by\\nserving the summons without the complaint, the removal\\ndeadline could expire before the defendant could access the\\ncomplaint. H.R. Rep. No. 81-352, at 14 (1949); Murphy\\nBros., 526 U.S. at 351.\\n\\n Recognizing this disparity, the next Congress amended\\nthe statute to provide each defendant with “20 days after he\\nhas received (or it has been made available to him) a copy of\\nthe initial pleading.” S. Rep. No. 81-303, at 6 (1949). In all\\nrelevant respects, the 1949 version of § 1446 mirrors today’s\\nversion: “The petition for removal of a civil action or\\nproceeding shall be filed within twenty days after the receipt\\n\\f ANDERSON V. STATE FARM 7\\n\\nby the defendant, through service or otherwise, of a copy of\\nthe initial pleading . . . .” 4 Act of May 24, 1949, Pub. L. No.\\n81-72, § 83, 63 Stat. 89, 101. In both versions of the statute,\\nthe key point was to peg the time calculation to receipt by\\nthe defendant.\\n\\n Intertwining the removal statute with state-specific\\nidiosyncrasies would thwart Congress’ twin aims of\\nadequate time to respond and uniform application of the rule.\\nMurphy Bros., 526 U.S. at 351. Indeed, if delivery to a\\nstatutorily designated agent began the removal clock, the\\neffective time a defendant had to remove would depend not\\nonly on differences in state law, but also on the efficiency of\\nstate agencies in each instance. The interpretation the\\nAndersons advance would compound disfavored interstate\\ndifferences by adding case-by-case, intrastate discrepancies.\\nThat arrangement cannot be reconciled with Congress’\\nunambiguous intent to provide each defendant with a fixed\\nand adequate amount of time, after obtaining access to the\\ncomplaint, to decide whether to remove. See Elliott,\\n883 F.3d at 393 (“Serving a statutory agent does not\\nguarantee that the defendant is provided with actual notice\\nof the complaint or adequate time to decide whether to\\nremove a case.”).\\n\\n As the Andersons point out, the Court held in Murphy\\nBrothers that “a named defendant’s time to remove is\\ntriggered by simultaneous service of the summons and\\ncomplaint.” 526 U.S. at 347–48. But this holding addressed\\nwhat must be received, not who must receive it. Id. The\\nCourt did not consider the implications of serving a statutory\\n\\n 4\\n Congress later extended the removal window from twenty days to\\nthirty days. Act of Sept. 29, 1965, Pub. L. No. 89-215, 79 Stat. 887, 887;\\n28 U.S.C. § 1446(b)(1).\\n\\f8 ANDERSON V. STATE FARM\\n\\nagent before actual receipt by the defendant. And although\\nit addressed a different question, the Court announced a\\nprinciple of fairness that is pertinent here:\\n\\n We read Congress’ provisions for removal in\\n light of a bedrock principle: An individual or\\n entity named as a defendant is not obliged to\\n engage in litigation unless notified of the\\n action, and brought under a court’s authority,\\n by formal process.\\n\\nId. at 347. That “bedrock principle” regarding the\\nimportance of actual notice confirms that serving the\\nCommissioner did not provide the necessary notice to State\\nFarm of the suit—that occurred only when State Farm’s\\ndesignated recipient received the complaint.\\n\\n We are also unconvinced by the Andersons’ invocation\\nof the interpretive canon counseling strict construction of\\nremoval statutes. See Gaus v. Miles, Inc., 980 F.2d 564, 566\\n(9th Cir. 1992). The canon is instructive, but it is not\\nabsolute and cannot overcome a “clear [contrary] command\\nfrom both Congress and the Supreme Court.” Durham v.\\nLockheed Martin Corp., 445 F.3d 1247, 1252 (9th Cir.\\n2006). The legislative history and Murphy Brothers provide\\njust such a clear command.\\n\\n Nor are we persuaded by the Andersons’ suggestion that\\nour ruling invites collateral litigation over “who qualifies” as\\na recipient. Our decision cleaves to “settled [district court]\\nlaw,” 14C Charles Alan Wright, et al., Federal Practice &\\nProcedure § 3731 & n.20 (4th ed.), and the current regime\\nhas hardly spawned uncertainty on this point.\\n\\n We join the Fourth Circuit and hold that the thirty-day\\nremoval clock under 28 U.S.C. § 1446(b)(1) does not begin\\n\\f ANDERSON V. STATE FARM 9\\n\\nupon service on and receipt by a statutorily designated agent,\\nand began in this case only when State Farm actually\\nreceived the Andersons’ complaint. 5 Elliott, 883 F.3d\\nat 391–94. State Farm’s removal was timely, and we affirm\\nthe district court’s denial of the Andersons’ motion to\\nremand. For reasons explained in our contemporaneously\\nfiled memorandum disposition, we affirm in part and vacate\\nin part the remainder of the judgment and remand for further\\nproceedings.\\n\\n AFFIRMED WITH RESPECT TO THE DENIAL\\nOF THE MOTION TO REMAND.\\n\\n Each party shall bear its own costs on appeal.\\n\\n\\n\\n\\n 5\\n The Washington law at issue here, RCW 48.05.200(1), and the\\nNorth Carolina statute at issue in Elliott, 883 F.3d at 390–91 (addressing\\nN.C. Gen. Stat. § 58-16-5), designate a statutory agent that foreign\\ninsurers must authorize to accept for service of process. Other states,\\nsuch as California, require foreign insurers to designate an agent, but do\\nnot designate who that agent is. See Cal. Ins. Code § 1600 (“[E]very\\nforeign insurer [must designate] an agent for service of process. The\\nagent designated may be any person residing in this state, including, but\\nnot limited to, any corporate officer of the insurer.”). Such a requirement\\nis not before us, and we take no position regarding its interplay with the\\nfederal removal statute.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1277,"cells":{"id":{"kind":"number","value":4598508,"string":"4,598,508"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Ipc (Usa), Inc. v. Kathryn Ellis"},"decision_date":{"kind":"string","value":"2019-03-11"},"docket_number":{"kind":"string","value":"17-60081"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: William A. Fletcher and Jay S. Bybee, Circuit Judges, and Larry A. Burns, * Chief District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"BURNS, Chief District Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nIN RE PETTIT OIL COMPANY, No. 17-60081\\n Debtor,\\n BAP No.\\n WW-16-1424-\\nIPC (USA), INC., a California KuFB\\ncorporation,\\n Appellant,\\n OPINION\\n v.\\n\\nKATHRYN A. ELLIS, Chapter 7\\nTrustee,\\n Appellee.\\n\\n\\n\\nAppeal from the Ninth Circuit Bankruptcy Appellate Panel\\n Kurtz, Faris, and Brand, Bankruptcy Judges\\n\\n Argued and Submitted December 6, 2018\\n Seattle, Washington\\n\\n Filed March 11, 2019\\n\\f2 IN RE PETTIT OIL CO.\\n\\n Before: William A. Fletcher and Jay S. Bybee, Circuit\\n Judges, and Larry A. Burns, * Chief District Judge.\\n\\n Opinion by Judge Burns\\n\\n\\n SUMMARY **\\n\\n\\n Bankruptcy\\n\\n The panel affirmed the Bankruptcy Appellate Panel’s\\naffirmance of the bankruptcy court’s summary judgment in\\nfavor of a bankruptcy trustee who brought an adversary\\nproceeding seeking avoidance of transfers.\\n\\n The debtor, a distributor of bulk petroleum products,\\nentered into a consignment agreement with IPC (USA), Inc.\\nUnder the agreement, IPC delivered fuel to “card lock” sites\\nfrom which the debtor’s commercial customers purchased\\nfuel using access cards. When the debtor filed for\\nbankruptcy, it had in its possession IPC fuel as well as\\nproceeds from sold fuel, in the form of cash and accounts\\nreceivable, that had not yet been remitted to IPC.\\n\\n U.C.C. § 9-319(a) grants a consignee “rights and title to\\nthe goods.” If a consignee files for bankruptcy, any\\nconsigned “goods” in its possession become property of the\\nbankruptcy estate unless the seller has previously provided\\n\\n *\\n The Honorable Larry Alan Burns, Chief United States District\\nJudge for the Southern District of California, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f IN RE PETTIT OIL CO. 3\\n\\npublic notice of its interest in the goods (normally by filing\\na document known as a “financing statement”) and thereby\\n“perfected” its interest. The panel held that this rule also\\nextends to the proceeds from goods sold that are held by the\\nconsignee on the date it files for bankruptcy. Thus, IPC’s\\nunperfected security interest in the fuel and the proceeds was\\nsubordinate to the trustee’s interest.\\n\\n\\n COUNSEL\\n\\nEdwin K. Sato (argued), Bucknell Stehlik Sato & Orth LLP,\\nSeattle, Washington, for Appellant.\\n\\nAndrew H. Morton (argued), Deborah A. Crabbe, Foster\\nPepper PLLC, Seattle, Washington, for Appellee.\\n\\n\\n OPINION\\n\\nBURNS, Chief District Judge:\\n\\n In a consignment transaction, a seller (the “consignor”)\\ndelivers goods to a middleman (the “consignee”) who holds\\nthe goods until they are sold to a buyer, at which point the\\nsale proceeds are transferred back to the seller. Under settled\\nbankruptcy law, if a consignee files for bankruptcy, any\\nconsigned “goods” in its possession become property of the\\nbankruptcy estate unless the seller has previously provided\\npublic notice of its interest in the goods (normally by filing\\na document known as a “financing statement”) and thereby\\n“perfected” its interest. At issue in this case is whether this\\nrule also extends to the proceeds from goods sold that are\\nheld by the consignee on the date it files for bankruptcy. We\\n\\f4 IN RE PETTIT OIL CO.\\n\\nconclude that it does, and affirm the judgment of the\\nBankruptcy Court and the Bankruptcy Appellate Panel.\\n\\n I.\\n\\n The Debtor here, Pettit Oil Company, was a distributor\\nof bulk petroleum products. Part of Pettit’s business\\ninvolved operating “card lock” sites, where commercial\\ncustomers purchased fuel products using access cards. In\\n2013, Pettit entered into a consignment agreement with IPC\\n(USA), Inc. (“IPC”), under which IPC was to deliver\\nconsigned fuel to card lock sites so Pettit could sell the fuel\\nto its customers. The aim was to reduce Pettit’s working\\ncapital needs by outsourcing its fuel sales to IPC. In return\\nfor being able to sell its fuel at Pettit’s stations, IPC paid\\nPettit a monthly commission.\\n\\n As with all “true” consignments, ownership of the fuel\\nremained with IPC until it was sold, at which time title\\ntransferred to the purchaser. Whenever a customer\\npurchased consigned fuel, Pettit prepared an invoice and\\ninstructed the customer to remit payment to IPC directly.\\nDespite this instruction, some customers continued to pay\\nPettit for their purchases of IPC fuel. Anticipating this might\\noccur, the agreement provided that Pettit would “promptly\\nforward such payment[s] to IPC,” and Pettit did so regularly.\\nNonetheless, when Pettit ultimately filed for bankruptcy, it\\nhad in its possession not just IPC fuel but also proceeds from\\nsold fuel that had not yet been remitted to IPC. These\\nproceeds took two forms: (1) cash and (2) accounts\\nreceivable—that is, balances owed by customers that had not\\nyet been paid. It is undisputed that IPC never filed a\\nfinancing statement or otherwise perfected its interests in the\\nconsigned fuel, the accounts receivable, or the cash.\\n\\f IN RE PETTIT OIL CO. 5\\n\\n After Pettit filed for bankruptcy, the Trustee commenced\\nthis proceeding seeking, among other things, the value of the\\nfuel, accounts receivable, and cash proceeds for the benefit\\nof the bankruptcy estate. The Trustee maintained that IPC’s\\ninterest in the fuel, the cash proceeds, and accounts\\nreceivable was subordinate to the Trustee’s because IPC\\nhadn’t filed a financing statement or otherwise perfected its\\ninterest. The Bankruptcy Court entered summary judgment\\nin the Trustee’s favor, and the Bankruptcy Appellate Panel\\naffirmed.\\n\\n II.\\n\\n We independently review decisions of the Bankruptcy\\nCourt and the Bankruptcy Appellate Panel. Carrillo v. Su\\n(In re Su), 290 F.3d 1140, 1142 (9th Cir. 2002). Conclusions\\nof law are reviewed de novo, while factual findings are\\nreviewed for clear error. Id.\\n\\n III.\\n\\n IPC’s principal argument is that the bankruptcy courts\\nerred in concluding that the Trustee’s interest in the cash and\\naccounts receivable was superior to IPC’s. In IPC’s view,\\nalthough the Trustee may have a superior interest in the\\n“goods”—i.e., the fuel—that interest does not extend to cash\\nproceeds and accounts receivable that happened to be in\\nPettit’s possession when it filed for bankruptcy. IPC’s\\nargument presents a single question of law: whether U.C.C.\\n§ 9-319(a), which grants a consignee “rights and title to the\\ngoods,” also grants the consignee an interest in the proceeds\\nof those goods that were generated prior to bankruptcy. We\\nhold that it does.\\n\\f6 IN RE PETTIT OIL CO.\\n\\n A.\\n\\n When a debtor goes into bankruptcy, the bankruptcy\\ntrustee is automatically granted a judicial lien over all\\nproperty the debtor owns as of the petition date. See 11\\nU.S.C. § 544(a)(1). A creditor wishing to shield a particular\\nasset from the reach of the trustee can do so only if the\\ncreditor can show that its interest in the asset is superior to a\\njudicial lien, a determination governed by various statutory\\npriority rules. Otherwise, the trustee’s judicial lien remains\\nsuperior and the trustee can “avoid” (i.e., block) any\\ntransfers of the asset outside the bankruptcy estate.\\n\\n IPC maintains that we should apply traditional property\\nlaw principles to hold that the proceeds in Pettit’s possession\\nare outside the scope of the Trustee’s avoidance powers\\nbecause the proceeds were not owned by Pettit. More\\nspecifically, IPC maintains that we should treat the proceeds\\nas if they were the product of a bailment—that is, a transfer\\nof possession without a transfer of ownership. See, e.g., 5\\nCollier on Bankruptcy ¶ 541.05[1][b] (16th ed.) (2018) (“In\\nordinary commercial practice, a consignment is equivalent\\nto a bailment for care or sale, wherein there is no obligation\\nto purchase in the consignee.”). Although this logic would\\nalso suggest that the goods themselves are outside the\\nTrustee’s avoidance powers, IPC concedes (as it must) that\\nArticle 9 of the Uniform Commercial Code (“U.C.C.”)\\nforecloses this argument. Section 9-319(a) of the U.C.C.\\nprovides, “for purposes of determining the rights of creditors\\nof . . . a consignee, while the goods are in the possession of\\nthe consignee, the consignee is deemed to have rights and\\ntitle to the goods identical to those the consignor had.”\\n(emphasis added). This provision means that even though a\\nconsignee doesn’t truly own the consigned goods, the U.C.C.\\ntreats the consignee as having an ownership interest. So,\\n\\f IN RE PETTIT OIL CO. 7\\n\\nhere, with Pettit’s ownership interest established, the parties\\nagree that IPC’s unperfected security interest in the fuel is\\nsubordinate to the Trustee’s judicial lien. See In re First T.D.\\n& Inv., Inc., 253 F.3d 520, 525 (9th Cir. 2001) (“Under 11\\nU.S.C. § 544(a), unperfected security interests are avoidable\\nand can be relegated to the status of general unsecured\\nclaims.”).\\n\\n Nonetheless, IPC argues that even if its interest in the\\nfuel is subordinate to that of the Trustee, its interest in the\\ncash and accounts receivable is superior because these assets\\naren’t “goods.” In other words, according to IPC, because\\nthe drafters of U.C.C. § 9-319(a) said “goods” instead of\\n“goods and proceeds,” Article 9—which governs priority\\nand perfection rules related to various types of security\\ninterests, including consignments—cannot dictate the rules\\nregarding the proceeds of the goods. The problem with this\\nstrained reading of section 9-319 is that it ignores numerous\\nreferences throughout the U.C.C. that treat a consignment as\\na security interest for all practical purposes. See, e.g., U.C.C.\\n§ 9-102(a)(73)(C) (defining “[s]ecured party” to include a\\n“consignor”); U.C.C. § 1-201(b)(35) (defining “[s]ecurity\\ninterest” to include “any interest of a consignor”) (emphasis\\nadded); U.C.C. § 9-102(a)(12)(C) (defining “[c]ollateral” to\\nmean “the property subject to a security interest” that\\nincludes “goods that are the subject of a consignment”). The\\nmost natural reading of these provisions is that a consignor’s\\ninterest in goods (and the related proceeds) is a security\\ninterest for all purposes—including for purposes of\\nperfection and priority—unless the U.C.C. specifically says\\notherwise.\\n\\n The best example of the inconsistency in IPC’s argument\\nis U.C.C. § 9-324(b), which states, “a perfected [interest] in\\ninventory has priority over a conflicting security interest in\\n\\f8 IN RE PETTIT OIL CO.\\n\\nthe same inventory . . . and . . . also has priority in\\nidentifiable cash proceeds of the inventory.” (emphasis\\nadded). Had IPC availed itself of this protection by\\nperfecting its interest in the fuel, it would have had priority\\nover the Trustee’s judicial lien extending not just to the fuel,\\nbut also to the proceeds. And there is no persuasive reason\\nto interpret Article 9 as limiting the reciprocal effect—that a\\nconsignor loses priority in the proceeds when it fails to\\nperfect its interest.\\n\\n Moreover, the “goods” provision of section 9-319 can’t\\nbe read in a vacuum. “We must interpret the statute as a\\nwhole, giving effect to each word and making every effort\\nnot to interpret a provision in a manner that renders other\\nprovisions of the same statute inconsistent, meaningless or\\nsuperfluous.” United States v. Neal, 776 F.3d 645, 652 (9th\\nCir. 2015) (internal alterations and quotations omitted).\\nConsistent with this rule of construction, and read in light of\\nArticle 9’s other provisions, we hold that the term “goods”\\nin section 9-319(a) includes the proceeds of those goods, and\\nthat Article 9’s priority and perfection rules apply with equal\\nforce to such proceeds.\\n\\n Although IPC argues the result should be different\\nbecause it retained title to the proceeds, the U.C.C. is clear\\nthat IPC’s retention of title does not matter. Section 9-202\\nof the U.C.C. states that “[e]xcept as otherwise provided\\nwith respect to consignments . . . , the provisions of [Article\\n9] with regard to rights and obligations apply whether title\\nto collateral is in the secured party or the debtor.” Retention\\nof title affects the remedies IPC could employ to recover the\\ngoods in the event of default, but title is irrelevant to whether\\nIPC or the Trustee has priority in the goods and proceeds.\\nSee U.C.C. § 9-202, cmt. 3.a.\\n\\f IN RE PETTIT OIL CO. 9\\n\\n Our conclusion that the term “goods” in section 9-319\\nincludes the proceeds of those goods is bolstered by the\\npolicy rationale underlying these rules. To the outside\\nworld, goods and proceeds held by a consignee appear to be\\nowned by the consignee, and creditors might reasonably\\nbelieve as much when they decide to lend the consignee\\nmoney. The perfection and priority rules—which require\\nthat the consignor publicly announce its interest in the\\nconsigned goods or else go to the back of the line when the\\nconsignee goes bankrupt—serve to protect unwary creditors\\nand prevent “secret liens” in the goods that might otherwise\\ndissuade such lending. See In re Valley Media, Inc., 279\\nB.R. 105, 125 (D. Del. Bankr. 2002) (“The purpose of . . . 9-\\n319(a) is to protect general creditors of the consignee from\\nclaims of consignors that have undisclosed consignment\\narrangements with the consignee that create secret liens on\\nthe inventory.”). A ruling that “proceeds” are outside the\\nscope of the perfection rules would disrupt the delicate\\nbalance the U.C.C. drafters struck between the interests of\\nconsignors and the interests of the consignee’s other\\ncreditors. IPC has not provided a convincing basis for\\ndisrupting this intended balance.\\n\\n B.\\n\\n We also reject IPC’s argument that the Trustee can’t\\nhave an interest in the proceeds because 11 U.S.C.\\n§ 544(a)—which establishes the Trustee as a judicial lien\\nholder—doesn’t contain a “reachback” provision that would\\nallow it to claim an interest in cash and accounts receivable\\nthat arose before Pettit filed its bankruptcy petition. Section\\n544 grants the Trustee “a judicial lien on all property on\\nwhich a creditor on a simple contract could have obtained\\nsuch a judicial lien.” 11 U.S.C. § 544(a)(1) (emphasis\\nadded). However we characterize the Trustee’s interest in\\n\\f10 IN RE PETTIT OIL CO.\\n\\nthe “property” (i.e., the proceeds), that interest is identical to\\nthe interest held by IPC. See U.C.C. § 9-319(a) (“[T]he\\nconsignee is deemed to have rights and title to the goods\\nidentical to those the consignor had.”). IPC undoubtedly\\ncould have secured a lien on the proceeds, so the Trustee\\ncould have as well. Accordingly, the lack of a “reachback”\\nprovision is no barrier to the Trustee claiming an interest in\\nproceeds that arose pre-petition.\\n\\n AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4375761/\", \"author_raw\": \"BURNS, Chief District Judge:\"}]}"},"judge1":{"kind":"string","value":"WILLIAM A FLETCHER"},"judge2":{"kind":"string","value":"JAY S BYBEE"},"judge3":{"kind":"string","value":"LARRY A BURNS"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4598508/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Ipc (Usa), Inc. v. Kathryn Ellis\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-11\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nIN RE PETTIT OIL COMPANY, No. 17-60081\n Debtor,\n BAP No.\n WW-16-1424-\nIPC (USA), INC., a California KuFB\ncorporation,\n Appellant,\n OPINION\n v.\n\nKATHRYN A. ELLIS, Chapter 7\nTrustee,\n Appellee.\n\n\n\nAppeal from the Ninth Circuit Bankruptcy Appellate Panel\n Kurtz, Faris, and Brand, Bankruptcy Judges\n\n Argued and Submitted December 6, 2018\n Seattle, Washington\n\n Filed March 11, 2019\n\f2 IN RE PETTIT OIL CO.\n\n Before: William A. Fletcher and Jay S. Bybee, Circuit\n Judges, and Larry A. Burns, * Chief District Judge.\n\n Opinion by Judge Burns\n\n\n SUMMARY **\n\n\n Bankruptcy\n\n The panel affirmed the Bankruptcy Appellate Panel’s\naffirmance of the bankruptcy court’s summary judgment in\nfavor of a bankruptcy trustee who brought an adversary\nproceeding seeking avoidance of transfers.\n\n The debtor, a distributor of bulk petroleum products,\nentered into a consignment agreement with IPC (USA), Inc.\nUnder the agreement, IPC delivered fuel to “card lock” sites\nfrom which the debtor’s commercial customers purchased\nfuel using access cards. When the debtor filed for\nbankruptcy, it had in its possession IPC fuel as well as\nproceeds from sold fuel, in the form of cash and accounts\nreceivable, that had not yet been remitted to IPC.\n\n U.C.C. § 9-319(a) grants a consignee “rights and title to\nthe goods.” If a consignee files for bankruptcy, any\nconsigned “goods” in its possession become property of the\nbankruptcy estate unless the seller has previously provided\n\n *\n The Honorable Larry Alan Burns, Chief United States District\nJudge for the Southern District of California, sitting by designation.\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f IN RE PETTIT OIL CO. 3\n\npublic notice of its interest in the goods (normally by filing\na document known as a “financing statement”) and thereby\n“perfected” its interest. The panel held that this rule also\nextends to the proceeds from goods sold that are held by the\nconsignee on the date it files for bankruptcy. Thus, IPC’s\nunperfected security interest in the fuel and the proceeds was\nsubordinate to the trustee’s interest.\n\n\n COUNSEL\n\nEdwin K. Sato (argued), Bucknell Stehlik Sato & Orth LLP,\nSeattle, Washington, for Appellant.\n\nAndrew H. Morton (argued), Deborah A. Crabbe, Foster\nPepper PLLC, Seattle, Washington, for Appellee.\n\n\n OPINION\n\nBURNS, Chief District Judge:\n\n In a consignment transaction, a seller (the “consignor”)\ndelivers goods to a middleman (the “consignee”) who holds\nthe goods until they are sold to a buyer, at which point the\nsale proceeds are transferred back to the seller. Under settled\nbankruptcy law, if a consignee files for bankruptcy, any\nconsigned “goods” in its possession become property of the\nbankruptcy estate unless the seller has previously provided\npublic notice of its interest in the goods (normally by filing\na document known as a “financing statement”) and thereby\n“perfected” its interest. At issue in this case is whether this\nrule also extends to the proceeds from goods sold that are\nheld by the consignee on the date it files for bankruptcy. We\n\f4 IN RE PETTIT OIL CO.\n\nconclude that it does, and affirm the judgment of the\nBankruptcy Court and the Bankruptcy Appellate Panel.\n\n I.\n\n The Debtor here, Pettit Oil Company, was a distributor\nof bulk petroleum products. Part of Pettit’s business\ninvolved operating “card lock” sites, where commercial\ncustomers purchased fuel products using access cards. In\n2013, Pettit entered into a consignment agreement with IPC\n(USA), Inc. (“IPC”), under which IPC was to deliver\nconsigned fuel to card lock sites so Pettit could sell the fuel\nto its customers. The aim was to reduce Pettit’s working\ncapital needs by outsourcing its fuel sales to IPC. In return\nfor being able to sell its fuel at Pettit’s stations, IPC paid\nPettit a monthly commission.\n\n As with all “true” consignments, ownership of the fuel\nremained with IPC until it was sold, at which time title\ntransferred to the purchaser. Whenever a customer\npurchased consigned fuel, Pettit prepared an invoice and\ninstructed the customer to remit payment to IPC directly.\nDespite this instruction, some customers continued to pay\nPettit for their purchases of IPC fuel. Anticipating this might\noccur, the agreement provided that Pettit would “promptly\nforward such payment[s] to IPC,” and Pettit did so regularly.\nNonetheless, when Pettit ultimately filed for bankruptcy, it\nhad in its possession not just IPC fuel but also proceeds from\nsold fuel that had not yet been remitted to IPC. These\nproceeds took two forms: (1) cash and (2) accounts\nreceivable—that is, balances owed by customers that had not\nyet been paid. It is undisputed that IPC never filed a\nfinancing statement or otherwise perfected its interests in the\nconsigned fuel, the accounts receivable, or the cash.\n\f IN RE PETTIT OIL CO. 5\n\n After Pettit filed for bankruptcy, the Trustee commenced\nthis proceeding seeking, among other things, the value of the\nfuel, accounts receivable, and cash proceeds for the benefit\nof the bankruptcy estate. The Trustee maintained that IPC’s\ninterest in the fuel, the cash proceeds, and accounts\nreceivable was subordinate to the Trustee’s because IPC\nhadn’t filed a financing statement or otherwise perfected its\ninterest. The Bankruptcy Court entered summary judgment\nin the Trustee’s favor, and the Bankruptcy Appellate Panel\naffirmed.\n\n II.\n\n We independently review decisions of the Bankruptcy\nCourt and the Bankruptcy Appellate Panel. Carrillo v. Su\n(In re Su), 290 F.3d 1140, 1142 (9th Cir. 2002). Conclusions\nof law are reviewed de novo, while factual findings are\nreviewed for clear error. Id.\n\n III.\n\n IPC’s principal argument is that the bankruptcy courts\nerred in concluding that the Trustee’s interest in the cash and\naccounts receivable was superior to IPC’s. In IPC’s view,\nalthough the Trustee may have a superior interest in the\n“goods”—i.e., the fuel—that interest does not extend to cash\nproceeds and accounts receivable that happened to be in\nPettit’s possession when it filed for bankruptcy. IPC’s\nargument presents a single question of law: whether U.C.C.\n§ 9-319(a), which grants a consignee “rights and title to the\ngoods,” also grants the consignee an interest in the proceeds\nof those goods that were generated prior to bankruptcy. We\nhold that it does.\n\f6 IN RE PETTIT OIL CO.\n\n A.\n\n When a debtor goes into bankruptcy, the bankruptcy\ntrustee is automatically granted a judicial lien over all\nproperty the debtor owns as of the petition date. See 11\nU.S.C. § 544(a)(1). A creditor wishing to shield a particular\nasset from the reach of the trustee can do so only if the\ncreditor can show that its interest in the asset is superior to a\njudicial lien, a determination governed by various statutory\npriority rules. Otherwise, the trustee’s judicial lien remains\nsuperior and the trustee can “avoid” (i.e., block) any\ntransfers of the asset outside the bankruptcy estate.\n\n IPC maintains that we should apply traditional property\nlaw principles to hold that the proceeds in Pettit’s possession\nare outside the scope of the Trustee’s avoidance powers\nbecause the proceeds were not owned by Pettit. More\nspecifically, IPC maintains that we should treat the proceeds\nas if they were the product of a bailment—that is, a transfer\nof possession without a transfer of ownership. See, e.g., 5\nCollier on Bankruptcy ¶ 541.05[1][b] (16th ed.) (2018) (“In\nordinary commercial practice, a consignment is equivalent\nto a bailment for care or sale, wherein there is no obligation\nto purchase in the consignee.”). Although this logic would\nalso suggest that the goods themselves are outside the\nTrustee’s avoidance powers, IPC concedes (as it must) that\nArticle 9 of the Uniform Commercial Code (“U.C.C.”)\nforecloses this argument. Section 9-319(a) of the U.C.C.\nprovides, “for purposes of determining the rights of creditors\nof . . . a consignee, while the goods are in the possession of\nthe consignee, the consignee is deemed to have rights and\ntitle to the goods identical to those the consignor had.”\n(emphasis added). This provision means that even though a\nconsignee doesn’t truly own the consigned goods, the U.C.C.\ntreats the consignee as having an ownership interest. So,\n\f IN RE PETTIT OIL CO. 7\n\nhere, with Pettit’s ownership interest established, the parties\nagree that IPC’s unperfected security interest in the fuel is\nsubordinate to the Trustee’s judicial lien. See In re First T.D.\n& Inv., Inc., 253 F.3d 520, 525 (9th Cir. 2001) (“Under 11\nU.S.C. § 544(a), unperfected security interests are avoidable\nand can be relegated to the status of general unsecured\nclaims.”).\n\n Nonetheless, IPC argues that even if its interest in the\nfuel is subordinate to that of the Trustee, its interest in the\ncash and accounts receivable is superior because these assets\naren’t “goods.” In other words, according to IPC, because\nthe drafters of U.C.C. § 9-319(a) said “goods” instead of\n“goods and proceeds,” Article 9—which governs priority\nand perfection rules related to various types of security\ninterests, including consignments—cannot dictate the rules\nregarding the proceeds of the goods. The problem with this\nstrained reading of section 9-319 is that it ignores numerous\nreferences throughout the U.C.C. that treat a consignment as\na security interest for all practical purposes. See, e.g., U.C.C.\n§ 9-102(a)(73)(C) (defining “[s]ecured party” to include a\n“consignor”); U.C.C. § 1-201(b)(35) (defining “[s]ecurity\ninterest” to include “any interest of a consignor”) (emphasis\nadded); U.C.C. § 9-102(a)(12)(C) (defining “[c]ollateral” to\nmean “the property subject to a security interest” that\nincludes “goods that are the subject of a consignment”). The\nmost natural reading of these provisions is that a consignor’s\ninterest in goods (and the related proceeds) is a security\ninterest for all purposes—including for purposes of\nperfection and priority—unless the U.C.C. specifically says\notherwise.\n\n The best example of the inconsistency in IPC’s argument\nis U.C.C. § 9-324(b), which states, “a perfected [interest] in\ninventory has priority over a conflicting security interest in\n\f8 IN RE PETTIT OIL CO.\n\nthe same inventory . . . and . . . also has priority in\nidentifiable cash proceeds of the inventory.” (emphasis\nadded). Had IPC availed itself of this protection by\nperfecting its interest in the fuel, it would have had priority\nover the Trustee’s judicial lien extending not just to the fuel,\nbut also to the proceeds. And there is no persuasive reason\nto interpret Article 9 as limiting the reciprocal effect—that a\nconsignor loses priority in the proceeds when it fails to\nperfect its interest.\n\n Moreover, the “goods” provision of section 9-319 can’t\nbe read in a vacuum. “We must interpret the statute as a\nwhole, giving effect to each word and making every effort\nnot to interpret a provision in a manner that renders other\nprovisions of the same statute inconsistent, meaningless or\nsuperfluous.” United States v. Neal, 776 F.3d 645, 652 (9th\nCir. 2015) (internal alterations and quotations omitted).\nConsistent with this rule of construction, and read in light of\nArticle 9’s other provisions, we hold that the term “goods”\nin section 9-319(a) includes the proceeds of those goods, and\nthat Article 9’s priority and perfection rules apply with equal\nforce to such proceeds.\n\n Although IPC argues the result should be different\nbecause it retained title to the proceeds, the U.C.C. is clear\nthat IPC’s retention of title does not matter. Section 9-202\nof the U.C.C. states that “[e]xcept as otherwise provided\nwith respect to consignments . . . , the provisions of [Article\n9] with regard to rights and obligations apply whether title\nto collateral is in the secured party or the debtor.” Retention\nof title affects the remedies IPC could employ to recover the\ngoods in the event of default, but title is irrelevant to whether\nIPC or the Trustee has priority in the goods and proceeds.\nSee U.C.C. § 9-202, cmt. 3.a.\n\f IN RE PETTIT OIL CO. 9\n\n Our conclusion that the term “goods” in section 9-319\nincludes the proceeds of those goods is bolstered by the\npolicy rationale underlying these rules. To the outside\nworld, goods and proceeds held by a consignee appear to be\nowned by the consignee, and creditors might reasonably\nbelieve as much when they decide to lend the consignee\nmoney. The perfection and priority rules—which require\nthat the consignor publicly announce its interest in the\nconsigned goods or else go to the back of the line when the\nconsignee goes bankrupt—serve to protect unwary creditors\nand prevent “secret liens” in the goods that might otherwise\ndissuade such lending. See In re Valley Media, Inc., 279\nB.R. 105, 125 (D. Del. Bankr. 2002) (“The purpose of . . . 9-\n319(a) is to protect general creditors of the consignee from\nclaims of consignors that have undisclosed consignment\narrangements with the consignee that create secret liens on\nthe inventory.”). A ruling that “proceeds” are outside the\nscope of the perfection rules would disrupt the delicate\nbalance the U.C.C. drafters struck between the interests of\nconsignors and the interests of the consignee’s other\ncreditors. IPC has not provided a convincing basis for\ndisrupting this intended balance.\n\n B.\n\n We also reject IPC’s argument that the Trustee can’t\nhave an interest in the proceeds because 11 U.S.C.\n§ 544(a)—which establishes the Trustee as a judicial lien\nholder—doesn’t contain a “reachback” provision that would\nallow it to claim an interest in cash and accounts receivable\nthat arose before Pettit filed its bankruptcy petition. Section\n544 grants the Trustee “a judicial lien on all property on\nwhich a creditor on a simple contract could have obtained\nsuch a judicial lien.” 11 U.S.C. § 544(a)(1) (emphasis\nadded). However we characterize the Trustee’s interest in\n\f10 IN RE PETTIT OIL CO.\n\nthe “property” (i.e., the proceeds), that interest is identical to\nthe interest held by IPC. See U.C.C. § 9-319(a) (“[T]he\nconsignee is deemed to have rights and title to the goods\nidentical to those the consignor had.”). IPC undoubtedly\ncould have secured a lien on the proceeds, so the Trustee\ncould have as well. Accordingly, the lack of a “reachback”\nprovision is no barrier to the Trustee claiming an interest in\nproceeds that arose pre-petition.\n\n AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Ipc (Usa), Inc. v. Kathryn Ellis\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-11\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nIN RE PETTIT OIL COMPANY, No. 17-60081\\n Debtor,\\n BAP No.\\n WW-16-1424-\\nIPC (USA), INC., a California KuFB\\ncorporation,\\n Appellant,\\n OPINION\\n v.\\n\\nKATHRYN A. ELLIS, Chapter 7\\nTrustee,\\n Appellee.\\n\\n\\n\\nAppeal from the Ninth Circuit Bankruptcy Appellate Panel\\n Kurtz, Faris, and Brand, Bankruptcy Judges\\n\\n Argued and Submitted December 6, 2018\\n Seattle, Washington\\n\\n Filed March 11, 2019\\n\\f2 IN RE PETTIT OIL CO.\\n\\n Before: William A. Fletcher and Jay S. Bybee, Circuit\\n Judges, and Larry A. Burns, * Chief District Judge.\\n\\n Opinion by Judge Burns\\n\\n\\n SUMMARY **\\n\\n\\n Bankruptcy\\n\\n The panel affirmed the Bankruptcy Appellate Panel’s\\naffirmance of the bankruptcy court’s summary judgment in\\nfavor of a bankruptcy trustee who brought an adversary\\nproceeding seeking avoidance of transfers.\\n\\n The debtor, a distributor of bulk petroleum products,\\nentered into a consignment agreement with IPC (USA), Inc.\\nUnder the agreement, IPC delivered fuel to “card lock” sites\\nfrom which the debtor’s commercial customers purchased\\nfuel using access cards. When the debtor filed for\\nbankruptcy, it had in its possession IPC fuel as well as\\nproceeds from sold fuel, in the form of cash and accounts\\nreceivable, that had not yet been remitted to IPC.\\n\\n U.C.C. § 9-319(a) grants a consignee “rights and title to\\nthe goods.” If a consignee files for bankruptcy, any\\nconsigned “goods” in its possession become property of the\\nbankruptcy estate unless the seller has previously provided\\n\\n *\\n The Honorable Larry Alan Burns, Chief United States District\\nJudge for the Southern District of California, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f IN RE PETTIT OIL CO. 3\\n\\npublic notice of its interest in the goods (normally by filing\\na document known as a “financing statement”) and thereby\\n“perfected” its interest. The panel held that this rule also\\nextends to the proceeds from goods sold that are held by the\\nconsignee on the date it files for bankruptcy. Thus, IPC’s\\nunperfected security interest in the fuel and the proceeds was\\nsubordinate to the trustee’s interest.\\n\\n\\n COUNSEL\\n\\nEdwin K. Sato (argued), Bucknell Stehlik Sato & Orth LLP,\\nSeattle, Washington, for Appellant.\\n\\nAndrew H. Morton (argued), Deborah A. Crabbe, Foster\\nPepper PLLC, Seattle, Washington, for Appellee.\\n\\n\\n OPINION\\n\\nBURNS, Chief District Judge:\\n\\n In a consignment transaction, a seller (the “consignor”)\\ndelivers goods to a middleman (the “consignee”) who holds\\nthe goods until they are sold to a buyer, at which point the\\nsale proceeds are transferred back to the seller. Under settled\\nbankruptcy law, if a consignee files for bankruptcy, any\\nconsigned “goods” in its possession become property of the\\nbankruptcy estate unless the seller has previously provided\\npublic notice of its interest in the goods (normally by filing\\na document known as a “financing statement”) and thereby\\n“perfected” its interest. At issue in this case is whether this\\nrule also extends to the proceeds from goods sold that are\\nheld by the consignee on the date it files for bankruptcy. We\\n\\f4 IN RE PETTIT OIL CO.\\n\\nconclude that it does, and affirm the judgment of the\\nBankruptcy Court and the Bankruptcy Appellate Panel.\\n\\n I.\\n\\n The Debtor here, Pettit Oil Company, was a distributor\\nof bulk petroleum products. Part of Pettit’s business\\ninvolved operating “card lock” sites, where commercial\\ncustomers purchased fuel products using access cards. In\\n2013, Pettit entered into a consignment agreement with IPC\\n(USA), Inc. (“IPC”), under which IPC was to deliver\\nconsigned fuel to card lock sites so Pettit could sell the fuel\\nto its customers. The aim was to reduce Pettit’s working\\ncapital needs by outsourcing its fuel sales to IPC. In return\\nfor being able to sell its fuel at Pettit’s stations, IPC paid\\nPettit a monthly commission.\\n\\n As with all “true” consignments, ownership of the fuel\\nremained with IPC until it was sold, at which time title\\ntransferred to the purchaser. Whenever a customer\\npurchased consigned fuel, Pettit prepared an invoice and\\ninstructed the customer to remit payment to IPC directly.\\nDespite this instruction, some customers continued to pay\\nPettit for their purchases of IPC fuel. Anticipating this might\\noccur, the agreement provided that Pettit would “promptly\\nforward such payment[s] to IPC,” and Pettit did so regularly.\\nNonetheless, when Pettit ultimately filed for bankruptcy, it\\nhad in its possession not just IPC fuel but also proceeds from\\nsold fuel that had not yet been remitted to IPC. These\\nproceeds took two forms: (1) cash and (2) accounts\\nreceivable—that is, balances owed by customers that had not\\nyet been paid. It is undisputed that IPC never filed a\\nfinancing statement or otherwise perfected its interests in the\\nconsigned fuel, the accounts receivable, or the cash.\\n\\f IN RE PETTIT OIL CO. 5\\n\\n After Pettit filed for bankruptcy, the Trustee commenced\\nthis proceeding seeking, among other things, the value of the\\nfuel, accounts receivable, and cash proceeds for the benefit\\nof the bankruptcy estate. The Trustee maintained that IPC’s\\ninterest in the fuel, the cash proceeds, and accounts\\nreceivable was subordinate to the Trustee’s because IPC\\nhadn’t filed a financing statement or otherwise perfected its\\ninterest. The Bankruptcy Court entered summary judgment\\nin the Trustee’s favor, and the Bankruptcy Appellate Panel\\naffirmed.\\n\\n II.\\n\\n We independently review decisions of the Bankruptcy\\nCourt and the Bankruptcy Appellate Panel. Carrillo v. Su\\n(In re Su), 290 F.3d 1140, 1142 (9th Cir. 2002). Conclusions\\nof law are reviewed de novo, while factual findings are\\nreviewed for clear error. Id.\\n\\n III.\\n\\n IPC’s principal argument is that the bankruptcy courts\\nerred in concluding that the Trustee’s interest in the cash and\\naccounts receivable was superior to IPC’s. In IPC’s view,\\nalthough the Trustee may have a superior interest in the\\n“goods”—i.e., the fuel—that interest does not extend to cash\\nproceeds and accounts receivable that happened to be in\\nPettit’s possession when it filed for bankruptcy. IPC’s\\nargument presents a single question of law: whether U.C.C.\\n§ 9-319(a), which grants a consignee “rights and title to the\\ngoods,” also grants the consignee an interest in the proceeds\\nof those goods that were generated prior to bankruptcy. We\\nhold that it does.\\n\\f6 IN RE PETTIT OIL CO.\\n\\n A.\\n\\n When a debtor goes into bankruptcy, the bankruptcy\\ntrustee is automatically granted a judicial lien over all\\nproperty the debtor owns as of the petition date. See 11\\nU.S.C. § 544(a)(1). A creditor wishing to shield a particular\\nasset from the reach of the trustee can do so only if the\\ncreditor can show that its interest in the asset is superior to a\\njudicial lien, a determination governed by various statutory\\npriority rules. Otherwise, the trustee’s judicial lien remains\\nsuperior and the trustee can “avoid” (i.e., block) any\\ntransfers of the asset outside the bankruptcy estate.\\n\\n IPC maintains that we should apply traditional property\\nlaw principles to hold that the proceeds in Pettit’s possession\\nare outside the scope of the Trustee’s avoidance powers\\nbecause the proceeds were not owned by Pettit. More\\nspecifically, IPC maintains that we should treat the proceeds\\nas if they were the product of a bailment—that is, a transfer\\nof possession without a transfer of ownership. See, e.g., 5\\nCollier on Bankruptcy ¶ 541.05[1][b] (16th ed.) (2018) (“In\\nordinary commercial practice, a consignment is equivalent\\nto a bailment for care or sale, wherein there is no obligation\\nto purchase in the consignee.”). Although this logic would\\nalso suggest that the goods themselves are outside the\\nTrustee’s avoidance powers, IPC concedes (as it must) that\\nArticle 9 of the Uniform Commercial Code (“U.C.C.”)\\nforecloses this argument. Section 9-319(a) of the U.C.C.\\nprovides, “for purposes of determining the rights of creditors\\nof . . . a consignee, while the goods are in the possession of\\nthe consignee, the consignee is deemed to have rights and\\ntitle to the goods identical to those the consignor had.”\\n(emphasis added). This provision means that even though a\\nconsignee doesn’t truly own the consigned goods, the U.C.C.\\ntreats the consignee as having an ownership interest. So,\\n\\f IN RE PETTIT OIL CO. 7\\n\\nhere, with Pettit’s ownership interest established, the parties\\nagree that IPC’s unperfected security interest in the fuel is\\nsubordinate to the Trustee’s judicial lien. See In re First T.D.\\n& Inv., Inc., 253 F.3d 520, 525 (9th Cir. 2001) (“Under 11\\nU.S.C. § 544(a), unperfected security interests are avoidable\\nand can be relegated to the status of general unsecured\\nclaims.”).\\n\\n Nonetheless, IPC argues that even if its interest in the\\nfuel is subordinate to that of the Trustee, its interest in the\\ncash and accounts receivable is superior because these assets\\naren’t “goods.” In other words, according to IPC, because\\nthe drafters of U.C.C. § 9-319(a) said “goods” instead of\\n“goods and proceeds,” Article 9—which governs priority\\nand perfection rules related to various types of security\\ninterests, including consignments—cannot dictate the rules\\nregarding the proceeds of the goods. The problem with this\\nstrained reading of section 9-319 is that it ignores numerous\\nreferences throughout the U.C.C. that treat a consignment as\\na security interest for all practical purposes. See, e.g., U.C.C.\\n§ 9-102(a)(73)(C) (defining “[s]ecured party” to include a\\n“consignor”); U.C.C. § 1-201(b)(35) (defining “[s]ecurity\\ninterest” to include “any interest of a consignor”) (emphasis\\nadded); U.C.C. § 9-102(a)(12)(C) (defining “[c]ollateral” to\\nmean “the property subject to a security interest” that\\nincludes “goods that are the subject of a consignment”). The\\nmost natural reading of these provisions is that a consignor’s\\ninterest in goods (and the related proceeds) is a security\\ninterest for all purposes—including for purposes of\\nperfection and priority—unless the U.C.C. specifically says\\notherwise.\\n\\n The best example of the inconsistency in IPC’s argument\\nis U.C.C. § 9-324(b), which states, “a perfected [interest] in\\ninventory has priority over a conflicting security interest in\\n\\f8 IN RE PETTIT OIL CO.\\n\\nthe same inventory . . . and . . . also has priority in\\nidentifiable cash proceeds of the inventory.” (emphasis\\nadded). Had IPC availed itself of this protection by\\nperfecting its interest in the fuel, it would have had priority\\nover the Trustee’s judicial lien extending not just to the fuel,\\nbut also to the proceeds. And there is no persuasive reason\\nto interpret Article 9 as limiting the reciprocal effect—that a\\nconsignor loses priority in the proceeds when it fails to\\nperfect its interest.\\n\\n Moreover, the “goods” provision of section 9-319 can’t\\nbe read in a vacuum. “We must interpret the statute as a\\nwhole, giving effect to each word and making every effort\\nnot to interpret a provision in a manner that renders other\\nprovisions of the same statute inconsistent, meaningless or\\nsuperfluous.” United States v. Neal, 776 F.3d 645, 652 (9th\\nCir. 2015) (internal alterations and quotations omitted).\\nConsistent with this rule of construction, and read in light of\\nArticle 9’s other provisions, we hold that the term “goods”\\nin section 9-319(a) includes the proceeds of those goods, and\\nthat Article 9’s priority and perfection rules apply with equal\\nforce to such proceeds.\\n\\n Although IPC argues the result should be different\\nbecause it retained title to the proceeds, the U.C.C. is clear\\nthat IPC’s retention of title does not matter. Section 9-202\\nof the U.C.C. states that “[e]xcept as otherwise provided\\nwith respect to consignments . . . , the provisions of [Article\\n9] with regard to rights and obligations apply whether title\\nto collateral is in the secured party or the debtor.” Retention\\nof title affects the remedies IPC could employ to recover the\\ngoods in the event of default, but title is irrelevant to whether\\nIPC or the Trustee has priority in the goods and proceeds.\\nSee U.C.C. § 9-202, cmt. 3.a.\\n\\f IN RE PETTIT OIL CO. 9\\n\\n Our conclusion that the term “goods” in section 9-319\\nincludes the proceeds of those goods is bolstered by the\\npolicy rationale underlying these rules. To the outside\\nworld, goods and proceeds held by a consignee appear to be\\nowned by the consignee, and creditors might reasonably\\nbelieve as much when they decide to lend the consignee\\nmoney. The perfection and priority rules—which require\\nthat the consignor publicly announce its interest in the\\nconsigned goods or else go to the back of the line when the\\nconsignee goes bankrupt—serve to protect unwary creditors\\nand prevent “secret liens” in the goods that might otherwise\\ndissuade such lending. See In re Valley Media, Inc., 279\\nB.R. 105, 125 (D. Del. Bankr. 2002) (“The purpose of . . . 9-\\n319(a) is to protect general creditors of the consignee from\\nclaims of consignors that have undisclosed consignment\\narrangements with the consignee that create secret liens on\\nthe inventory.”). A ruling that “proceeds” are outside the\\nscope of the perfection rules would disrupt the delicate\\nbalance the U.C.C. drafters struck between the interests of\\nconsignors and the interests of the consignee’s other\\ncreditors. IPC has not provided a convincing basis for\\ndisrupting this intended balance.\\n\\n B.\\n\\n We also reject IPC’s argument that the Trustee can’t\\nhave an interest in the proceeds because 11 U.S.C.\\n§ 544(a)—which establishes the Trustee as a judicial lien\\nholder—doesn’t contain a “reachback” provision that would\\nallow it to claim an interest in cash and accounts receivable\\nthat arose before Pettit filed its bankruptcy petition. Section\\n544 grants the Trustee “a judicial lien on all property on\\nwhich a creditor on a simple contract could have obtained\\nsuch a judicial lien.” 11 U.S.C. § 544(a)(1) (emphasis\\nadded). However we characterize the Trustee’s interest in\\n\\f10 IN RE PETTIT OIL CO.\\n\\nthe “property” (i.e., the proceeds), that interest is identical to\\nthe interest held by IPC. See U.C.C. § 9-319(a) (“[T]he\\nconsignee is deemed to have rights and title to the goods\\nidentical to those the consignor had.”). IPC undoubtedly\\ncould have secured a lien on the proceeds, so the Trustee\\ncould have as well. Accordingly, the lack of a “reachback”\\nprovision is no barrier to the Trustee claiming an interest in\\nproceeds that arose pre-petition.\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1278,"cells":{"id":{"kind":"number","value":4598813,"string":"4,598,813"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Victoria Dieringer v. Cir"},"decision_date":{"kind":"string","value":"2019-03-12"},"docket_number":{"kind":"string","value":"16-72640"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: M. Margaret McKeown and Richard A. Paez, Circuit Judges, and Cynthia A. Bashant, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PAEZ, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n VICTORIA E. DIERINGER, Deceased; No. 16-72640\\n EUGENE DIERINGER, Executor,\\n Petitioners-Appellants, Tax Ct. No.\\n 21992-13\\n v.\\n\\n COMMISSIONER OF INTERNAL OPINION\\n REVENUE,\\n Respondent-Appellee.\\n\\n\\n\\n Appeal from a Decision of the\\n United States Tax Court\\n\\n Argued and Submitted May 15, 2018\\n Portland, Oregon\\n\\n Filed March 12, 2019\\n\\n Before: M. Margaret McKeown and Richard A. Paez,\\n Circuit Judges, and Cynthia A. Bashant, * District Judge.\\n\\n Opinion by Judge Paez\\n\\n\\n\\n *\\n The Honorable Cynthia A. Bashant, United States District Judge\\nfor the Southern District of California, sitting by designation.\\n\\f2 DIERINGER V. CIR\\n\\n SUMMARY **\\n\\n\\n Tax\\n\\n The panel affirmed the Tax Court’s decision sustaining\\na deficiency against an estate for overstating the amount of\\na charitable deduction, and sustaining an accuracy-related\\npenalty.\\n\\n Petitioner, the estate executor and heir of the decedent,\\ndeclared a large charitable deduction based on the value of\\ncertain estate property at the time of death. The charity\\nreceived less than the amount of the claimed deduction, and\\nthe estate received a tax windfall in the process. Petitioner\\ncontended that the Tax Court should have taken into account\\nevents that occurred after the decedent’s death in\\ndetermining the value of the charitable deduction.\\n\\n In light of this court’s holding in Ahmanson Foundation\\nv. United States, which underscored “the principle that the\\ntestator may only be allowed a deduction for estate tax\\npurposes for what is actually received by the charity,”\\n674 F.2d 761, 772 (9th Cir. 1981), the panel affirmed the\\nTax Court’s decision upholding the reduction of the\\ncharitable deduction and the deficiency assessment. The\\npanel found no clear error in the Tax Court’s finding that\\nthere was no evidence of a significant decline in the\\neconomy that would have decreased the value of the\\nproperty being donated. The panel also found no error in the\\nTax Court’s upholding of the accuracy-related penalty.\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f DIERINGER V. CIR 3\\n\\n COUNSEL\\n\\nW. Michael Gillette (argued), Sara Kobak, and Marc K.\\nSellers, Schwabe Williamson & Wyatt, P.C., Portland,\\nOregon, for Petitioners-Appellants.\\n\\nDouglas Campbell Rennie (argued), Jennifer M. Rubin, and\\nJoan I. Oppenheiner, Attorneys; David A. Hubbert, Acting\\nAssistant Attorney General; Tax Division, United States\\nDepartment of Justice, Washington, D.C.; for Respondent-\\nAppellee.\\n\\n\\n OPINION\\n\\nPAEZ, Circuit Judge:\\n\\n I.\\n\\n In this estate tax case, the Internal Revenue Service\\nCommissioner (“Commissioner”) assessed a deficiency\\nagainst the decedent’s estate for overstating the amount of a\\ncharitable contribution. The estate executor and heir\\ndeclared a large charitable deduction based on the value of\\nestate property at the time of death, only to manipulate the\\nproperty for personal gain, deliver assets to the charity worth\\nsubstantially less than those claimed as a deduction, and\\nreceive a tax windfall in the process. The Tax Court\\nsustained the deficiency. We affirm in light of our holding\\nin Ahmanson Foundation v. United States, where we\\nunderscored “the principle that the testator may only be\\nallowed a deduction for estate tax purposes for what is\\nactually received by the charity.” 674 F.2d 761, 772 (9th\\nCir. 1981). We also affirm the Tax Court’s ruling sustaining\\nan accuracy-related penalty under I.R.C. § 6662.\\n\\f4 DIERINGER V. CIR\\n\\n A.\\n\\n This case involves the Victoria E. Dieringer Estate\\n(“Estate”). Victoria and her late husband Robert E.\\nDieringer (“Robert”) had twelve children together, including\\nEugene Dieringer (“Eugene”), Patrick Dieringer (“Patrick”),\\nand Timothy Dieringer (“Timothy”).\\n\\n The Dieringer family owns Dieringer Properties, Inc.\\n(“DPI”), a closely held corporation that manages\\ncommercial and residential properties in Portland, Oregon,\\nas well as a Wendy’s restaurant in Texas. After Robert\\npassed away and before Victoria’s death, Eugene was the\\npresident of DPI, Patrick was the executive vice-president\\nand secretary, Victoria was the vice-president, and Timothy\\nwas the office manager. DPI’s Board of Directors (“Board”)\\nconsisted of Victoria as chairperson, and Eugene, Patrick,\\nTimothy, and Thomas Keepes (“Keepes”), who is unrelated\\nto the Dieringers, as directors. None of the other nine\\nDieringer children were involved in DPI in any capacity.\\n\\n Before Victoria’s death, the only shareholders of DPI\\nwere Victoria, Eugene, and Patrick. Victoria owned 425 out\\nof 525 voting shares and 7,736.5 out of 9,220.5 nonvoting\\nshares. Eugene owned the remaining 100 voting shares, and\\nEugene and Patrick each owned 742 nonvoting shares.\\n\\n According to Victoria’s will, dated November 10, 2000,\\nupon her death all of her Estate would pass to the Victoria\\nEvelyn Dieringer Trust (“Trust”). The Trust, as amended on\\nApril 22, 2005, provided for Victoria’s children to receive\\nsome personal effects, but no other proceeds from her estate.\\nThe Trust also provided for $600,000 in donations to various\\ncharitable organizations. Any assets remaining in the Estate\\nwould then pass to the Bob and Evelyn Dieringer Family\\nFoundation (“Foundation”), an I.R.C. § 501(c)(3)\\n\\f DIERINGER V. CIR 5\\n\\norganization, as a charitable contribution. Under Victoria’s\\nestate plan, Eugene was appointed as the sole trustee of both\\nthe Trust and the Foundation. Patrick and Keepes served as\\nadvisory trustees of the Foundation.\\n\\n Prior to Victoria’s death, the DPI Board had preliminary\\ndiscussions about purchasing Victoria’s DPI shares as part\\nof ongoing succession planning. A November 24, 2008\\nBoard resolution reported that the issue had been discussed\\nand that the Board resolved to “periodically purchase”\\nVictoria’s shares based on terms acceptable to all parties.\\nVictoria was “completely agreeable” to such a plan. At a\\nFebruary 13, 2009 Board meeting, Victoria reiterated her\\npotential interest in having DPI purchase her shares. In\\nanticipation of entering a purchase agreement for Victoria’s\\nshares, DPI paid the Trust $45,000 prior to Victoria’s death.\\nWhen Victoria unexpectedly died on April 14, 2009, there\\nwere no specific redemption agreements in place. 1\\n\\n When Victoria died, Eugene was appointed executor of\\nthe Estate. To determine the value of Victoria’s DPI shares\\nfor Estate administration purposes, the Estate’s law firm\\nrequested that Lewis Olds & Associates perform an\\nindependent appraisal of the net asset value of DPI. The\\nappraisal determined that the value of DPI as of the date of\\n\\n\\n\\n\\n 1\\n In addition to the Board resolutions that reflect conversations about\\nredemption of Victoria’s stock in 2008 and 2009, the record reflects that\\nas early as 2000 Victoria and Robert contemplated a buy/sell agreement\\nthat would require DPI or Eugene to purchase all of Victoria or Robert’s\\nshares upon the later death of the two of them.\\n\\f6 DIERINGER V. CIR\\n\\nVictoria’s death was $17,777,626 and that Victoria’s shares\\nin DPI were worth $14,182,471. 2\\n\\n Effective November 20, 2009, DPI’s Board converted\\nthe corporate structure from a C corporation to an\\nS corporation on the advice of Keepes. The DPI Board made\\nthis change in corporate structure to accomplish its long-\\nterm tax planning goals and to avoid certain adverse tax\\nconsequences under I.R.C. § 1374. As a result of electing\\nS corporation status, the DPI Board decided that it should\\nalso redeem Victoria’s DPI shares that were to pass to the\\nFoundation. 3\\n\\n The Board entered into an agreement with the Trust to\\nredeem Victoria’s shares, prior to the shares “pouring over”\\ninto the Foundation. Initially, DPI agreed to redeem all of\\nVictoria’s shares for $6,071,558, effective November 30,\\n2009. This amount was based on a 2002 appraisal, since the\\ndate-of-death appraisal had not yet been completed. As a\\nresult, the redemption agreement provided that the stated\\nprice would be “reconciled and adjusted retroactively” to\\nreflect the fair market value of the shares as of November 30,\\n2009. DPI executed two promissory notes payable to the\\n 2\\n Victoria’s voting shares had a market value of $1,824.19 per share\\nand her nonvoting shares had a market value of $1,732.97 per share.\\n 3\\n The DPI Board put forth a number of business rationales for the\\ndecision to redeem the Trust’s DPI shares. These included concerns\\nabout the tax consequences of the Foundation owning shares in an\\nS corporation, that the DPI shares did not provide enough liquidity for\\nthe Foundation to distribute five percent of its funds annually as required\\nby I.R.C. § 4942, and that freezing the value of the Foundation’s DPI\\nshares into a promissory note could prevent future decline in the value\\nof DPI shares given the poor economic climate. The Tax Court did not\\nquestion any of these rationales.\\n\\f DIERINGER V. CIR 7\\n\\nTrust in exchange for the DPI shares. 4 Eugene, Patrick, and\\nTimothy entered into separate subscription agreements to\\npurchase additional DPI shares, in order to provide funding\\nfor DPI to meet the required payments on the promissory\\nnotes. 5 At the later Tax Court trial, the court found that there\\nwere appropriate business purposes for these subscription\\nagreements.\\n\\n At the direction of DPI, Lewis Olds & Associates\\nperformed another appraisal of the value of Victoria’s DPI\\nshares as of November 30, 2009, for the purpose of\\nredemption (“redemption appraisal”). The redemption\\nappraisal valued Victoria’s DPI shares at $916 per voting\\nshare and $870 per nonvoting share. Lewis Olds testified,\\nand the Tax Court found, that Eugene (through his lawyer)\\ninstructed him to value Victoria’s DPI shares as if they were\\na minority interest in DPI for the purposes of this appraisal,\\nand that he would not have done so without these\\ninstructions. Olds’s appraisal therefore included a\\n15-percent discount for lack of control and a 35-percent\\ndiscount for lack of marketability. As a result, Victoria’s\\n\\n\\n 4\\n One of the promissory notes was a short-term note for $2,250,000.\\nThe other was a long-term note for $3,776,558. Each note was payable\\nto the Trust, included interest, and was retroactively adjustable\\ndepending on the new appraisal value.\\n 5\\n These separate subscription agreements provided that voting\\nshares would cost $779 per share and nonvoting shares would cost\\n$742 per share, subject to a retroactive price adjustment. Initially, the\\nsubscription agreements provided that Eugene would purchase\\n2,695 nonvoting shares; Patrick would purchase 86 voting shares and\\n2,695 nonvoting shares; and Timothy would purchase 25 voting shares\\nand 108 nonvoting shares. These amounts were later modified. See infra\\nfn. 7.\\n\\f8 DIERINGER V. CIR\\n\\nDPI shares were valued significantly less in the redemption\\nappraisal than in the date-of-death appraisal. 6\\n\\n DPI determined that it could not afford to redeem all of\\nVictoria’s shares at the new valuation price. The redemption\\nagreement was then amended, and consequently DPI\\nredeemed all 425 voting shares and 5,600.5 nonvoting shares\\nfor a total purchase price of $5,263,462. 7 The balance on the\\nlong-term promissory note was amended to reflect the\\nchanges in the number and price of shares under the\\namended redemption agreement. After the redemption\\nagreement was implemented, the distribution of DPI shares\\nwas as follows: (1) the Trust owned 2,136 nonvoting shares;\\n(2) Eugene owned 200 voting shares and 2,932 nonvoting\\nshares; (3) Patrick owned 65 voting shares and\\n893 nonvoting shares; and (4) Timothy owned 25 voting\\nshares and 108 nonvoting shares. 8\\n\\n In January 2011, the Trust distributed the promissory\\nnotes and the remaining DPI shares to the Foundation. The\\n\\n 6\\n Lewis Olds & Associates worked on producing the appraisals\\nsimultaneously; they are dated March 24, 2010, and March 25, 2010.\\n\\n 7\\n Correspondingly, the subscription agreements also were modified.\\nUnder the modified agreements, Eugene purchased 100 voting shares\\nand 2,190 nonvoting shares, Patrick purchased 65 voting shares and\\n86 nonvoting shares, and Timothy purchased 25 voting shares and\\n108 nonvoting shares.\\n\\n 8\\n There are minor discrepancies between the number of shares\\nindicated in the parties’ stipulation and those indicated in the Tax Court’s\\nopinion. For instance, the number of nonvoting shares owned by Patrick\\ndiffers between the parties’ stipulation (893) and the Tax Court’s opinion\\n(828). We use the number of shares to which the parties stipulated, but\\nthe analysis is the same regardless.\\n\\f DIERINGER V. CIR 9\\n\\nstate probate court approved the redemption agreement and\\nindicated that the redemption agreement and promissory\\nnotes would not be prohibited self-dealing under I.R.C.\\n§ 4941. 9 For the 2011 tax year, the Foundation reported the\\nfollowing contributions on its Form 990-PF, Return of\\nPrivate Foundation: (1) a noncash contribution of DPI shares\\nwith a fair market value of $1,858,961; (2) a long-term note\\nreceivable with a fair market value of $2,921,312; and (3) a\\nshort-term note receivable with a fair market value of\\n$2,250,000. The Trust reported a capital loss of $385,934\\nfor the sale of the 425 voting shares and $4,831,439 for the\\nsale of 5,600.5 nonvoting shares for the taxable year ending\\nin December 31, 2009, on its Form 1041 Tax Return. The\\nEstate filed its Form 706, United States Estate (and\\nGeneration-Skipping Transfer) Tax Return on July 12, 2010,\\nreporting no estate tax liability. The Estate claimed a\\ncharitable contribution deduction of $18,812,181, based on\\nthe date-of-death value of Victoria’s DPI shares.\\n\\n B.\\n\\n In June 2013, the Commissioner issued a notice of\\ndeficiency to the Estate based on its July 2010 tax return\\n(Form 706). The notice stated that there was a deficiency of\\n$4,124,717 and imposed an accuracy-related penalty of\\n$824,943 under I.R.C. § 6662 for error and negligence in\\nusing the date-of-death appraisal as the value of the\\ncharitable contribution of Victoria’s DPI shares.\\n\\n Upon receiving the notice of deficiency, the Estate filed\\na timely petition in the Tax Court challenging the\\n\\n 9\\n I.R.C. § 4941 allows the Commissioner to impose a tax on certain\\ntransactions between a private foundation and a “disqualified person” as\\ndefined by I.R.C. § 4946(a)(1).\\n\\f10 DIERINGER V. CIR\\n\\nCommissioner’s deficiency notice and penalty assessment.\\nThe Estate argued that it correctly used the date-of-death\\nappraisal to determine the value of Victoria’s DPI shares for\\nthe purpose of the charitable contribution deduction. The\\nCommissioner responded that post-death events should be\\nconsidered in determining the value of the charitable\\ncontribution, as the actions by Eugene, Patrick, and Timothy\\nreduced the value of Victoria’s contribution to the\\nFoundation.\\n\\n Following trial, the Tax Court issued a decision\\nupholding the Commissioner’s reduction of the Estate’s\\ncharitable deduction and the deficiency assessment. The\\nTax Court found that the redemption was not part of\\nVictoria’s estate plan and there were valid business reasons\\nfor many of the transactions that took place after her death.\\nThe Tax Court also concluded, however, that post-death\\nevents—primarily Eugene’s decision to apply a minority\\ninterest discount to the redemption value of Victoria’s DPI\\nshares—reduced the value of the contribution to the\\nFoundation and therefore reduced the value of the Estate’s\\ncharitable deduction. But the evidence did not support the\\nconclusion that a poor business climate caused the reported\\ndecline in share values, as the Estate argued. Further, the\\nTax Court affirmed the Commissioner’s determination that\\nthe Estate was liable for a penalty under I.R.C. § 6662(a).\\nAfter denying the Estate’s motion for reconsideration, the\\nTax Court entered a final decision in favor of the\\nCommissioner, which sustained an estate tax deficiency of\\n$4,124,717 and an accuracy-related penalty of $824,943.\\nThe Estate timely appealed.\\n\\n II.\\n\\n We review de novo the Tax Court’s legal conclusions,\\nsee DJB Holding Corp. v. Comm’r, 803 F.3d 1014, 1022 (9th\\n\\f DIERINGER V. CIR 11\\n\\nCir. 2015), as well as the Tax Court’s interpretation of the\\nInternal Revenue Code, see Metro One Telecomms., Inc. v.\\nComm’r, 704 F.3d 1057, 1059 (9th Cir. 2012). We review\\nfor clear error the Tax Court’s factual findings. See DJB\\nHolding Corp., 803 F.3d at 1022. Under clear error review,\\nif the Tax Court’s interpretation of the evidence is plausible,\\nwe must uphold it. See Leslie v. Comm’r, 146 F.3d 643, 646\\n(9th Cir. 1998).\\n\\n Where the Tax Court sustains an accuracy-related\\npenalty, we review for clear error the Tax Court’s finding of\\nnegligence. Hansen v. Comm’r, 471 F.3d 1021, 1028 (9th\\nCir. 2006). We review de novo whether substantial authority\\nsupported the taxpayer’s position, but we review for clear\\nerror whether the taxpayer acted with reasonable cause and\\nin good faith. DJB Holding Corp., 803 F.3d at 1022.\\n\\n III.\\n\\n The Estate challenges the Tax Court’s final decision on\\nthree grounds. First, it argues that the Tax Court erred by\\ntaking into account events that occurred after Victoria’s\\ndeath in determining the value of the charitable deduction.\\nInstead, the Estate argues that the charitable deduction\\nshould have been valued as of Victoria’s date of death.\\nSecond, the Estate argues that even if post-death events\\ncould be considered, the Tax Court erred by not accounting\\nfor a decline in value of Victoria’s shares caused by\\neconomic forces. Third, the Estate argues that the Tax Court\\nerred by upholding the accuracy-related penalty under I.R.C.\\n§ 6662. We reject all three arguments.\\n\\n A.\\n\\n At the center of the parties’ dispute is whether the\\nEstate’s charitable deduction should be valued at the time of\\n\\f12 DIERINGER V. CIR\\n\\nVictoria’s death, or whether the post-death events that\\ndecreased value of the property delivered to charity should\\nbe considered.\\n\\nThe Estate Tax and Charitable Deductions\\n\\n It is well established that the “estate tax is a tax on the\\nprivilege of transfer[r]ing property” after one’s death.\\nPropstra v. United States, 680 F.2d 1248, 1250 (9th Cir.\\n1982). The estate tax “is on the act of the testator not on the\\nreceipt of the property by the legatees.” Estate of Simplot v.\\nComm’r, 249 F.3d 1191, 1194 (9th Cir. 2001) (citing Ithaca\\nTr. Co. v. United States, 279 U.S. 151, 155 (1929)).\\n\\n Because the estate tax is a tax on the decedent’s bequest\\nof property, the valuation of the gross estate is typically done\\nas of the date of death. See I.R.C. § 2031; Tr. Servs. of Am.,\\nInc. v. United States, 885 F.2d 561, 568–69 (9th Cir. 1989).\\nThe pertinent statute provides that “[t]he value of the gross\\nestate of the decedent shall be determined by including to the\\nextent provided for in this part, the value at the time of his\\ndeath of all property . . . wherever situated.” I.R.C.\\n§ 2031(a); see also id. § 2033 (mandating that property in\\nwhich the decedent had an interest be valued on the date of\\ndecedent’s death); Treas. Reg. § 20.2031-1(b) (requiring the\\nvaluation of property at the “fair market value at the time of\\nthe decedent’s death” unless an exception applies). Except\\nin some limited circumstances, such as when the executor\\nelects to use an alternative valuation under I.R.C. § 2032,\\npost-death events are generally not considered in\\ndetermining the estate’s gross value for purposes of the\\nestate tax. The parties agree that the executor did not use the\\nalternative valuation method in this case.\\n\\n A related provision allows for deductions from the value\\nof the gross estate for transfers of assets to qualified\\n\\f DIERINGER V. CIR 13\\n\\ncharitable entities. I.R.C. § 2055(a). This deduction\\ngenerally is allowed “for the value of property included in\\nthe decedent’s gross estate and transferred by the decedent\\n. . . by will.” Treas. Reg. § 20.2055-1(a). Congress’s\\nallowance for a charitable deduction was a “liberalization[]\\nof the law in the taxpayer’s favor.” Helvering v. Bliss,\\n293 U.S. 144, 151 (1934). Yet, as we have recognized,\\ndeductions are acts of “legislative grace.” Comm’r v.\\nShoong, 177 F.2d 131, 132 (9th Cir. 1949). “[T]he purpose\\nof the charitable deduction [is] to encourage gifts to charity.”\\nAhmanson, 674 F.2d at 772; see also Underwood v. United\\nStates, 407 F.2d 608, 610 (6th Cir. 1969) (“The purpose of\\nallowing charitable deductions is to encourage testators to\\nmake charitable bequests, not to permit executors and\\nbeneficiaries to rewrite a will so as to achieve tax savings.”).\\n\\nValuing the Charitable Deduction\\n\\n Deductions are valued separately from the valuation of\\nthe gross estate. See Ahmanson, 674 F.2d at 772 (“The\\nstatute does not ordain equal valuation as between an item in\\nthe gross estate and the same item under the charitable\\ndeduction.”). Separate valuations allow for the\\nconsideration of post-death events, as required by Ahmanson\\nand provisions of the tax code.\\n\\n We addressed valuation of a charitable deduction in\\nAhmanson. There, the decedent’s estate plan provided for\\nthe voting shares in a corporation to be left to family\\nmembers and the nonvoting shares to be left to a charitable\\nfoundation. Id. at 765–66. We held that when valuing the\\ncharitable deduction for the nonvoting shares, a discount\\nshould be applied to account for the fact that the shares\\ndonated to charity had been stripped of their voting power.\\nId. at 772. That a discount was not applied to the value of\\nthe nonvoting shares in the gross estate did not impact our\\n\\f14 DIERINGER V. CIR\\n\\nholding. Id. Importantly, we recognized that a charitable\\ndeduction “is subject to the principle that the testator may\\nonly be allowed a deduction for estate tax purposes for what\\nis actually received by the charity.” Id.\\n\\n In contrast, the Estate argues the charitable deduction\\nmust be valued as of the date of Victoria’s death, in keeping\\nwith the date-of-death valuation of an estate. We disagree.\\nAlthough the Supreme Court and our court have applied that\\nvaluation method where the remainder of an estate is\\ndonated to charity after a post-death contingency, there is no\\nuniform rule for all circumstances. Ithaca Trust, 279 U.S. at\\n154; Wells Fargo Bank & Union Tr. Co. v. Comm’r,\\n145 F.2d 132 (9th Cir. 1944). In Ithaca Trust, the decedent\\nleft the remainder of his estate to his wife for her life, with\\nany assets remaining after her death to be donated to charity.\\n279 U.S. at 154. The decedent’s wife unexpectedly died\\nonly six months after the decedent. Id. at 155. The Court\\nconsidered whether the charitable deduction should reflect\\nthe amount actually given to charity after the wife’s\\npremature death, but ultimately determined that the\\ndeduction should instead reflect a valuation based on\\nmortality tables for the wife on the date of the decedent’s\\ndeath. Id.\\n\\n We applied Ithaca Trust in Wells Fargo, where the\\ndecedent had created a trust with instructions that the trust’s\\nincome be paid to the decedent’s sister from the date of the\\ndecedent’s death to the date of the sister’s death, and after\\nthe sister’s death that the corpus of the trust be given to\\ncharity. 145 F.2d at 133. Due to commingling of assets after\\nthe decedent’s death and an unrelated tax dispute, part of the\\ncorpus of the trust was used to support the decedent’s sister,\\nwhich reduced the amount that went to charity upon her\\ndeath. Id. The Tax Court relied on these payments to\\n\\f DIERINGER V. CIR 15\\n\\ndisallow part of the charitable deduction, but we reversed,\\nconcluding that the charitable deduction must be determined\\nfrom data available at the time of death. Id.; see also Estate\\nof Van Horne v. Comm’r, 720 F.2d 1114, 1117 (9th Cir.\\n1983) (relying on Ithaca Trust to “hold that legally\\nenforceable claims valued by reference to an actuarial table\\nmeet the test of certainty for estate tax purposes” when\\nvaluing spousal support obligation).\\n\\n Neither Ithaca Trust nor Wells Fargo, however, set in\\nstone the date of death as the date of the valuation of assets\\nfor purposes of a deduction. See Shedd’s Estate v. Comm’r,\\n320 F.2d 638, 639 (9th Cir. 1963) (“Congress did not intend\\nto make events at the date of death invariably determinative\\nin computing the federal estate tax obligation.”). Nor could\\nthey. Certain deductions not only permit consideration of\\npost-death events, but require them. For example, I.R.C.\\n§ 2053(a) authorizes a deduction for funeral expenses and\\nestate administration expenses—costs that cannot accrue\\nuntil after the death of the testator. Similarly, I.R.C.\\n§ 2055(c) specifies that where death taxes are payable out of\\na charitable bequest, any charitable deduction is limited to\\nthe value remaining in the estate after such post-death tax\\npayment. Still another provision of the tax code, I.R.C.\\n§ 2055(d), prohibits the amount of a charitable deduction\\nfrom exceeding the value of transferred property included in\\na gross estate—but, by negative implication, permits such a\\ndeduction to be lower than the value of donated assets at the\\nmoment of death. The Third Circuit also recognized that\\nvaluations of the gross estate and a charitable deduction are\\nseparate and may differ. In Re Sage’s Estate, 122 F.2d 480,\\n484 (3d Cir. 1941) (“[W]hile a decedent’s gross estate is\\nfixed as of the date of his death, deductions claimed in\\ndetermining the net estate subject to tax may not be\\n\\f16 DIERINGER V. CIR\\n\\nascertainable or even accrue until the happening of events\\nsubsequent to death.”).\\n\\n Indeed, “[t]he proper administration of the charitable\\ndeduction cannot ignore such differences in the value\\nactually received by the charity.” Ahmanson, 674 F.2d at\\n772; accord Irving Tr. Co. v. United States, 221 F.2d 303,\\n306 (2d Cir. 1955); Thompson’s Estate v. Comm’r, 123 F.2d\\n816, 817 (2d Cir. 1941); In Re Sage’s Estate, 122 F.2d at\\n484. This rule prohibits crafting an estate plan or will so as\\nto game the system and guarantee a charitable deduction that\\nis larger than the amount actually given to charity.\\nAhmanson, 674 F.2d at 772.\\n\\nApplying Ahmanson’s Rule\\n\\n Ahmanson compels affirming the Tax Court’s ruling\\nhere. Victoria structured her Estate so as not to donate her\\nDPI shares directly to a charity, or even directly to the\\nFoundation, but to the Trust. Victoria enabled Eugene to\\ncommit almost unchecked abuse of the Estate by setting him\\nup to be executor of the Estate, trustee of the Trust, and\\ntrustee of the Foundation, in addition to his roles as\\npresident, director, and majority shareholder of DPI. As the\\nTax Court found, Eugene improperly directed Lewis Olds to\\ndetermine the redemption value of the DPI shares by\\napplying a minority interest valuation, when he knew a\\nmajority interest applied and the Estate had claimed a\\ncharitable deduction based upon a majority interest\\nvaluation. Through his actions, Eugene manipulated the\\ncharitable deduction so that the Foundation only received a\\nfraction of the charitable deduction claimed by the Estate.\\n\\n The Estate attempts to evade Ahmanson by arguing that\\nit is limited to situations where the testamentary plan\\ndiminishes the value of the charitable property. Read in\\n\\f DIERINGER V. CIR 17\\n\\ncontext, Ahmanson is not limited to abuses in the four\\ncorners of the testamentary plan. Id. Ahmanson extends to\\nsituations where “the testator would be able to produce an\\nartificially low valuation by manipulat[ion],” which includes\\nthe present situation. Id. Moreover, Victoria’s testamentary\\nplan laid the groundwork for Eugene’s manipulation by\\nconcentrating power in his hands—in his roles as executor\\nof the Estate and trustee of the Trust and Foundation—even\\nafter she knew of and assented to early discussions of the\\nshare redemption plan.\\n\\n The Tax Court correctly considered the difference\\nbetween the deduction and the property actually received by\\nthe charity due to Eugene’s manipulation of the redemption\\nappraisal value. 10\\n\\n B.\\n\\n The Estate argues that any consideration of post-death\\nevents also requires finding that the decline in value of DPI\\nstock was due, at least in part, to market forces. The Tax\\nCourt, however, found that “[t]he evidence does not support\\na significant decline in the economy that resulted in a large\\ndecrease in value in only seven months.” The Tax Court\\nacknowledged that the adjusted net asset value of the DPI\\nstock decreased from $17,777,626 at the date of Victoria’s\\ndeath to $16,159,167 at the date used for the redemption\\nappraisal. 11 Still, the Tax Court found, “[t]he reported\\n\\n 10\\n The Commissioner argues in the alternative that the events in this\\ncase fall under the exception in Treas. Reg. § 20.2055-2(b)(1). In light\\nof our holding that Ahmanson compels affirmance, we need not address\\nthis argument.\\n 11\\n The Estate suggests that this change in adjusted net asset value is\\nentirely attributable to a changed real estate market. This argument\\n\\f18 DIERINGER V. CIR\\n\\ndecline in per share value was primarily due to the specific\\ninstruction to value decedent’s majority interest as a\\nminority interest with a 50% discount.” We find nothing in\\nthe record—nor does the Estate point to anything in the\\nrecord—that suggests the Tax Court’s findings were clearly\\nerroneous.\\n\\n C.\\n\\n Finally, the Estate challenges the accuracy-related\\npenalty that was imposed pursuant to I.R.C. § 6662(a).\\nSection 6662(a) imposes a penalty of 20 percent of the\\namount of the underpayment attributable to either\\nnegligence or disregard of rules or regulations. A penalty\\nwill not be imposed if the taxpayer can show “that there was\\na reasonable cause [for the underpayment] and that the\\ntaxpayer acted in good faith.” I.R.C. § 6664(c)(1). The\\nCommissioner bears the initial burden of production\\nregarding the applicability of the penalty, but once this\\nburden is met, the taxpayer bears the burden of persuasion\\nas to defenses to the penalty. Id. § 7491(c); Higbee v.\\nComm’r, 116 T.C. 438, 446–47 (2001).\\n\\n The Tax Court first found that the Estate acted\\nnegligently. Section 6662(c) defines negligence as “any\\nfailure to make a reasonable attempt to comply” with the\\nI.R.C. See also Allen v. Comm’r, 925 F.2d 348, 353 (9th Cir.\\n1991) (defining negligence as a “lack of due care or the\\nfailure to do what a reasonable and prudent person would do\\nunder similar circumstances”). The Tax Court found that the\\nEstate was negligent because it “not only failed to inform the\\n\\n\\nmisses the point. The Estate claimed an $18,205,476 charitable\\ndeduction largely based on the value of assets earmarked for the\\nFoundation—but the Foundation only received assets worth $6,434,578.\\n\\f DIERINGER V. CIR 19\\n\\nappraiser that the redemption was for a majority interest, but\\nalso instructed the appraiser to value the redeemed DPI stock\\nas a minority interest.” We agree; the Tax Court did not\\nclearly err in finding negligence.\\n\\n The Tax Court next found that the Estate did not have\\nreasonable cause and good faith for its negligent act. “The\\ndetermination of whether a taxpayer acted with reasonable\\ncause and in good faith is made on a case-by-case basis,\\ntaking into account all pertinent facts and circumstances.”\\nTreas. Reg. § 1.6664-4(b)(1). The most important factor “is\\nthe extent of the taxpayer’s effort to assess the taxpayer’s\\nproper tax liability.” Id. A reliance on professional advice\\nmay be used to show reasonable cause and good faith, but\\nonly if the professional advice meets certain requirements.\\nId. (instructing that “the circumstances under which the\\nappraisal was obtained” are considered in determining\\nwhether there was reasonable cause and good faith).\\n\\n The record evidence supports the Tax Court’s finding\\nthat the Estate did not rely in good faith on its attorney’s\\njudgment. The redemption appraisal was inaccurate because\\nof instructions from Eugene on behalf of the Estate.\\nTherefore, the Estate “knew” that Eugene and his brothers\\nwere acquiring the DPI stock at a discount. Further, the\\nappraiser’s credible testimony that he was instructed to\\nundertake a minority interest valuation, which he ordinarily\\nwould not have done in this situation, demonstrates the lack\\nof good faith. See id. On these facts we find no error in the\\nTax Court’s holding that the Commissioner properly\\nimposed the accuracy-related penalty under I.R.C.\\n§ 6662(a).\\n\\f20 DIERINGER V. CIR\\n\\n IV.\\n\\n The Estate claimed an $18,205,476 charitable deduction\\nbased on the value of assets earmarked for the Foundation,\\nbut the Foundation received assets worth only $6,434,578.\\nThe Estate, at Eugene’s direction, claimed a large charitable\\ndeduction representing that such assets would be delivered\\nto the Foundation. Eugene then revalued and delivered\\nassets—promissory notes and DPI shares—worth far less\\nthan the claimed charitable deduction. By doing so, the\\nEstate—and various Dieringer family entities—ended up\\navoiding $4.1 million in taxes. We agree with the Tax\\nCourt’s decision upholding the Commissioner’s reduction of\\nthe Estate’s charitable deduction and the deficiency\\nassessment. To hold otherwise would only “invite abuse.”\\nAhmanson, 674 F.2d at 768.\\n\\n AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4376066/\", \"author_raw\": \"PAEZ, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"M MARGARET MCKEOWN"},"judge2":{"kind":"string","value":"RICHARD A PAEZ"},"judge3":{"kind":"string","value":"CYNTHIA A BASHANT"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4598813/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Victoria Dieringer v. Cir\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-12\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n VICTORIA E. DIERINGER, Deceased; No. 16-72640\n EUGENE DIERINGER, Executor,\n Petitioners-Appellants, Tax Ct. No.\n 21992-13\n v.\n\n COMMISSIONER OF INTERNAL OPINION\n REVENUE,\n Respondent-Appellee.\n\n\n\n Appeal from a Decision of the\n United States Tax Court\n\n Argued and Submitted May 15, 2018\n Portland, Oregon\n\n Filed March 12, 2019\n\n Before: M. Margaret McKeown and Richard A. Paez,\n Circuit Judges, and Cynthia A. Bashant, * District Judge.\n\n Opinion by Judge Paez\n\n\n\n *\n The Honorable Cynthia A. Bashant, United States District Judge\nfor the Southern District of California, sitting by designation.\n\f2 DIERINGER V. CIR\n\n SUMMARY **\n\n\n Tax\n\n The panel affirmed the Tax Court’s decision sustaining\na deficiency against an estate for overstating the amount of\na charitable deduction, and sustaining an accuracy-related\npenalty.\n\n Petitioner, the estate executor and heir of the decedent,\ndeclared a large charitable deduction based on the value of\ncertain estate property at the time of death. The charity\nreceived less than the amount of the claimed deduction, and\nthe estate received a tax windfall in the process. Petitioner\ncontended that the Tax Court should have taken into account\nevents that occurred after the decedent’s death in\ndetermining the value of the charitable deduction.\n\n In light of this court’s holding in Ahmanson Foundation\nv. United States, which underscored “the principle that the\ntestator may only be allowed a deduction for estate tax\npurposes for what is actually received by the charity,”\n674 F.2d 761, 772 (9th Cir. 1981), the panel affirmed the\nTax Court’s decision upholding the reduction of the\ncharitable deduction and the deficiency assessment. The\npanel found no clear error in the Tax Court’s finding that\nthere was no evidence of a significant decline in the\neconomy that would have decreased the value of the\nproperty being donated. The panel also found no error in the\nTax Court’s upholding of the accuracy-related penalty.\n\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f DIERINGER V. CIR 3\n\n COUNSEL\n\nW. Michael Gillette (argued), Sara Kobak, and Marc K.\nSellers, Schwabe Williamson & Wyatt, P.C., Portland,\nOregon, for Petitioners-Appellants.\n\nDouglas Campbell Rennie (argued), Jennifer M. Rubin, and\nJoan I. Oppenheiner, Attorneys; David A. Hubbert, Acting\nAssistant Attorney General; Tax Division, United States\nDepartment of Justice, Washington, D.C.; for Respondent-\nAppellee.\n\n\n OPINION\n\nPAEZ, Circuit Judge:\n\n I.\n\n In this estate tax case, the Internal Revenue Service\nCommissioner (“Commissioner”) assessed a deficiency\nagainst the decedent’s estate for overstating the amount of a\ncharitable contribution. The estate executor and heir\ndeclared a large charitable deduction based on the value of\nestate property at the time of death, only to manipulate the\nproperty for personal gain, deliver assets to the charity worth\nsubstantially less than those claimed as a deduction, and\nreceive a tax windfall in the process. The Tax Court\nsustained the deficiency. We affirm in light of our holding\nin Ahmanson Foundation v. United States, where we\nunderscored “the principle that the testator may only be\nallowed a deduction for estate tax purposes for what is\nactually received by the charity.” 674 F.2d 761, 772 (9th\nCir. 1981). We also affirm the Tax Court’s ruling sustaining\nan accuracy-related penalty under I.R.C. § 6662.\n\f4 DIERINGER V. CIR\n\n A.\n\n This case involves the Victoria E. Dieringer Estate\n(“Estate”). Victoria and her late husband Robert E.\nDieringer (“Robert”) had twelve children together, including\nEugene Dieringer (“Eugene”), Patrick Dieringer (“Patrick”),\nand Timothy Dieringer (“Timothy”).\n\n The Dieringer family owns Dieringer Properties, Inc.\n(“DPI”), a closely held corporation that manages\ncommercial and residential properties in Portland, Oregon,\nas well as a Wendy’s restaurant in Texas. After Robert\npassed away and before Victoria’s death, Eugene was the\npresident of DPI, Patrick was the executive vice-president\nand secretary, Victoria was the vice-president, and Timothy\nwas the office manager. DPI’s Board of Directors (“Board”)\nconsisted of Victoria as chairperson, and Eugene, Patrick,\nTimothy, and Thomas Keepes (“Keepes”), who is unrelated\nto the Dieringers, as directors. None of the other nine\nDieringer children were involved in DPI in any capacity.\n\n Before Victoria’s death, the only shareholders of DPI\nwere Victoria, Eugene, and Patrick. Victoria owned 425 out\nof 525 voting shares and 7,736.5 out of 9,220.5 nonvoting\nshares. Eugene owned the remaining 100 voting shares, and\nEugene and Patrick each owned 742 nonvoting shares.\n\n According to Victoria’s will, dated November 10, 2000,\nupon her death all of her Estate would pass to the Victoria\nEvelyn Dieringer Trust (“Trust”). The Trust, as amended on\nApril 22, 2005, provided for Victoria’s children to receive\nsome personal effects, but no other proceeds from her estate.\nThe Trust also provided for $600,000 in donations to various\ncharitable organizations. Any assets remaining in the Estate\nwould then pass to the Bob and Evelyn Dieringer Family\nFoundation (“Foundation”), an I.R.C. § 501(c)(3)\n\f DIERINGER V. CIR 5\n\norganization, as a charitable contribution. Under Victoria’s\nestate plan, Eugene was appointed as the sole trustee of both\nthe Trust and the Foundation. Patrick and Keepes served as\nadvisory trustees of the Foundation.\n\n Prior to Victoria’s death, the DPI Board had preliminary\ndiscussions about purchasing Victoria’s DPI shares as part\nof ongoing succession planning. A November 24, 2008\nBoard resolution reported that the issue had been discussed\nand that the Board resolved to “periodically purchase”\nVictoria’s shares based on terms acceptable to all parties.\nVictoria was “completely agreeable” to such a plan. At a\nFebruary 13, 2009 Board meeting, Victoria reiterated her\npotential interest in having DPI purchase her shares. In\nanticipation of entering a purchase agreement for Victoria’s\nshares, DPI paid the Trust $45,000 prior to Victoria’s death.\nWhen Victoria unexpectedly died on April 14, 2009, there\nwere no specific redemption agreements in place. 1\n\n When Victoria died, Eugene was appointed executor of\nthe Estate. To determine the value of Victoria’s DPI shares\nfor Estate administration purposes, the Estate’s law firm\nrequested that Lewis Olds & Associates perform an\nindependent appraisal of the net asset value of DPI. The\nappraisal determined that the value of DPI as of the date of\n\n\n\n\n 1\n In addition to the Board resolutions that reflect conversations about\nredemption of Victoria’s stock in 2008 and 2009, the record reflects that\nas early as 2000 Victoria and Robert contemplated a buy/sell agreement\nthat would require DPI or Eugene to purchase all of Victoria or Robert’s\nshares upon the later death of the two of them.\n\f6 DIERINGER V. CIR\n\nVictoria’s death was $17,777,626 and that Victoria’s shares\nin DPI were worth $14,182,471. 2\n\n Effective November 20, 2009, DPI’s Board converted\nthe corporate structure from a C corporation to an\nS corporation on the advice of Keepes. The DPI Board made\nthis change in corporate structure to accomplish its long-\nterm tax planning goals and to avoid certain adverse tax\nconsequences under I.R.C. § 1374. As a result of electing\nS corporation status, the DPI Board decided that it should\nalso redeem Victoria’s DPI shares that were to pass to the\nFoundation. 3\n\n The Board entered into an agreement with the Trust to\nredeem Victoria’s shares, prior to the shares “pouring over”\ninto the Foundation. Initially, DPI agreed to redeem all of\nVictoria’s shares for $6,071,558, effective November 30,\n2009. This amount was based on a 2002 appraisal, since the\ndate-of-death appraisal had not yet been completed. As a\nresult, the redemption agreement provided that the stated\nprice would be “reconciled and adjusted retroactively” to\nreflect the fair market value of the shares as of November 30,\n2009. DPI executed two promissory notes payable to the\n 2\n Victoria’s voting shares had a market value of $1,824.19 per share\nand her nonvoting shares had a market value of $1,732.97 per share.\n 3\n The DPI Board put forth a number of business rationales for the\ndecision to redeem the Trust’s DPI shares. These included concerns\nabout the tax consequences of the Foundation owning shares in an\nS corporation, that the DPI shares did not provide enough liquidity for\nthe Foundation to distribute five percent of its funds annually as required\nby I.R.C. § 4942, and that freezing the value of the Foundation’s DPI\nshares into a promissory note could prevent future decline in the value\nof DPI shares given the poor economic climate. The Tax Court did not\nquestion any of these rationales.\n\f DIERINGER V. CIR 7\n\nTrust in exchange for the DPI shares. 4 Eugene, Patrick, and\nTimothy entered into separate subscription agreements to\npurchase additional DPI shares, in order to provide funding\nfor DPI to meet the required payments on the promissory\nnotes. 5 At the later Tax Court trial, the court found that there\nwere appropriate business purposes for these subscription\nagreements.\n\n At the direction of DPI, Lewis Olds & Associates\nperformed another appraisal of the value of Victoria’s DPI\nshares as of November 30, 2009, for the purpose of\nredemption (“redemption appraisal”). The redemption\nappraisal valued Victoria’s DPI shares at $916 per voting\nshare and $870 per nonvoting share. Lewis Olds testified,\nand the Tax Court found, that Eugene (through his lawyer)\ninstructed him to value Victoria’s DPI shares as if they were\na minority interest in DPI for the purposes of this appraisal,\nand that he would not have done so without these\ninstructions. Olds’s appraisal therefore included a\n15-percent discount for lack of control and a 35-percent\ndiscount for lack of marketability. As a result, Victoria’s\n\n\n 4\n One of the promissory notes was a short-term note for $2,250,000.\nThe other was a long-term note for $3,776,558. Each note was payable\nto the Trust, included interest, and was retroactively adjustable\ndepending on the new appraisal value.\n 5\n These separate subscription agreements provided that voting\nshares would cost $779 per share and nonvoting shares would cost\n$742 per share, subject to a retroactive price adjustment. Initially, the\nsubscription agreements provided that Eugene would purchase\n2,695 nonvoting shares; Patrick would purchase 86 voting shares and\n2,695 nonvoting shares; and Timothy would purchase 25 voting shares\nand 108 nonvoting shares. These amounts were later modified. See infra\nfn. 7.\n\f8 DIERINGER V. CIR\n\nDPI shares were valued significantly less in the redemption\nappraisal than in the date-of-death appraisal. 6\n\n DPI determined that it could not afford to redeem all of\nVictoria’s shares at the new valuation price. The redemption\nagreement was then amended, and consequently DPI\nredeemed all 425 voting shares and 5,600.5 nonvoting shares\nfor a total purchase price of $5,263,462. 7 The balance on the\nlong-term promissory note was amended to reflect the\nchanges in the number and price of shares under the\namended redemption agreement. After the redemption\nagreement was implemented, the distribution of DPI shares\nwas as follows: (1) the Trust owned 2,136 nonvoting shares;\n(2) Eugene owned 200 voting shares and 2,932 nonvoting\nshares; (3) Patrick owned 65 voting shares and\n893 nonvoting shares; and (4) Timothy owned 25 voting\nshares and 108 nonvoting shares. 8\n\n In January 2011, the Trust distributed the promissory\nnotes and the remaining DPI shares to the Foundation. The\n\n 6\n Lewis Olds & Associates worked on producing the appraisals\nsimultaneously; they are dated March 24, 2010, and March 25, 2010.\n\n 7\n Correspondingly, the subscription agreements also were modified.\nUnder the modified agreements, Eugene purchased 100 voting shares\nand 2,190 nonvoting shares, Patrick purchased 65 voting shares and\n86 nonvoting shares, and Timothy purchased 25 voting shares and\n108 nonvoting shares.\n\n 8\n There are minor discrepancies between the number of shares\nindicated in the parties’ stipulation and those indicated in the Tax Court’s\nopinion. For instance, the number of nonvoting shares owned by Patrick\ndiffers between the parties’ stipulation (893) and the Tax Court’s opinion\n(828). We use the number of shares to which the parties stipulated, but\nthe analysis is the same regardless.\n\f DIERINGER V. CIR 9\n\nstate probate court approved the redemption agreement and\nindicated that the redemption agreement and promissory\nnotes would not be prohibited self-dealing under I.R.C.\n§ 4941. 9 For the 2011 tax year, the Foundation reported the\nfollowing contributions on its Form 990-PF, Return of\nPrivate Foundation: (1) a noncash contribution of DPI shares\nwith a fair market value of $1,858,961; (2) a long-term note\nreceivable with a fair market value of $2,921,312; and (3) a\nshort-term note receivable with a fair market value of\n$2,250,000. The Trust reported a capital loss of $385,934\nfor the sale of the 425 voting shares and $4,831,439 for the\nsale of 5,600.5 nonvoting shares for the taxable year ending\nin December 31, 2009, on its Form 1041 Tax Return. The\nEstate filed its Form 706, United States Estate (and\nGeneration-Skipping Transfer) Tax Return on July 12, 2010,\nreporting no estate tax liability. The Estate claimed a\ncharitable contribution deduction of $18,812,181, based on\nthe date-of-death value of Victoria’s DPI shares.\n\n B.\n\n In June 2013, the Commissioner issued a notice of\ndeficiency to the Estate based on its July 2010 tax return\n(Form 706). The notice stated that there was a deficiency of\n$4,124,717 and imposed an accuracy-related penalty of\n$824,943 under I.R.C. § 6662 for error and negligence in\nusing the date-of-death appraisal as the value of the\ncharitable contribution of Victoria’s DPI shares.\n\n Upon receiving the notice of deficiency, the Estate filed\na timely petition in the Tax Court challenging the\n\n 9\n I.R.C. § 4941 allows the Commissioner to impose a tax on certain\ntransactions between a private foundation and a “disqualified person” as\ndefined by I.R.C. § 4946(a)(1).\n\f10 DIERINGER V. CIR\n\nCommissioner’s deficiency notice and penalty assessment.\nThe Estate argued that it correctly used the date-of-death\nappraisal to determine the value of Victoria’s DPI shares for\nthe purpose of the charitable contribution deduction. The\nCommissioner responded that post-death events should be\nconsidered in determining the value of the charitable\ncontribution, as the actions by Eugene, Patrick, and Timothy\nreduced the value of Victoria’s contribution to the\nFoundation.\n\n Following trial, the Tax Court issued a decision\nupholding the Commissioner’s reduction of the Estate’s\ncharitable deduction and the deficiency assessment. The\nTax Court found that the redemption was not part of\nVictoria’s estate plan and there were valid business reasons\nfor many of the transactions that took place after her death.\nThe Tax Court also concluded, however, that post-death\nevents—primarily Eugene’s decision to apply a minority\ninterest discount to the redemption value of Victoria’s DPI\nshares—reduced the value of the contribution to the\nFoundation and therefore reduced the value of the Estate’s\ncharitable deduction. But the evidence did not support the\nconclusion that a poor business climate caused the reported\ndecline in share values, as the Estate argued. Further, the\nTax Court affirmed the Commissioner’s determination that\nthe Estate was liable for a penalty under I.R.C. § 6662(a).\nAfter denying the Estate’s motion for reconsideration, the\nTax Court entered a final decision in favor of the\nCommissioner, which sustained an estate tax deficiency of\n$4,124,717 and an accuracy-related penalty of $824,943.\nThe Estate timely appealed.\n\n II.\n\n We review de novo the Tax Court’s legal conclusions,\nsee DJB Holding Corp. v. Comm’r, 803 F.3d 1014, 1022 (9th\n\f DIERINGER V. CIR 11\n\nCir. 2015), as well as the Tax Court’s interpretation of the\nInternal Revenue Code, see Metro One Telecomms., Inc. v.\nComm’r, 704 F.3d 1057, 1059 (9th Cir. 2012). We review\nfor clear error the Tax Court’s factual findings. See DJB\nHolding Corp., 803 F.3d at 1022. Under clear error review,\nif the Tax Court’s interpretation of the evidence is plausible,\nwe must uphold it. See Leslie v. Comm’r, 146 F.3d 643, 646\n(9th Cir. 1998).\n\n Where the Tax Court sustains an accuracy-related\npenalty, we review for clear error the Tax Court’s finding of\nnegligence. Hansen v. Comm’r, 471 F.3d 1021, 1028 (9th\nCir. 2006). We review de novo whether substantial authority\nsupported the taxpayer’s position, but we review for clear\nerror whether the taxpayer acted with reasonable cause and\nin good faith. DJB Holding Corp., 803 F.3d at 1022.\n\n III.\n\n The Estate challenges the Tax Court’s final decision on\nthree grounds. First, it argues that the Tax Court erred by\ntaking into account events that occurred after Victoria’s\ndeath in determining the value of the charitable deduction.\nInstead, the Estate argues that the charitable deduction\nshould have been valued as of Victoria’s date of death.\nSecond, the Estate argues that even if post-death events\ncould be considered, the Tax Court erred by not accounting\nfor a decline in value of Victoria’s shares caused by\neconomic forces. Third, the Estate argues that the Tax Court\nerred by upholding the accuracy-related penalty under I.R.C.\n§ 6662. We reject all three arguments.\n\n A.\n\n At the center of the parties’ dispute is whether the\nEstate’s charitable deduction should be valued at the time of\n\f12 DIERINGER V. CIR\n\nVictoria’s death, or whether the post-death events that\ndecreased value of the property delivered to charity should\nbe considered.\n\nThe Estate Tax and Charitable Deductions\n\n It is well established that the “estate tax is a tax on the\nprivilege of transfer[r]ing property” after one’s death.\nPropstra v. United States, 680 F.2d 1248, 1250 (9th Cir.\n1982). The estate tax “is on the act of the testator not on the\nreceipt of the property by the legatees.” Estate of Simplot v.\nComm’r, 249 F.3d 1191, 1194 (9th Cir. 2001) (citing Ithaca\nTr. Co. v. United States, 279 U.S. 151, 155 (1929)).\n\n Because the estate tax is a tax on the decedent’s bequest\nof property, the valuation of the gross estate is typically done\nas of the date of death. See I.R.C. § 2031; Tr. Servs. of Am.,\nInc. v. United States, 885 F.2d 561, 568–69 (9th Cir. 1989).\nThe pertinent statute provides that “[t]he value of the gross\nestate of the decedent shall be determined by including to the\nextent provided for in this part, the value at the time of his\ndeath of all property . . . wherever situated.” I.R.C.\n§ 2031(a); see also id. § 2033 (mandating that property in\nwhich the decedent had an interest be valued on the date of\ndecedent’s death); Treas. Reg. § 20.2031-1(b) (requiring the\nvaluation of property at the “fair market value at the time of\nthe decedent’s death” unless an exception applies). Except\nin some limited circumstances, such as when the executor\nelects to use an alternative valuation under I.R.C. § 2032,\npost-death events are generally not considered in\ndetermining the estate’s gross value for purposes of the\nestate tax. The parties agree that the executor did not use the\nalternative valuation method in this case.\n\n A related provision allows for deductions from the value\nof the gross estate for transfers of assets to qualified\n\f DIERINGER V. CIR 13\n\ncharitable entities. I.R.C. § 2055(a). This deduction\ngenerally is allowed “for the value of property included in\nthe decedent’s gross estate and transferred by the decedent\n. . . by will.” Treas. Reg. § 20.2055-1(a). Congress’s\nallowance for a charitable deduction was a “liberalization[]\nof the law in the taxpayer’s favor.” Helvering v. Bliss,\n293 U.S. 144, 151 (1934). Yet, as we have recognized,\ndeductions are acts of “legislative grace.” Comm’r v.\nShoong, 177 F.2d 131, 132 (9th Cir. 1949). “[T]he purpose\nof the charitable deduction [is] to encourage gifts to charity.”\nAhmanson, 674 F.2d at 772; see also Underwood v. United\nStates, 407 F.2d 608, 610 (6th Cir. 1969) (“The purpose of\nallowing charitable deductions is to encourage testators to\nmake charitable bequests, not to permit executors and\nbeneficiaries to rewrite a will so as to achieve tax savings.”).\n\nValuing the Charitable Deduction\n\n Deductions are valued separately from the valuation of\nthe gross estate. See Ahmanson, 674 F.2d at 772 (“The\nstatute does not ordain equal valuation as between an item in\nthe gross estate and the same item under the charitable\ndeduction.”). Separate valuations allow for the\nconsideration of post-death events, as required by Ahmanson\nand provisions of the tax code.\n\n We addressed valuation of a charitable deduction in\nAhmanson. There, the decedent’s estate plan provided for\nthe voting shares in a corporation to be left to family\nmembers and the nonvoting shares to be left to a charitable\nfoundation. Id. at 765–66. We held that when valuing the\ncharitable deduction for the nonvoting shares, a discount\nshould be applied to account for the fact that the shares\ndonated to charity had been stripped of their voting power.\nId. at 772. That a discount was not applied to the value of\nthe nonvoting shares in the gross estate did not impact our\n\f14 DIERINGER V. CIR\n\nholding. Id. Importantly, we recognized that a charitable\ndeduction “is subject to the principle that the testator may\nonly be allowed a deduction for estate tax purposes for what\nis actually received by the charity.” Id.\n\n In contrast, the Estate argues the charitable deduction\nmust be valued as of the date of Victoria’s death, in keeping\nwith the date-of-death valuation of an estate. We disagree.\nAlthough the Supreme Court and our court have applied that\nvaluation method where the remainder of an estate is\ndonated to charity after a post-death contingency, there is no\nuniform rule for all circumstances. Ithaca Trust, 279 U.S. at\n154; Wells Fargo Bank & Union Tr. Co. v. Comm’r,\n145 F.2d 132 (9th Cir. 1944). In Ithaca Trust, the decedent\nleft the remainder of his estate to his wife for her life, with\nany assets remaining after her death to be donated to charity.\n279 U.S. at 154. The decedent’s wife unexpectedly died\nonly six months after the decedent. Id. at 155. The Court\nconsidered whether the charitable deduction should reflect\nthe amount actually given to charity after the wife’s\npremature death, but ultimately determined that the\ndeduction should instead reflect a valuation based on\nmortality tables for the wife on the date of the decedent’s\ndeath. Id.\n\n We applied Ithaca Trust in Wells Fargo, where the\ndecedent had created a trust with instructions that the trust’s\nincome be paid to the decedent’s sister from the date of the\ndecedent’s death to the date of the sister’s death, and after\nthe sister’s death that the corpus of the trust be given to\ncharity. 145 F.2d at 133. Due to commingling of assets after\nthe decedent’s death and an unrelated tax dispute, part of the\ncorpus of the trust was used to support the decedent’s sister,\nwhich reduced the amount that went to charity upon her\ndeath. Id. The Tax Court relied on these payments to\n\f DIERINGER V. CIR 15\n\ndisallow part of the charitable deduction, but we reversed,\nconcluding that the charitable deduction must be determined\nfrom data available at the time of death. Id.; see also Estate\nof Van Horne v. Comm’r, 720 F.2d 1114, 1117 (9th Cir.\n1983) (relying on Ithaca Trust to “hold that legally\nenforceable claims valued by reference to an actuarial table\nmeet the test of certainty for estate tax purposes” when\nvaluing spousal support obligation).\n\n Neither Ithaca Trust nor Wells Fargo, however, set in\nstone the date of death as the date of the valuation of assets\nfor purposes of a deduction. See Shedd’s Estate v. Comm’r,\n320 F.2d 638, 639 (9th Cir. 1963) (“Congress did not intend\nto make events at the date of death invariably determinative\nin computing the federal estate tax obligation.”). Nor could\nthey. Certain deductions not only permit consideration of\npost-death events, but require them. For example, I.R.C.\n§ 2053(a) authorizes a deduction for funeral expenses and\nestate administration expenses—costs that cannot accrue\nuntil after the death of the testator. Similarly, I.R.C.\n§ 2055(c) specifies that where death taxes are payable out of\na charitable bequest, any charitable deduction is limited to\nthe value remaining in the estate after such post-death tax\npayment. Still another provision of the tax code, I.R.C.\n§ 2055(d), prohibits the amount of a charitable deduction\nfrom exceeding the value of transferred property included in\na gross estate—but, by negative implication, permits such a\ndeduction to be lower than the value of donated assets at the\nmoment of death. The Third Circuit also recognized that\nvaluations of the gross estate and a charitable deduction are\nseparate and may differ. In Re Sage’s Estate, 122 F.2d 480,\n484 (3d Cir. 1941) (“[W]hile a decedent’s gross estate is\nfixed as of the date of his death, deductions claimed in\ndetermining the net estate subject to tax may not be\n\f16 DIERINGER V. CIR\n\nascertainable or even accrue until the happening of events\nsubsequent to death.”).\n\n Indeed, “[t]he proper administration of the charitable\ndeduction cannot ignore such differences in the value\nactually received by the charity.” Ahmanson, 674 F.2d at\n772; accord Irving Tr. Co. v. United States, 221 F.2d 303,\n306 (2d Cir. 1955); Thompson’s Estate v. Comm’r, 123 F.2d\n816, 817 (2d Cir. 1941); In Re Sage’s Estate, 122 F.2d at\n484. This rule prohibits crafting an estate plan or will so as\nto game the system and guarantee a charitable deduction that\nis larger than the amount actually given to charity.\nAhmanson, 674 F.2d at 772.\n\nApplying Ahmanson’s Rule\n\n Ahmanson compels affirming the Tax Court’s ruling\nhere. Victoria structured her Estate so as not to donate her\nDPI shares directly to a charity, or even directly to the\nFoundation, but to the Trust. Victoria enabled Eugene to\ncommit almost unchecked abuse of the Estate by setting him\nup to be executor of the Estate, trustee of the Trust, and\ntrustee of the Foundation, in addition to his roles as\npresident, director, and majority shareholder of DPI. As the\nTax Court found, Eugene improperly directed Lewis Olds to\ndetermine the redemption value of the DPI shares by\napplying a minority interest valuation, when he knew a\nmajority interest applied and the Estate had claimed a\ncharitable deduction based upon a majority interest\nvaluation. Through his actions, Eugene manipulated the\ncharitable deduction so that the Foundation only received a\nfraction of the charitable deduction claimed by the Estate.\n\n The Estate attempts to evade Ahmanson by arguing that\nit is limited to situations where the testamentary plan\ndiminishes the value of the charitable property. Read in\n\f DIERINGER V. CIR 17\n\ncontext, Ahmanson is not limited to abuses in the four\ncorners of the testamentary plan. Id. Ahmanson extends to\nsituations where “the testator would be able to produce an\nartificially low valuation by manipulat[ion],” which includes\nthe present situation. Id. Moreover, Victoria’s testamentary\nplan laid the groundwork for Eugene’s manipulation by\nconcentrating power in his hands—in his roles as executor\nof the Estate and trustee of the Trust and Foundation—even\nafter she knew of and assented to early discussions of the\nshare redemption plan.\n\n The Tax Court correctly considered the difference\nbetween the deduction and the property actually received by\nthe charity due to Eugene’s manipulation of the redemption\nappraisal value. 10\n\n B.\n\n The Estate argues that any consideration of post-death\nevents also requires finding that the decline in value of DPI\nstock was due, at least in part, to market forces. The Tax\nCourt, however, found that “[t]he evidence does not support\na significant decline in the economy that resulted in a large\ndecrease in value in only seven months.” The Tax Court\nacknowledged that the adjusted net asset value of the DPI\nstock decreased from $17,777,626 at the date of Victoria’s\ndeath to $16,159,167 at the date used for the redemption\nappraisal. 11 Still, the Tax Court found, “[t]he reported\n\n 10\n The Commissioner argues in the alternative that the events in this\ncase fall under the exception in Treas. Reg. § 20.2055-2(b)(1). In light\nof our holding that Ahmanson compels affirmance, we need not address\nthis argument.\n 11\n The Estate suggests that this change in adjusted net asset value is\nentirely attributable to a changed real estate market. This argument\n\f18 DIERINGER V. CIR\n\ndecline in per share value was primarily due to the specific\ninstruction to value decedent’s majority interest as a\nminority interest with a 50% discount.” We find nothing in\nthe record—nor does the Estate point to anything in the\nrecord—that suggests the Tax Court’s findings were clearly\nerroneous.\n\n C.\n\n Finally, the Estate challenges the accuracy-related\npenalty that was imposed pursuant to I.R.C. § 6662(a).\nSection 6662(a) imposes a penalty of 20 percent of the\namount of the underpayment attributable to either\nnegligence or disregard of rules or regulations. A penalty\nwill not be imposed if the taxpayer can show “that there was\na reasonable cause [for the underpayment] and that the\ntaxpayer acted in good faith.” I.R.C. § 6664(c)(1). The\nCommissioner bears the initial burden of production\nregarding the applicability of the penalty, but once this\nburden is met, the taxpayer bears the burden of persuasion\nas to defenses to the penalty. Id. § 7491(c); Higbee v.\nComm’r, 116 T.C. 438, 446–47 (2001).\n\n The Tax Court first found that the Estate acted\nnegligently. Section 6662(c) defines negligence as “any\nfailure to make a reasonable attempt to comply” with the\nI.R.C. See also Allen v. Comm’r, 925 F.2d 348, 353 (9th Cir.\n1991) (defining negligence as a “lack of due care or the\nfailure to do what a reasonable and prudent person would do\nunder similar circumstances”). The Tax Court found that the\nEstate was negligent because it “not only failed to inform the\n\n\nmisses the point. The Estate claimed an $18,205,476 charitable\ndeduction largely based on the value of assets earmarked for the\nFoundation—but the Foundation only received assets worth $6,434,578.\n\f DIERINGER V. CIR 19\n\nappraiser that the redemption was for a majority interest, but\nalso instructed the appraiser to value the redeemed DPI stock\nas a minority interest.” We agree; the Tax Court did not\nclearly err in finding negligence.\n\n The Tax Court next found that the Estate did not have\nreasonable cause and good faith for its negligent act. “The\ndetermination of whether a taxpayer acted with reasonable\ncause and in good faith is made on a case-by-case basis,\ntaking into account all pertinent facts and circumstances.”\nTreas. Reg. § 1.6664-4(b)(1). The most important factor “is\nthe extent of the taxpayer’s effort to assess the taxpayer’s\nproper tax liability.” Id. A reliance on professional advice\nmay be used to show reasonable cause and good faith, but\nonly if the professional advice meets certain requirements.\nId. (instructing that “the circumstances under which the\nappraisal was obtained” are considered in determining\nwhether there was reasonable cause and good faith).\n\n The record evidence supports the Tax Court’s finding\nthat the Estate did not rely in good faith on its attorney’s\njudgment. The redemption appraisal was inaccurate because\nof instructions from Eugene on behalf of the Estate.\nTherefore, the Estate “knew” that Eugene and his brothers\nwere acquiring the DPI stock at a discount. Further, the\nappraiser’s credible testimony that he was instructed to\nundertake a minority interest valuation, which he ordinarily\nwould not have done in this situation, demonstrates the lack\nof good faith. See id. On these facts we find no error in the\nTax Court’s holding that the Commissioner properly\nimposed the accuracy-related penalty under I.R.C.\n§ 6662(a).\n\f20 DIERINGER V. CIR\n\n IV.\n\n The Estate claimed an $18,205,476 charitable deduction\nbased on the value of assets earmarked for the Foundation,\nbut the Foundation received assets worth only $6,434,578.\nThe Estate, at Eugene’s direction, claimed a large charitable\ndeduction representing that such assets would be delivered\nto the Foundation. Eugene then revalued and delivered\nassets—promissory notes and DPI shares—worth far less\nthan the claimed charitable deduction. By doing so, the\nEstate—and various Dieringer family entities—ended up\navoiding $4.1 million in taxes. We agree with the Tax\nCourt’s decision upholding the Commissioner’s reduction of\nthe Estate’s charitable deduction and the deficiency\nassessment. To hold otherwise would only “invite abuse.”\nAhmanson, 674 F.2d at 768.\n\n AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Victoria Dieringer v. Cir\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-12\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n VICTORIA E. DIERINGER, Deceased; No. 16-72640\\n EUGENE DIERINGER, Executor,\\n Petitioners-Appellants, Tax Ct. No.\\n 21992-13\\n v.\\n\\n COMMISSIONER OF INTERNAL OPINION\\n REVENUE,\\n Respondent-Appellee.\\n\\n\\n\\n Appeal from a Decision of the\\n United States Tax Court\\n\\n Argued and Submitted May 15, 2018\\n Portland, Oregon\\n\\n Filed March 12, 2019\\n\\n Before: M. Margaret McKeown and Richard A. Paez,\\n Circuit Judges, and Cynthia A. Bashant, * District Judge.\\n\\n Opinion by Judge Paez\\n\\n\\n\\n *\\n The Honorable Cynthia A. Bashant, United States District Judge\\nfor the Southern District of California, sitting by designation.\\n\\f2 DIERINGER V. CIR\\n\\n SUMMARY **\\n\\n\\n Tax\\n\\n The panel affirmed the Tax Court’s decision sustaining\\na deficiency against an estate for overstating the amount of\\na charitable deduction, and sustaining an accuracy-related\\npenalty.\\n\\n Petitioner, the estate executor and heir of the decedent,\\ndeclared a large charitable deduction based on the value of\\ncertain estate property at the time of death. The charity\\nreceived less than the amount of the claimed deduction, and\\nthe estate received a tax windfall in the process. Petitioner\\ncontended that the Tax Court should have taken into account\\nevents that occurred after the decedent’s death in\\ndetermining the value of the charitable deduction.\\n\\n In light of this court’s holding in Ahmanson Foundation\\nv. United States, which underscored “the principle that the\\ntestator may only be allowed a deduction for estate tax\\npurposes for what is actually received by the charity,”\\n674 F.2d 761, 772 (9th Cir. 1981), the panel affirmed the\\nTax Court’s decision upholding the reduction of the\\ncharitable deduction and the deficiency assessment. The\\npanel found no clear error in the Tax Court’s finding that\\nthere was no evidence of a significant decline in the\\neconomy that would have decreased the value of the\\nproperty being donated. The panel also found no error in the\\nTax Court’s upholding of the accuracy-related penalty.\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f DIERINGER V. CIR 3\\n\\n COUNSEL\\n\\nW. Michael Gillette (argued), Sara Kobak, and Marc K.\\nSellers, Schwabe Williamson & Wyatt, P.C., Portland,\\nOregon, for Petitioners-Appellants.\\n\\nDouglas Campbell Rennie (argued), Jennifer M. Rubin, and\\nJoan I. Oppenheiner, Attorneys; David A. Hubbert, Acting\\nAssistant Attorney General; Tax Division, United States\\nDepartment of Justice, Washington, D.C.; for Respondent-\\nAppellee.\\n\\n\\n OPINION\\n\\nPAEZ, Circuit Judge:\\n\\n I.\\n\\n In this estate tax case, the Internal Revenue Service\\nCommissioner (“Commissioner”) assessed a deficiency\\nagainst the decedent’s estate for overstating the amount of a\\ncharitable contribution. The estate executor and heir\\ndeclared a large charitable deduction based on the value of\\nestate property at the time of death, only to manipulate the\\nproperty for personal gain, deliver assets to the charity worth\\nsubstantially less than those claimed as a deduction, and\\nreceive a tax windfall in the process. The Tax Court\\nsustained the deficiency. We affirm in light of our holding\\nin Ahmanson Foundation v. United States, where we\\nunderscored “the principle that the testator may only be\\nallowed a deduction for estate tax purposes for what is\\nactually received by the charity.” 674 F.2d 761, 772 (9th\\nCir. 1981). We also affirm the Tax Court’s ruling sustaining\\nan accuracy-related penalty under I.R.C. § 6662.\\n\\f4 DIERINGER V. CIR\\n\\n A.\\n\\n This case involves the Victoria E. Dieringer Estate\\n(“Estate”). Victoria and her late husband Robert E.\\nDieringer (“Robert”) had twelve children together, including\\nEugene Dieringer (“Eugene”), Patrick Dieringer (“Patrick”),\\nand Timothy Dieringer (“Timothy”).\\n\\n The Dieringer family owns Dieringer Properties, Inc.\\n(“DPI”), a closely held corporation that manages\\ncommercial and residential properties in Portland, Oregon,\\nas well as a Wendy’s restaurant in Texas. After Robert\\npassed away and before Victoria’s death, Eugene was the\\npresident of DPI, Patrick was the executive vice-president\\nand secretary, Victoria was the vice-president, and Timothy\\nwas the office manager. DPI’s Board of Directors (“Board”)\\nconsisted of Victoria as chairperson, and Eugene, Patrick,\\nTimothy, and Thomas Keepes (“Keepes”), who is unrelated\\nto the Dieringers, as directors. None of the other nine\\nDieringer children were involved in DPI in any capacity.\\n\\n Before Victoria’s death, the only shareholders of DPI\\nwere Victoria, Eugene, and Patrick. Victoria owned 425 out\\nof 525 voting shares and 7,736.5 out of 9,220.5 nonvoting\\nshares. Eugene owned the remaining 100 voting shares, and\\nEugene and Patrick each owned 742 nonvoting shares.\\n\\n According to Victoria’s will, dated November 10, 2000,\\nupon her death all of her Estate would pass to the Victoria\\nEvelyn Dieringer Trust (“Trust”). The Trust, as amended on\\nApril 22, 2005, provided for Victoria’s children to receive\\nsome personal effects, but no other proceeds from her estate.\\nThe Trust also provided for $600,000 in donations to various\\ncharitable organizations. Any assets remaining in the Estate\\nwould then pass to the Bob and Evelyn Dieringer Family\\nFoundation (“Foundation”), an I.R.C. § 501(c)(3)\\n\\f DIERINGER V. CIR 5\\n\\norganization, as a charitable contribution. Under Victoria’s\\nestate plan, Eugene was appointed as the sole trustee of both\\nthe Trust and the Foundation. Patrick and Keepes served as\\nadvisory trustees of the Foundation.\\n\\n Prior to Victoria’s death, the DPI Board had preliminary\\ndiscussions about purchasing Victoria’s DPI shares as part\\nof ongoing succession planning. A November 24, 2008\\nBoard resolution reported that the issue had been discussed\\nand that the Board resolved to “periodically purchase”\\nVictoria’s shares based on terms acceptable to all parties.\\nVictoria was “completely agreeable” to such a plan. At a\\nFebruary 13, 2009 Board meeting, Victoria reiterated her\\npotential interest in having DPI purchase her shares. In\\nanticipation of entering a purchase agreement for Victoria’s\\nshares, DPI paid the Trust $45,000 prior to Victoria’s death.\\nWhen Victoria unexpectedly died on April 14, 2009, there\\nwere no specific redemption agreements in place. 1\\n\\n When Victoria died, Eugene was appointed executor of\\nthe Estate. To determine the value of Victoria’s DPI shares\\nfor Estate administration purposes, the Estate’s law firm\\nrequested that Lewis Olds & Associates perform an\\nindependent appraisal of the net asset value of DPI. The\\nappraisal determined that the value of DPI as of the date of\\n\\n\\n\\n\\n 1\\n In addition to the Board resolutions that reflect conversations about\\nredemption of Victoria’s stock in 2008 and 2009, the record reflects that\\nas early as 2000 Victoria and Robert contemplated a buy/sell agreement\\nthat would require DPI or Eugene to purchase all of Victoria or Robert’s\\nshares upon the later death of the two of them.\\n\\f6 DIERINGER V. CIR\\n\\nVictoria’s death was $17,777,626 and that Victoria’s shares\\nin DPI were worth $14,182,471. 2\\n\\n Effective November 20, 2009, DPI’s Board converted\\nthe corporate structure from a C corporation to an\\nS corporation on the advice of Keepes. The DPI Board made\\nthis change in corporate structure to accomplish its long-\\nterm tax planning goals and to avoid certain adverse tax\\nconsequences under I.R.C. § 1374. As a result of electing\\nS corporation status, the DPI Board decided that it should\\nalso redeem Victoria’s DPI shares that were to pass to the\\nFoundation. 3\\n\\n The Board entered into an agreement with the Trust to\\nredeem Victoria’s shares, prior to the shares “pouring over”\\ninto the Foundation. Initially, DPI agreed to redeem all of\\nVictoria’s shares for $6,071,558, effective November 30,\\n2009. This amount was based on a 2002 appraisal, since the\\ndate-of-death appraisal had not yet been completed. As a\\nresult, the redemption agreement provided that the stated\\nprice would be “reconciled and adjusted retroactively” to\\nreflect the fair market value of the shares as of November 30,\\n2009. DPI executed two promissory notes payable to the\\n 2\\n Victoria’s voting shares had a market value of $1,824.19 per share\\nand her nonvoting shares had a market value of $1,732.97 per share.\\n 3\\n The DPI Board put forth a number of business rationales for the\\ndecision to redeem the Trust’s DPI shares. These included concerns\\nabout the tax consequences of the Foundation owning shares in an\\nS corporation, that the DPI shares did not provide enough liquidity for\\nthe Foundation to distribute five percent of its funds annually as required\\nby I.R.C. § 4942, and that freezing the value of the Foundation’s DPI\\nshares into a promissory note could prevent future decline in the value\\nof DPI shares given the poor economic climate. The Tax Court did not\\nquestion any of these rationales.\\n\\f DIERINGER V. CIR 7\\n\\nTrust in exchange for the DPI shares. 4 Eugene, Patrick, and\\nTimothy entered into separate subscription agreements to\\npurchase additional DPI shares, in order to provide funding\\nfor DPI to meet the required payments on the promissory\\nnotes. 5 At the later Tax Court trial, the court found that there\\nwere appropriate business purposes for these subscription\\nagreements.\\n\\n At the direction of DPI, Lewis Olds & Associates\\nperformed another appraisal of the value of Victoria’s DPI\\nshares as of November 30, 2009, for the purpose of\\nredemption (“redemption appraisal”). The redemption\\nappraisal valued Victoria’s DPI shares at $916 per voting\\nshare and $870 per nonvoting share. Lewis Olds testified,\\nand the Tax Court found, that Eugene (through his lawyer)\\ninstructed him to value Victoria’s DPI shares as if they were\\na minority interest in DPI for the purposes of this appraisal,\\nand that he would not have done so without these\\ninstructions. Olds’s appraisal therefore included a\\n15-percent discount for lack of control and a 35-percent\\ndiscount for lack of marketability. As a result, Victoria’s\\n\\n\\n 4\\n One of the promissory notes was a short-term note for $2,250,000.\\nThe other was a long-term note for $3,776,558. Each note was payable\\nto the Trust, included interest, and was retroactively adjustable\\ndepending on the new appraisal value.\\n 5\\n These separate subscription agreements provided that voting\\nshares would cost $779 per share and nonvoting shares would cost\\n$742 per share, subject to a retroactive price adjustment. Initially, the\\nsubscription agreements provided that Eugene would purchase\\n2,695 nonvoting shares; Patrick would purchase 86 voting shares and\\n2,695 nonvoting shares; and Timothy would purchase 25 voting shares\\nand 108 nonvoting shares. These amounts were later modified. See infra\\nfn. 7.\\n\\f8 DIERINGER V. CIR\\n\\nDPI shares were valued significantly less in the redemption\\nappraisal than in the date-of-death appraisal. 6\\n\\n DPI determined that it could not afford to redeem all of\\nVictoria’s shares at the new valuation price. The redemption\\nagreement was then amended, and consequently DPI\\nredeemed all 425 voting shares and 5,600.5 nonvoting shares\\nfor a total purchase price of $5,263,462. 7 The balance on the\\nlong-term promissory note was amended to reflect the\\nchanges in the number and price of shares under the\\namended redemption agreement. After the redemption\\nagreement was implemented, the distribution of DPI shares\\nwas as follows: (1) the Trust owned 2,136 nonvoting shares;\\n(2) Eugene owned 200 voting shares and 2,932 nonvoting\\nshares; (3) Patrick owned 65 voting shares and\\n893 nonvoting shares; and (4) Timothy owned 25 voting\\nshares and 108 nonvoting shares. 8\\n\\n In January 2011, the Trust distributed the promissory\\nnotes and the remaining DPI shares to the Foundation. The\\n\\n 6\\n Lewis Olds & Associates worked on producing the appraisals\\nsimultaneously; they are dated March 24, 2010, and March 25, 2010.\\n\\n 7\\n Correspondingly, the subscription agreements also were modified.\\nUnder the modified agreements, Eugene purchased 100 voting shares\\nand 2,190 nonvoting shares, Patrick purchased 65 voting shares and\\n86 nonvoting shares, and Timothy purchased 25 voting shares and\\n108 nonvoting shares.\\n\\n 8\\n There are minor discrepancies between the number of shares\\nindicated in the parties’ stipulation and those indicated in the Tax Court’s\\nopinion. For instance, the number of nonvoting shares owned by Patrick\\ndiffers between the parties’ stipulation (893) and the Tax Court’s opinion\\n(828). We use the number of shares to which the parties stipulated, but\\nthe analysis is the same regardless.\\n\\f DIERINGER V. CIR 9\\n\\nstate probate court approved the redemption agreement and\\nindicated that the redemption agreement and promissory\\nnotes would not be prohibited self-dealing under I.R.C.\\n§ 4941. 9 For the 2011 tax year, the Foundation reported the\\nfollowing contributions on its Form 990-PF, Return of\\nPrivate Foundation: (1) a noncash contribution of DPI shares\\nwith a fair market value of $1,858,961; (2) a long-term note\\nreceivable with a fair market value of $2,921,312; and (3) a\\nshort-term note receivable with a fair market value of\\n$2,250,000. The Trust reported a capital loss of $385,934\\nfor the sale of the 425 voting shares and $4,831,439 for the\\nsale of 5,600.5 nonvoting shares for the taxable year ending\\nin December 31, 2009, on its Form 1041 Tax Return. The\\nEstate filed its Form 706, United States Estate (and\\nGeneration-Skipping Transfer) Tax Return on July 12, 2010,\\nreporting no estate tax liability. The Estate claimed a\\ncharitable contribution deduction of $18,812,181, based on\\nthe date-of-death value of Victoria’s DPI shares.\\n\\n B.\\n\\n In June 2013, the Commissioner issued a notice of\\ndeficiency to the Estate based on its July 2010 tax return\\n(Form 706). The notice stated that there was a deficiency of\\n$4,124,717 and imposed an accuracy-related penalty of\\n$824,943 under I.R.C. § 6662 for error and negligence in\\nusing the date-of-death appraisal as the value of the\\ncharitable contribution of Victoria’s DPI shares.\\n\\n Upon receiving the notice of deficiency, the Estate filed\\na timely petition in the Tax Court challenging the\\n\\n 9\\n I.R.C. § 4941 allows the Commissioner to impose a tax on certain\\ntransactions between a private foundation and a “disqualified person” as\\ndefined by I.R.C. § 4946(a)(1).\\n\\f10 DIERINGER V. CIR\\n\\nCommissioner’s deficiency notice and penalty assessment.\\nThe Estate argued that it correctly used the date-of-death\\nappraisal to determine the value of Victoria’s DPI shares for\\nthe purpose of the charitable contribution deduction. The\\nCommissioner responded that post-death events should be\\nconsidered in determining the value of the charitable\\ncontribution, as the actions by Eugene, Patrick, and Timothy\\nreduced the value of Victoria’s contribution to the\\nFoundation.\\n\\n Following trial, the Tax Court issued a decision\\nupholding the Commissioner’s reduction of the Estate’s\\ncharitable deduction and the deficiency assessment. The\\nTax Court found that the redemption was not part of\\nVictoria’s estate plan and there were valid business reasons\\nfor many of the transactions that took place after her death.\\nThe Tax Court also concluded, however, that post-death\\nevents—primarily Eugene’s decision to apply a minority\\ninterest discount to the redemption value of Victoria’s DPI\\nshares—reduced the value of the contribution to the\\nFoundation and therefore reduced the value of the Estate’s\\ncharitable deduction. But the evidence did not support the\\nconclusion that a poor business climate caused the reported\\ndecline in share values, as the Estate argued. Further, the\\nTax Court affirmed the Commissioner’s determination that\\nthe Estate was liable for a penalty under I.R.C. § 6662(a).\\nAfter denying the Estate’s motion for reconsideration, the\\nTax Court entered a final decision in favor of the\\nCommissioner, which sustained an estate tax deficiency of\\n$4,124,717 and an accuracy-related penalty of $824,943.\\nThe Estate timely appealed.\\n\\n II.\\n\\n We review de novo the Tax Court’s legal conclusions,\\nsee DJB Holding Corp. v. Comm’r, 803 F.3d 1014, 1022 (9th\\n\\f DIERINGER V. CIR 11\\n\\nCir. 2015), as well as the Tax Court’s interpretation of the\\nInternal Revenue Code, see Metro One Telecomms., Inc. v.\\nComm’r, 704 F.3d 1057, 1059 (9th Cir. 2012). We review\\nfor clear error the Tax Court’s factual findings. See DJB\\nHolding Corp., 803 F.3d at 1022. Under clear error review,\\nif the Tax Court’s interpretation of the evidence is plausible,\\nwe must uphold it. See Leslie v. Comm’r, 146 F.3d 643, 646\\n(9th Cir. 1998).\\n\\n Where the Tax Court sustains an accuracy-related\\npenalty, we review for clear error the Tax Court’s finding of\\nnegligence. Hansen v. Comm’r, 471 F.3d 1021, 1028 (9th\\nCir. 2006). We review de novo whether substantial authority\\nsupported the taxpayer’s position, but we review for clear\\nerror whether the taxpayer acted with reasonable cause and\\nin good faith. DJB Holding Corp., 803 F.3d at 1022.\\n\\n III.\\n\\n The Estate challenges the Tax Court’s final decision on\\nthree grounds. First, it argues that the Tax Court erred by\\ntaking into account events that occurred after Victoria’s\\ndeath in determining the value of the charitable deduction.\\nInstead, the Estate argues that the charitable deduction\\nshould have been valued as of Victoria’s date of death.\\nSecond, the Estate argues that even if post-death events\\ncould be considered, the Tax Court erred by not accounting\\nfor a decline in value of Victoria’s shares caused by\\neconomic forces. Third, the Estate argues that the Tax Court\\nerred by upholding the accuracy-related penalty under I.R.C.\\n§ 6662. We reject all three arguments.\\n\\n A.\\n\\n At the center of the parties’ dispute is whether the\\nEstate’s charitable deduction should be valued at the time of\\n\\f12 DIERINGER V. CIR\\n\\nVictoria’s death, or whether the post-death events that\\ndecreased value of the property delivered to charity should\\nbe considered.\\n\\nThe Estate Tax and Charitable Deductions\\n\\n It is well established that the “estate tax is a tax on the\\nprivilege of transfer[r]ing property” after one’s death.\\nPropstra v. United States, 680 F.2d 1248, 1250 (9th Cir.\\n1982). The estate tax “is on the act of the testator not on the\\nreceipt of the property by the legatees.” Estate of Simplot v.\\nComm’r, 249 F.3d 1191, 1194 (9th Cir. 2001) (citing Ithaca\\nTr. Co. v. United States, 279 U.S. 151, 155 (1929)).\\n\\n Because the estate tax is a tax on the decedent’s bequest\\nof property, the valuation of the gross estate is typically done\\nas of the date of death. See I.R.C. § 2031; Tr. Servs. of Am.,\\nInc. v. United States, 885 F.2d 561, 568–69 (9th Cir. 1989).\\nThe pertinent statute provides that “[t]he value of the gross\\nestate of the decedent shall be determined by including to the\\nextent provided for in this part, the value at the time of his\\ndeath of all property . . . wherever situated.” I.R.C.\\n§ 2031(a); see also id. § 2033 (mandating that property in\\nwhich the decedent had an interest be valued on the date of\\ndecedent’s death); Treas. Reg. § 20.2031-1(b) (requiring the\\nvaluation of property at the “fair market value at the time of\\nthe decedent’s death” unless an exception applies). Except\\nin some limited circumstances, such as when the executor\\nelects to use an alternative valuation under I.R.C. § 2032,\\npost-death events are generally not considered in\\ndetermining the estate’s gross value for purposes of the\\nestate tax. The parties agree that the executor did not use the\\nalternative valuation method in this case.\\n\\n A related provision allows for deductions from the value\\nof the gross estate for transfers of assets to qualified\\n\\f DIERINGER V. CIR 13\\n\\ncharitable entities. I.R.C. § 2055(a). This deduction\\ngenerally is allowed “for the value of property included in\\nthe decedent’s gross estate and transferred by the decedent\\n. . . by will.” Treas. Reg. § 20.2055-1(a). Congress’s\\nallowance for a charitable deduction was a “liberalization[]\\nof the law in the taxpayer’s favor.” Helvering v. Bliss,\\n293 U.S. 144, 151 (1934). Yet, as we have recognized,\\ndeductions are acts of “legislative grace.” Comm’r v.\\nShoong, 177 F.2d 131, 132 (9th Cir. 1949). “[T]he purpose\\nof the charitable deduction [is] to encourage gifts to charity.”\\nAhmanson, 674 F.2d at 772; see also Underwood v. United\\nStates, 407 F.2d 608, 610 (6th Cir. 1969) (“The purpose of\\nallowing charitable deductions is to encourage testators to\\nmake charitable bequests, not to permit executors and\\nbeneficiaries to rewrite a will so as to achieve tax savings.”).\\n\\nValuing the Charitable Deduction\\n\\n Deductions are valued separately from the valuation of\\nthe gross estate. See Ahmanson, 674 F.2d at 772 (“The\\nstatute does not ordain equal valuation as between an item in\\nthe gross estate and the same item under the charitable\\ndeduction.”). Separate valuations allow for the\\nconsideration of post-death events, as required by Ahmanson\\nand provisions of the tax code.\\n\\n We addressed valuation of a charitable deduction in\\nAhmanson. There, the decedent’s estate plan provided for\\nthe voting shares in a corporation to be left to family\\nmembers and the nonvoting shares to be left to a charitable\\nfoundation. Id. at 765–66. We held that when valuing the\\ncharitable deduction for the nonvoting shares, a discount\\nshould be applied to account for the fact that the shares\\ndonated to charity had been stripped of their voting power.\\nId. at 772. That a discount was not applied to the value of\\nthe nonvoting shares in the gross estate did not impact our\\n\\f14 DIERINGER V. CIR\\n\\nholding. Id. Importantly, we recognized that a charitable\\ndeduction “is subject to the principle that the testator may\\nonly be allowed a deduction for estate tax purposes for what\\nis actually received by the charity.” Id.\\n\\n In contrast, the Estate argues the charitable deduction\\nmust be valued as of the date of Victoria’s death, in keeping\\nwith the date-of-death valuation of an estate. We disagree.\\nAlthough the Supreme Court and our court have applied that\\nvaluation method where the remainder of an estate is\\ndonated to charity after a post-death contingency, there is no\\nuniform rule for all circumstances. Ithaca Trust, 279 U.S. at\\n154; Wells Fargo Bank & Union Tr. Co. v. Comm’r,\\n145 F.2d 132 (9th Cir. 1944). In Ithaca Trust, the decedent\\nleft the remainder of his estate to his wife for her life, with\\nany assets remaining after her death to be donated to charity.\\n279 U.S. at 154. The decedent’s wife unexpectedly died\\nonly six months after the decedent. Id. at 155. The Court\\nconsidered whether the charitable deduction should reflect\\nthe amount actually given to charity after the wife’s\\npremature death, but ultimately determined that the\\ndeduction should instead reflect a valuation based on\\nmortality tables for the wife on the date of the decedent’s\\ndeath. Id.\\n\\n We applied Ithaca Trust in Wells Fargo, where the\\ndecedent had created a trust with instructions that the trust’s\\nincome be paid to the decedent’s sister from the date of the\\ndecedent’s death to the date of the sister’s death, and after\\nthe sister’s death that the corpus of the trust be given to\\ncharity. 145 F.2d at 133. Due to commingling of assets after\\nthe decedent’s death and an unrelated tax dispute, part of the\\ncorpus of the trust was used to support the decedent’s sister,\\nwhich reduced the amount that went to charity upon her\\ndeath. Id. The Tax Court relied on these payments to\\n\\f DIERINGER V. CIR 15\\n\\ndisallow part of the charitable deduction, but we reversed,\\nconcluding that the charitable deduction must be determined\\nfrom data available at the time of death. Id.; see also Estate\\nof Van Horne v. Comm’r, 720 F.2d 1114, 1117 (9th Cir.\\n1983) (relying on Ithaca Trust to “hold that legally\\nenforceable claims valued by reference to an actuarial table\\nmeet the test of certainty for estate tax purposes” when\\nvaluing spousal support obligation).\\n\\n Neither Ithaca Trust nor Wells Fargo, however, set in\\nstone the date of death as the date of the valuation of assets\\nfor purposes of a deduction. See Shedd’s Estate v. Comm’r,\\n320 F.2d 638, 639 (9th Cir. 1963) (“Congress did not intend\\nto make events at the date of death invariably determinative\\nin computing the federal estate tax obligation.”). Nor could\\nthey. Certain deductions not only permit consideration of\\npost-death events, but require them. For example, I.R.C.\\n§ 2053(a) authorizes a deduction for funeral expenses and\\nestate administration expenses—costs that cannot accrue\\nuntil after the death of the testator. Similarly, I.R.C.\\n§ 2055(c) specifies that where death taxes are payable out of\\na charitable bequest, any charitable deduction is limited to\\nthe value remaining in the estate after such post-death tax\\npayment. Still another provision of the tax code, I.R.C.\\n§ 2055(d), prohibits the amount of a charitable deduction\\nfrom exceeding the value of transferred property included in\\na gross estate—but, by negative implication, permits such a\\ndeduction to be lower than the value of donated assets at the\\nmoment of death. The Third Circuit also recognized that\\nvaluations of the gross estate and a charitable deduction are\\nseparate and may differ. In Re Sage’s Estate, 122 F.2d 480,\\n484 (3d Cir. 1941) (“[W]hile a decedent’s gross estate is\\nfixed as of the date of his death, deductions claimed in\\ndetermining the net estate subject to tax may not be\\n\\f16 DIERINGER V. CIR\\n\\nascertainable or even accrue until the happening of events\\nsubsequent to death.”).\\n\\n Indeed, “[t]he proper administration of the charitable\\ndeduction cannot ignore such differences in the value\\nactually received by the charity.” Ahmanson, 674 F.2d at\\n772; accord Irving Tr. Co. v. United States, 221 F.2d 303,\\n306 (2d Cir. 1955); Thompson’s Estate v. Comm’r, 123 F.2d\\n816, 817 (2d Cir. 1941); In Re Sage’s Estate, 122 F.2d at\\n484. This rule prohibits crafting an estate plan or will so as\\nto game the system and guarantee a charitable deduction that\\nis larger than the amount actually given to charity.\\nAhmanson, 674 F.2d at 772.\\n\\nApplying Ahmanson’s Rule\\n\\n Ahmanson compels affirming the Tax Court’s ruling\\nhere. Victoria structured her Estate so as not to donate her\\nDPI shares directly to a charity, or even directly to the\\nFoundation, but to the Trust. Victoria enabled Eugene to\\ncommit almost unchecked abuse of the Estate by setting him\\nup to be executor of the Estate, trustee of the Trust, and\\ntrustee of the Foundation, in addition to his roles as\\npresident, director, and majority shareholder of DPI. As the\\nTax Court found, Eugene improperly directed Lewis Olds to\\ndetermine the redemption value of the DPI shares by\\napplying a minority interest valuation, when he knew a\\nmajority interest applied and the Estate had claimed a\\ncharitable deduction based upon a majority interest\\nvaluation. Through his actions, Eugene manipulated the\\ncharitable deduction so that the Foundation only received a\\nfraction of the charitable deduction claimed by the Estate.\\n\\n The Estate attempts to evade Ahmanson by arguing that\\nit is limited to situations where the testamentary plan\\ndiminishes the value of the charitable property. Read in\\n\\f DIERINGER V. CIR 17\\n\\ncontext, Ahmanson is not limited to abuses in the four\\ncorners of the testamentary plan. Id. Ahmanson extends to\\nsituations where “the testator would be able to produce an\\nartificially low valuation by manipulat[ion],” which includes\\nthe present situation. Id. Moreover, Victoria’s testamentary\\nplan laid the groundwork for Eugene’s manipulation by\\nconcentrating power in his hands—in his roles as executor\\nof the Estate and trustee of the Trust and Foundation—even\\nafter she knew of and assented to early discussions of the\\nshare redemption plan.\\n\\n The Tax Court correctly considered the difference\\nbetween the deduction and the property actually received by\\nthe charity due to Eugene’s manipulation of the redemption\\nappraisal value. 10\\n\\n B.\\n\\n The Estate argues that any consideration of post-death\\nevents also requires finding that the decline in value of DPI\\nstock was due, at least in part, to market forces. The Tax\\nCourt, however, found that “[t]he evidence does not support\\na significant decline in the economy that resulted in a large\\ndecrease in value in only seven months.” The Tax Court\\nacknowledged that the adjusted net asset value of the DPI\\nstock decreased from $17,777,626 at the date of Victoria’s\\ndeath to $16,159,167 at the date used for the redemption\\nappraisal. 11 Still, the Tax Court found, “[t]he reported\\n\\n 10\\n The Commissioner argues in the alternative that the events in this\\ncase fall under the exception in Treas. Reg. § 20.2055-2(b)(1). In light\\nof our holding that Ahmanson compels affirmance, we need not address\\nthis argument.\\n 11\\n The Estate suggests that this change in adjusted net asset value is\\nentirely attributable to a changed real estate market. This argument\\n\\f18 DIERINGER V. CIR\\n\\ndecline in per share value was primarily due to the specific\\ninstruction to value decedent’s majority interest as a\\nminority interest with a 50% discount.” We find nothing in\\nthe record—nor does the Estate point to anything in the\\nrecord—that suggests the Tax Court’s findings were clearly\\nerroneous.\\n\\n C.\\n\\n Finally, the Estate challenges the accuracy-related\\npenalty that was imposed pursuant to I.R.C. § 6662(a).\\nSection 6662(a) imposes a penalty of 20 percent of the\\namount of the underpayment attributable to either\\nnegligence or disregard of rules or regulations. A penalty\\nwill not be imposed if the taxpayer can show “that there was\\na reasonable cause [for the underpayment] and that the\\ntaxpayer acted in good faith.” I.R.C. § 6664(c)(1). The\\nCommissioner bears the initial burden of production\\nregarding the applicability of the penalty, but once this\\nburden is met, the taxpayer bears the burden of persuasion\\nas to defenses to the penalty. Id. § 7491(c); Higbee v.\\nComm’r, 116 T.C. 438, 446–47 (2001).\\n\\n The Tax Court first found that the Estate acted\\nnegligently. Section 6662(c) defines negligence as “any\\nfailure to make a reasonable attempt to comply” with the\\nI.R.C. See also Allen v. Comm’r, 925 F.2d 348, 353 (9th Cir.\\n1991) (defining negligence as a “lack of due care or the\\nfailure to do what a reasonable and prudent person would do\\nunder similar circumstances”). The Tax Court found that the\\nEstate was negligent because it “not only failed to inform the\\n\\n\\nmisses the point. The Estate claimed an $18,205,476 charitable\\ndeduction largely based on the value of assets earmarked for the\\nFoundation—but the Foundation only received assets worth $6,434,578.\\n\\f DIERINGER V. CIR 19\\n\\nappraiser that the redemption was for a majority interest, but\\nalso instructed the appraiser to value the redeemed DPI stock\\nas a minority interest.” We agree; the Tax Court did not\\nclearly err in finding negligence.\\n\\n The Tax Court next found that the Estate did not have\\nreasonable cause and good faith for its negligent act. “The\\ndetermination of whether a taxpayer acted with reasonable\\ncause and in good faith is made on a case-by-case basis,\\ntaking into account all pertinent facts and circumstances.”\\nTreas. Reg. § 1.6664-4(b)(1). The most important factor “is\\nthe extent of the taxpayer’s effort to assess the taxpayer’s\\nproper tax liability.” Id. A reliance on professional advice\\nmay be used to show reasonable cause and good faith, but\\nonly if the professional advice meets certain requirements.\\nId. (instructing that “the circumstances under which the\\nappraisal was obtained” are considered in determining\\nwhether there was reasonable cause and good faith).\\n\\n The record evidence supports the Tax Court’s finding\\nthat the Estate did not rely in good faith on its attorney’s\\njudgment. The redemption appraisal was inaccurate because\\nof instructions from Eugene on behalf of the Estate.\\nTherefore, the Estate “knew” that Eugene and his brothers\\nwere acquiring the DPI stock at a discount. Further, the\\nappraiser’s credible testimony that he was instructed to\\nundertake a minority interest valuation, which he ordinarily\\nwould not have done in this situation, demonstrates the lack\\nof good faith. See id. On these facts we find no error in the\\nTax Court’s holding that the Commissioner properly\\nimposed the accuracy-related penalty under I.R.C.\\n§ 6662(a).\\n\\f20 DIERINGER V. CIR\\n\\n IV.\\n\\n The Estate claimed an $18,205,476 charitable deduction\\nbased on the value of assets earmarked for the Foundation,\\nbut the Foundation received assets worth only $6,434,578.\\nThe Estate, at Eugene’s direction, claimed a large charitable\\ndeduction representing that such assets would be delivered\\nto the Foundation. Eugene then revalued and delivered\\nassets—promissory notes and DPI shares—worth far less\\nthan the claimed charitable deduction. By doing so, the\\nEstate—and various Dieringer family entities—ended up\\navoiding $4.1 million in taxes. We agree with the Tax\\nCourt’s decision upholding the Commissioner’s reduction of\\nthe Estate’s charitable deduction and the deficiency\\nassessment. To hold otherwise would only “invite abuse.”\\nAhmanson, 674 F.2d at 768.\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1279,"cells":{"id":{"kind":"number","value":4598814,"string":"4,598,814"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. Kenneth Door"},"decision_date":{"kind":"string","value":"2019-03-12"},"docket_number":{"kind":"string","value":"17-30165"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Sidney R. Thomas, Chief Judge, and M. Margaret McKeown and Morgan Christen, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"CHRISTEN, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-30165\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 3:12-cr-05126-RBL-1\\n\\nKENNETH RANDALE DOOR,\\n Defendant-Appellant. OPINION\\n\\n\\n Appeal from the United States District Court\\n for the Western District of Washington\\n Ronald B. Leighton, District Judge, Presiding\\n\\n Argued and Submitted December 7, 2018\\n Seattle, Washington\\n\\n Filed March 12, 2019\\n\\nBefore: Sidney R. Thomas, Chief Judge, and M. Margaret\\n McKeown and Morgan Christen, Circuit Judges.\\n\\n Opinion by Judge Christen\\n\\f2 UNITED STATES V. DOOR\\n\\n SUMMARY*\\n\\n\\n Criminal Law\\n\\n Vacating a sentence and remanding for resentencing, the\\npanel held that, in setting the defendant’s base offense level\\nunder U.S.S.G. § 2K2.1(a)(2), his prior Washington State\\nconviction for felony harassment qualified as a crime of\\nviolence but his prior conviction for second degree assault did\\nnot.\\n\\n Consistent with United States v. Werle, 877 F.3d 879 (9th\\nCir. 2017), the panel held that the defendant’s 1997\\nconviction for felony harassment, in violation of Wash. Rev.\\nCode §§ 9A.46.020(1)(a)(i) and (2)(B), qualified as a crime\\nof violence, as defined in U.S.S.G. § 4B1.2. Applying the\\ncategorical approach, the panel held that the conviction\\nqualified as a crime of violence under § 4B1.2(a)’s force\\nclause because it necessarily entailed the threatened use of\\nviolent physical force.\\n\\n The panel held that the district court erred in concluding\\nthat the defendant’s 2002 conviction for second degree\\nassault, in violation of Wash. Rev. Code § 9A.36.021(1)(c),\\nqualified as a crime of violence. Under United States v.\\nRobinson, 869 F.3d 933 (9th Cir. 2017), the conviction did\\nnot qualify under the force clause of § 4B1.2(a). The panel\\nheld that second degree assault also did not qualify as a crime\\nof violence under § 4B1.2(a)’s residual clause because the\\noffense, in the ordinary case, does not present a serious\\n\\n *\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f UNITED STATES V. DOOR 3\\n\\npotential risk of physical injury to another, and it is not\\nsimilar in kind to the crimes listed in the enumerated offenses\\nclause.\\n\\n The panel therefore vacated the sentence and remanded\\nfor resentencing. The panel addressed other issues in a\\nconcurrently-filed memorandum disposition.\\n\\n\\n COUNSEL\\n\\nCarlton Gunn (argued), Pasadena, California, for Defendant-\\nAppellant.\\n\\nHelen J. Brunner (argued), First Assistant United States\\nAttorney; Annette L. Hayes, United States Attorney; United\\nStates Attorney’s Office, Seattle, Washington; for Plaintiff-\\nAppellee.\\n\\n\\n OPINION\\n\\nCHRISTEN, Circuit Judge:\\n\\n Defendant Kenneth Randale Door appeals the sentence\\nthe district court imposed after he was convicted of several\\noffenses in 2014. At his 2017 sentencing hearing, the district\\ncourt determined that Door’s base offense level should be 24\\npursuant to United States Sentencing Guideline (U.S.S.G.)\\n§ 2K2.1(a)(2) after concluding that Door’s prior Washington\\nstate convictions for second-degree assault and felony\\nharassment qualify as crimes of violence. Door contends\\nthese offenses do not constitute crimes of violence and that\\nhis offense level was thus calculated incorrectly. Consistent\\n\\f4 UNITED STATES V. DOOR\\n\\nwith United States v. Werle, 877 F.3d 879, 884 (9th Cir.\\n2017), we hold that Door’s 1997 conviction for felony\\nharassment, in violation of the Revised Code of Washington\\n(Wash. Rev. Code) §§ 9A.46.020(1)(a)(i) and (2)(b), qualifies\\nas a crime of violence. Door’s argument to the contrary\\ndisregards that the framework for the “crime of violence”\\nanalysis differs depending on whether the prior offense is\\nalleged to constitute a crime of violence pursuant to the force\\nclause, the enumerated offenses clause, or the residual clause\\nof U.S.S.G. § 4B1.2(a). The district court did err, however,\\nin concluding that Door’s 2002 conviction for second-degree\\nassault, in violation of Wash. Rev. Code. § 9A.36.021(1)(c),\\nqualifies as a crime of violence. See United States v.\\nRobinson, 869 F.3d 933, 941 (9th Cir. 2017); see also United\\nStates v. Vederoff, 914 F.3d 1238, 1244–46 (9th Cir. 2019).\\nAccordingly, we vacate Door’s sentence and remand for\\nresentencing.1\\n\\n FACTUAL AND PROCEDURAL BACKGROUND\\n\\n A search of Door’s home in 2011 led to the discovery of\\ntwo handguns, some magazines loaded with ammunition, two\\nmilitary grade ballistic vests (body armor), an explosive\\ndevice known as a “seal bomb,” two digital scales, drug\\npackaging materials, and two drug pipes containing\\nmethamphetamine residue. These discoveries led to Door’s\\nindictment in 2012; he was charged with being a Felon in\\n\\n\\n 1\\n Door also contends that he was improperly convicted of possessing\\nbody armor, challenges the application of an enhancement for possessing\\na firearm in connection with another felony, and argues that his case\\nshould be reassigned if remanded. In a concurrently filed memorandum\\ndisposition, we affirm the body armor conviction and sentencing\\nenhancement, and decline to reassign the case on remand.\\n\\f UNITED STATES V. DOOR 5\\n\\nPossession of a Firearm, a Violent Felon in Possession of\\nBody Armor, and a Felon in Possession of an Explosive. In\\n2014, a jury convicted Door of each count. Door received a\\n300-month sentence, but for reasons not pertinent to this\\nappeal, we vacated Door’s sentence and remanded for\\nresentencing.\\n\\n On remand, the probation officer recommended a base\\noffense level of 24 because the officer concluded that Door’s\\nprior Washington state convictions for second-degree assault\\nand felony harassment constitute crimes of violence. See\\nU.S.S.G. § 2K2.1(a)(2) (providing that a base offense level of\\n24 applies “if the defendant committed any part of the instant\\noffense subsequent to sustaining at least two felony\\nconvictions of either a crime of violence or a controlled\\nsubstance offense”). A total offense level of 32 and Door’s\\ncriminal history category of VI yielded a guideline\\nimprisonment range of 210 to 262 months. The probation\\nofficer recommended 276 months of imprisonment due to\\nDoor’s extensive criminal history. This represented the\\nmaximum sentence for each count, served consecutively.\\nDoor argued that his prior convictions for second-degree\\nassault and felony harassment were not “crime[s] of violence”\\nas defined in U.S.S.G. § 4B1.2, and thus his base offense\\nlevel should only be 14.\\n\\n At the resentencing hearing held in 2017, the district court\\nruled that Door’s prior convictions for second-degree assault\\nand felony harassment qualified as crimes of violence\\npursuant to U.S.S.G. § 4B1.2. The court concluded that Door\\nhad a total offense level of 32 and imposed a sentence of\\n276 months.\\n\\f6 UNITED STATES V. DOOR\\n\\n ANALYSIS\\n\\n We have jurisdiction pursuant to 28 U.S.C. § 1291. “We\\nreview de novo whether a state-law crime constitutes a crime\\nof violence under the [Federal Sentencing] Guidelines.”\\nRobinson, 869 F.3d at 936.\\n\\n To determine whether a prior conviction qualifies as a\\ncrime of violence, we apply the categorical approach first\\noutlined in Taylor v. United States, 495 U.S. 575 (1990). The\\ncategorical approach requires courts to compare the elements\\nof the statute of conviction with the federal definition of\\n“crime of violence” to determine whether the statute of\\nconviction criminalizes a broader range of conduct than the\\nfederal definition captures. United States v. Edling, 895 F.3d\\n1153, 1155 (9th Cir. 2018). The 2013 Sentencing Guidelines\\ndefine the term “crime of violence” as follows:\\n\\n [A]ny offense under federal or state law,\\n punishable by imprisonment for a term\\n exceeding one year, that—\\n\\n (1) has as an element the use, attempted\\n use, or threatened use of physical\\n force against the person of another\\n [known as the force clause or the\\n elements clause], or\\n\\n (2) is burglary of a dwelling, arson, or\\n extortion, involves use of explosives\\n [known as the enumerated offenses\\n clause], or otherwise involves conduct\\n that presents a serious potential risk of\\n\\f UNITED STATES V. DOOR 7\\n\\n physical injury to another [known as\\n the residual clause].\\n\\nU.S.S.G. § 4B1.2(a) (2013).2\\n\\n When determining whether a prior conviction constitutes\\na crime of violence, the precise inquiry differs depending on\\nwhether the offense is alleged to qualify as a crime of\\nviolence pursuant to the force clause, the enumerated offenses\\nclause, or the residual clause. See, e.g., Edling, 895 F.3d at\\n1156–58 (determining whether prior convictions qualified as\\ncrimes of violence pursuant to the force clause and the\\nenumerated offenses clause); United States v. Adkins,\\n883 F.3d 1207, 1213–15 (9th Cir. 2018) (determining\\nwhether prior convictions qualified as crimes of violence\\npursuant to the residual clause). An offense constitutes a\\n“crime of violence” if it qualifies under any one of the three\\nclauses. See Edling, 895 F.3d at 1155. Throughout this\\nanalysis, we “presume that the conviction ‘rested upon\\n[nothing] more than the least of th[e] acts’ criminalized” by\\nthe statute of conviction. Moncrieffe v. Holder, 569 U.S. 184,\\n191–92 (2013) (alterations in original) (quoting Johnson v.\\nUnited States, 559 U.S. 133, 137 (2010)).\\n\\n\\n\\n\\n 2\\n The relevant Sentencing Guidelines are those in effect on the date\\nof the defendant’s first sentencing hearing. See 18 U.S.C. § 3742(g)(1).\\nDoor’s first sentencing hearing occurred when the 2013 Guidelines were\\nin effect. Although the residual clause was later omitted from the\\nGuidelines’ definition of a crime of violence, we do not apply the\\namendment retroactively because it substantively changed the Guidelines.\\nUnited States v. Adkins, 883 F.3d 1207, 1211–12 (9th Cir. 2018).\\n\\f8 UNITED STATES V. DOOR\\n\\n A. Applying the Categorical Analysis to the\\n Sentencing Guidelines’ Force Clause, Enumerated\\n Offenses Clause, and Residual Clause\\n\\n To determine whether a prior conviction qualifies\\npursuant to the force clause, the question is whether the crime\\nof conviction “has as an element the use or threatened use of\\nphysical force against the person of another, with ‘physical\\nforce’ understood to mean in this context ‘violent force—that\\nis, force capable of causing physical pain or injury to another\\nperson.’” Edling, 895 F.3d at 1156 (quoting Johnson,\\n559 U.S. at 140). If the crime of conviction necessarily\\nentails the use or threatened use of violent physical force, it\\nis considered a categorical match for a crime of violence\\npursuant to the force clause of § 4B1.2(a)(1), and the inquiry\\nends. See id.; see also Stokeling v. United States, 139 S. Ct.\\n544, 554–55 (2019).\\n\\n A prior offense constitutes a crime of violence pursuant\\nto the enumerated offenses clause if the elements of one of\\nthe generic federal crimes listed in that clause fully subsume\\nthe elements of the crime of conviction. See Mathis v. United\\nStates, 136 S. Ct. 2243, 2248 (2016); United States v.\\nPeterson, 902 F.3d 1016, 1021–22 (9th Cir. 2018). A generic\\nfederal crime is defined by looking to the common law, the\\nModel Penal Code, treatises, and the laws of other states. See\\nVederoff, 914 F.3d at 1245; United States v. Esparza-\\nHerrera, 557 F.3d 1019, 1022–23 (9th Cir. 2009). If the\\ncrime of conviction falls within the generic federal\\ndefinition—meaning it does not punish a broader range of\\nconduct than the generic offense—the conviction qualifies as\\na crime of violence. See, e.g., Mathis, 136 S. Ct. at 2248.\\n\\f UNITED STATES V. DOOR 9\\n\\n We use a two-part test to assess whether an offense is\\ncategorically a crime of violence pursuant to the residual\\nclause in § 4B1.2(a)(2). Adkins, 883 F.3d at 1213. “First, the\\n‘conduct encompassed by the elements of the offense, in the\\nordinary case, must present a serious potential risk of\\nphysical injury to another.’” Id. (quoting United States v.\\nPark, 649 F.3d 1175, 1177–78 (9th Cir. 2011)). “Second, the\\nprior offense must be ‘roughly similar, in kind as well as in\\ndegree of risk posed’” to the crimes listed in the enumerated\\noffenses clause. Id. (quoting Park, 649 F.3d at 1178). To\\ndetermine whether the offense is “similar in kind” to the\\nlisted crimes, we consider “whether the predicate offense\\ninvolves ‘purposeful, violent, and aggressive conduct.’” Id.\\n(quoting Begay v. United States, 553 U.S. 137, 145 (2008)).\\nBoth criteria must be satisfied for a prior offense to constitute\\na crime of violence pursuant to the residual clause of\\n§ 4B1.2(a)(2). Id.\\n\\n If the statute of conviction does not qualify as a\\ncategorical match pursuant to the force clause, the\\nenumerated offenses clause, or the residual clause, the court\\nconsiders whether the statute of conviction’s elements are\\ndivisible. See, e.g., Edling, 895 F.3d at 1156, 1159; Adkins,\\n883 F.3d at 1215; Robinson, 869 F.3d at 938. A statute is\\ndivisible if it “list[s] elements in the alternative, and thereby\\ndefine[s] multiple crimes.” Mathis, 136 S. Ct. at 2249. If the\\nstatute of conviction is not a categorical match and is\\nindivisible, it is not a crime of violence. See id. at 2248–49.\\nIf the statute of conviction is not a categorical match and is\\ndivisible, then the court applies the modified categorical\\napproach and “looks to a limited class of documents (for\\nexample, the indictment, jury instructions, or plea agreement\\nand colloquy) to determine what crime, with what elements,\\na defendant was convicted of.” Id. at 2249 (citing Shepard v.\\n\\f10 UNITED STATES V. DOOR\\n\\nUnited States, 544 U.S. 13, 26 (2005)). We then apply the\\nforce clause, the enumerated offenses clause, or the residual\\nclause analysis to the specific crime of conviction to\\ndetermine whether it is a crime of violence. See id.\\n\\n B. Door’s Prior Conviction for the Washington\\n Crime of Felony Harassment Qualifies as a Crime\\n of Violence Pursuant to the Force Clause.\\n\\n Door was convicted in 1997 of felony harassment in\\nviolation of Wash. Rev. Code §§ 9A.46.020(1)(a)(i) and\\n(2)(b) for “threatening to kill” a person.3 We previously\\n\\n\\n\\n\\n 3\\n At the time of Door’s conviction in 1997, the Washington\\nharassment statute provided the following:\\n\\n (1) A person is guilty of harassment if:\\n\\n (a) Without lawful authority, the person knowingly\\n threatens:\\n\\n (i) To cause bodily injury in the future to the\\n person threatened or to any other person; or\\n\\n (ii) To cause physical damage to the property\\n of a person other than the actor; or\\n\\n (iii) To subject the person threatened or any\\n other person to physical confinement or\\n restraint; or\\n\\n (iv) Maliciously to do any other act which is\\n intended to substantially harm the person\\n threatened or another with respect to his or her\\n physical or mental health or safety; and\\n\\f UNITED STATES V. DOOR 11\\n\\nanalyzed this statute in Werle, and held that a conviction\\npursuant to the “threat to kill” subsection of the statute\\nconstitutes a crime of violence pursuant to the force clause of\\n§ 4B1.2(a)(1).4 Werle, 877 F.3d at 883–84. Because a\\nconviction pursuant to Wash. Rev. Code § 9A.46.020(2)(b)\\nrequires that a defendant be found guilty of knowingly\\nthreatening to kill a person, this offense necessarily entails\\nthe threatened use of violent physical force against another\\nperson, as required by § 4B1.2(a)(1). Id. Thus, Door’s\\n\\n\\n\\n (b) The person by words or conduct places the\\n person threatened in reasonable fear that the threat\\n will be carried out.\\n\\n (2) A person who harasses another is guilty of a gross\\n misdemeanor punishable under chapter 9A.20 RCW,\\n except that the person is guilty of a class C felony if\\n either of the following applies:\\n\\n (a) The person has previously been convicted in\\n this or any other state of any crime of harassment,\\n as defined in RCW 9A.46.060, of the same victim\\n or members of the victim’s family or household or\\n any person specifically named in a no-contact or\\n no-harassment order; or\\n\\n (b) the person harasses another person under subsection\\n (1)(a)(i) of this section by threatening to kill the\\n person threatened or any other person.\\n\\nWash. Rev. Code § 9A.46.020.\\n 4\\n In Werle, we assessed the amended version of Wash. Rev. Code\\n§ 9A.46.020. 877 F.3d at 882. Although Door’s conviction occurred prior\\nto the statutory amendment, the relevant language in the earlier version of\\nthe felony harassment statute is substantially the same as the amended\\nversion reviewed in Werle.\\n\\f12 UNITED STATES V. DOOR\\n\\nconviction for the Washington crime of felony harassment\\nqualifies as a crime of violence for sentencing purposes.\\n\\n Door contends that we are not bound by Werle’s holding,\\nobserving that in Werle, the defendant never argued his prior\\nWashington felony harassment conviction failed to qualify as\\na crime of violence on account of Washington’s version of\\naiding and abetting, which is broader than the federal\\ndefinition of aiding and abetting. As Door’s argument goes,\\nbecause every Washington criminal statute incorporates\\naiding and abetting, all Washington criminal statutes are\\noverbroad, and therefore all Washington state convictions fail\\nto qualify as crimes of violence—an argument inspired by\\nUnited States v. Valdivia-Flores, 876 F.3d 1201 (9th Cir.\\n2017). Door’s argument, however, overlooks the analytical\\ndifference between the force clause and the enumerated\\noffenses clause of U.S.S.G. § 4B1.2(a).\\n\\n In Valdivia-Flores, this court assessed whether a\\nconviction for violating a Washington drug-trafficking law,\\nWash. Rev. Code § 69.50.401, constituted an “aggravated\\nfelony” for purposes of the Immigration and Nationality Act\\n(INA). Valdivia-Flores, 876 F.3d at 1206. Applying\\nTaylor’s categorical approach, we considered “whether the\\nstate statute defining the crime of conviction categorically\\nfit[] within the generic federal definition of a corresponding\\naggravated felony.” Id. (quoting Roman-Suaste v. Holder,\\n766 F.3d 1035, 1038 (9th Cir. 2014)). Valdivia-Flores’s\\ncrime of conviction was overbroad because: (1) aiding and\\nabetting is included in every Washington criminal statute;\\n(2) Washington defines aiding and abetting more broadly than\\nfederal law because it merely requires knowledge, whereas\\nfederal law requires a mens rea of specific intent; and\\n(3) accomplice and principal liability are indivisible under\\n\\f UNITED STATES V. DOOR 13\\n\\nWashington law. Id. at 1207–10. Concluding that Valdivia-\\nFlores’s prior offense did not qualify as an “aggravated\\nfelony,” we observed that it may be that “no Washington state\\nconviction can serve as an aggravated felony at all.” Id. at\\n1209–10.\\n\\n Door’s reliance on Valdivia-Flores is misplaced. The\\ncategorical analysis in Valdivia-Flores involved comparing\\nthe elements of the Washington drug trafficking crime with\\nthe generic federal offense of drug trafficking because “drug\\ntrafficking” is listed in the INA as an “aggravated felony.”\\nSee id. at 1206–07. In other words, the categorical analysis\\nemployed in Valdivia-Flores mirrors the inquiry under the\\nenumerated offenses clause of U.S.S.G. § 4B1.2(a)(2).\\nWerle, on the other hand, held that a prior conviction for\\nWashington felony harassment constitutes a crime of violence\\npursuant to the force clause of § 4B1.2(a)(1). Because a\\nconviction for violating Wash. Rev. Code § 9A.46.020(2)(b)\\nnecessarily entails the threatened use of violent physical\\nforce, it qualifies as a crime of violence pursuant to the force\\nclause, and our inquiry ends there. We need not compare the\\nelements of the crime of conviction with the elements of the\\ngeneric federal crime when analyzing whether an offense\\nqualifies as a crime of violence pursuant to the force clause\\nof § 4B1.2(a). Compare Peterson, 902 F.3d at 1016,\\n1021–22, with Edling, 895 F.3d at 1156–58. Valdivia-Flores\\nis consistent with Werle.\\n\\f14 UNITED STATES V. DOOR\\n\\n C. Door’s Prior Conviction for the Washington\\n Crime of Second-Degree Assault Does Not Qualify\\n as a Crime of Violence.\\n\\n Door was convicted in 2002 of second-degree assault in\\nviolation of Wash. Rev. Code § 9A.36.021(1)(c).5 In\\nRobinson, we held that Washington’s crime of second-degree\\nassault did not qualify as a crime of violence pursuant to the\\n\\n\\n 5\\n At the time of Door’s conviction in 2002, the Washington second-\\ndegree assault statute provided the following:\\n\\n (1) A person is guilty of assault in the second degree if\\n he or she, under circumstances not amounting to assault\\n in the first degree:\\n\\n (a) Intentionally assaults another and thereby\\n recklessly inflicts substantial bodily harm; or\\n\\n (b) Intentionally and unlawfully causes substantial\\n bodily harm to an unborn quick child by\\n intentionally and unlawfully inflicting any injury\\n upon the mother of such child; or\\n\\n (c) Assaults another with a deadly weapon; or\\n\\n (d) With intent to inflict bodily harm, administers\\n to or causes to be taken by another, poison or any\\n other destructive or noxious substance; or\\n\\n (e) With intent to commit a felony, assaults\\n another; or\\n\\n (f) Knowingly inflicts bodily harm which by\\n design causes such pain or agony as to be the\\n equivalent of that produced by torture.\\n\\nWash. Rev. Code § 9A.36.021(1).\\n\\f UNITED STATES V. DOOR 15\\n\\nforce clause of U.S.S.G. § 4B1.2(a)(1).6 869 F.3d at 937–38.\\nHowever, Robinson did not address whether second-degree\\nassault constitutes a crime of violence pursuant to the residual\\nclause of § 4B1.2(a)(2).7\\n\\n Applying the two-part residual clause test, we hold that\\nWash. Rev. Code § 9A.36.021(1) is not a crime of violence\\nbecause the offense, in the ordinary case, does not “present a\\nserious potential risk of physical injury to another.” Adkins,\\n883 F.3d at 1213 (quoting Park, 649 F.3d at 1177–78). The\\nWashington second-degree assault statute encompasses\\nassault with intent to commit a felony. Wash. Rev. Code\\n§ 9A.36.021(1)(e). This includes the intent to commit any\\nnon-violent felony offense. The “assault” may also be non-\\nviolent because Washington defines assault broadly to\\ninclude “an intentional touching . . . that is harmful or\\noffensive regardless of whether any physical injury is done to\\nthe person.” State v. Smith, 154 P.3d 873, 875 (Wash. 2007)\\n(en banc); see also Robinson, 869 F.3d at 938 n.7. Thus, a\\ndefendant may violate Wash. Rev. Code § 9A.36.021(1)(e) in\\n\\n\\n\\n\\n 6\\n In Robinson, we assessed the amended version of Wash. Rev. Code\\n§ 9A.36.021(1). 869 F.3d at 937 n.6. The statute was amended after\\nDoor’s conviction to include a seventh subsection, (1)(g) which prohibits\\n“[a]ssaults . . . by strangulation or suffocation.” Id.\\n 7\\n Robinson also did not address the enumerated offenses clause. To\\nthe extent the government argues that second-degree assault qualifies as\\na crime of violence pursuant to the enumerated offenses clause, we decline\\nto address this argument, raised for the first time on appeal. See United\\nStates v. Fomichev, 899 F.3d 766, 770 (9th Cir. 2018), as amended by\\n909 F.3d 1078 (9th Cir. 2018). Further, the argument is foreclosed by\\nVederoff. See Vederoff, 914 F.3d at 1246.\\n\\f16 UNITED STATES V. DOOR\\n\\na way that poses no serious risk of physical injury to others.8\\nBecause Door “could have been convicted on the basis of\\nconduct that did not present a serious risk of physical injury\\nto another,” a prior conviction for Washington second-degree\\nassault does not qualify as a crime of violence. See United\\nStates v. Simmons, 782 F.3d 510, 519 (9th Cir. 2015) (quoting\\nUnited States v. Kelly, 422 F.3d 889, 893 (9th Cir. 2005)); see\\nalso United States v. Lee, 821 F.3d 1124, 1128–29 (9th Cir.\\n2016) (holding that California’s crime of battery committed\\nagainst a custodial officer does not qualify as a crime of\\nviolence pursuant to the residual clause because “the least\\ntouching may constitute battery” (quoting People v. Mesce,\\n60 Cal. Rptr. 2d 745, 756 (Cal. Ct. App. 1997)).\\n\\n Wash. Rev. Code § 9A.36.021(1) also fails to satisfy the\\nsecond part of the residual clause test because it is not\\n“similar in kind” to the crimes listed in the enumerated\\noffenses clause, which typically involve “purposeful, violent,\\nand aggressive conduct.” Adkins, 883 F.3d at 1213 (quoting\\nBegay, 553 U.S. at 145). As discussed, the Washington crime\\nof second-degree assault can be committed by offensive\\ntouching, see Smith, 154 P.3d at 875, which would not\\ninvolve “violent” or “aggressive” conduct.\\n\\n The Washington second-degree assault statute is\\nindivisible, thus the modified categorical approach is\\ninapplicable. Robinson, 869 F.3d at 941. The district court\\nerred by concluding that Door’s prior second-degree assault\\nconviction qualifies as a crime of violence pursuant to the\\nresidual clause.\\n\\n\\n 8\\n For example, a defendant may be convicted of assault with intent to\\ncommit a felony for touching a minor’s body in a sexual manner. See\\nRobinson, 869 F.3d at 938 n.7.\\n\\f UNITED STATES V. DOOR 17\\n\\n D. Remand is Required\\n\\n The Sentencing Guidelines direct that a base offense level\\nof 24 applies if the defendant has two or more felony\\nconvictions that qualify as a “crime of violence.” U.S.S.G.\\n§ 2K2.1(a)(2). A mistake in calculating the Guidelines\\nsentencing range generally requires us to remand for\\nresentencing. United States v. Bankston, 901 F.3d 1100, 1107\\n(9th Cir. 2018). Door’s conviction for the Washington crime\\nof felony harassment qualifies as a crime of violence, but the\\ndistrict court erred in holding that Door’s conviction for the\\nWashington crime of second-degree assault constitutes a\\ncrime of violence. Because Door only had one conviction\\nthat qualified as a crime of violence, his base offense level\\nshould have been 20. See U.S.S.G. § 2K2.1(a)(4). The\\ndistrict court therefore erred in calculating Door’s base\\noffense level and remand is required.\\n\\n Sentence VACATED; REMANDED for resentencing.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4376067/\", \"author_raw\": \"CHRISTEN, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"SIDNEY R THOMAS"},"judge2":{"kind":"string","value":"M MARGARET MCKEOWN"},"judge3":{"kind":"string","value":"MORGAN CHRISTEN"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4598814/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. Kenneth Door\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-12\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 17-30165\n Plaintiff-Appellee,\n D.C. No.\n v. 3:12-cr-05126-RBL-1\n\nKENNETH RANDALE DOOR,\n Defendant-Appellant. OPINION\n\n\n Appeal from the United States District Court\n for the Western District of Washington\n Ronald B. Leighton, District Judge, Presiding\n\n Argued and Submitted December 7, 2018\n Seattle, Washington\n\n Filed March 12, 2019\n\nBefore: Sidney R. Thomas, Chief Judge, and M. Margaret\n McKeown and Morgan Christen, Circuit Judges.\n\n Opinion by Judge Christen\n\f2 UNITED STATES V. DOOR\n\n SUMMARY*\n\n\n Criminal Law\n\n Vacating a sentence and remanding for resentencing, the\npanel held that, in setting the defendant’s base offense level\nunder U.S.S.G. § 2K2.1(a)(2), his prior Washington State\nconviction for felony harassment qualified as a crime of\nviolence but his prior conviction for second degree assault did\nnot.\n\n Consistent with United States v. Werle, 877 F.3d 879 (9th\nCir. 2017), the panel held that the defendant’s 1997\nconviction for felony harassment, in violation of Wash. Rev.\nCode §§ 9A.46.020(1)(a)(i) and (2)(B), qualified as a crime\nof violence, as defined in U.S.S.G. § 4B1.2. Applying the\ncategorical approach, the panel held that the conviction\nqualified as a crime of violence under § 4B1.2(a)’s force\nclause because it necessarily entailed the threatened use of\nviolent physical force.\n\n The panel held that the district court erred in concluding\nthat the defendant’s 2002 conviction for second degree\nassault, in violation of Wash. Rev. Code § 9A.36.021(1)(c),\nqualified as a crime of violence. Under United States v.\nRobinson, 869 F.3d 933 (9th Cir. 2017), the conviction did\nnot qualify under the force clause of § 4B1.2(a). The panel\nheld that second degree assault also did not qualify as a crime\nof violence under § 4B1.2(a)’s residual clause because the\noffense, in the ordinary case, does not present a serious\n\n *\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\f UNITED STATES V. DOOR 3\n\npotential risk of physical injury to another, and it is not\nsimilar in kind to the crimes listed in the enumerated offenses\nclause.\n\n The panel therefore vacated the sentence and remanded\nfor resentencing. The panel addressed other issues in a\nconcurrently-filed memorandum disposition.\n\n\n COUNSEL\n\nCarlton Gunn (argued), Pasadena, California, for Defendant-\nAppellant.\n\nHelen J. Brunner (argued), First Assistant United States\nAttorney; Annette L. Hayes, United States Attorney; United\nStates Attorney’s Office, Seattle, Washington; for Plaintiff-\nAppellee.\n\n\n OPINION\n\nCHRISTEN, Circuit Judge:\n\n Defendant Kenneth Randale Door appeals the sentence\nthe district court imposed after he was convicted of several\noffenses in 2014. At his 2017 sentencing hearing, the district\ncourt determined that Door’s base offense level should be 24\npursuant to United States Sentencing Guideline (U.S.S.G.)\n§ 2K2.1(a)(2) after concluding that Door’s prior Washington\nstate convictions for second-degree assault and felony\nharassment qualify as crimes of violence. Door contends\nthese offenses do not constitute crimes of violence and that\nhis offense level was thus calculated incorrectly. Consistent\n\f4 UNITED STATES V. DOOR\n\nwith United States v. Werle, 877 F.3d 879, 884 (9th Cir.\n2017), we hold that Door’s 1997 conviction for felony\nharassment, in violation of the Revised Code of Washington\n(Wash. Rev. Code) §§ 9A.46.020(1)(a)(i) and (2)(b), qualifies\nas a crime of violence. Door’s argument to the contrary\ndisregards that the framework for the “crime of violence”\nanalysis differs depending on whether the prior offense is\nalleged to constitute a crime of violence pursuant to the force\nclause, the enumerated offenses clause, or the residual clause\nof U.S.S.G. § 4B1.2(a). The district court did err, however,\nin concluding that Door’s 2002 conviction for second-degree\nassault, in violation of Wash. Rev. Code. § 9A.36.021(1)(c),\nqualifies as a crime of violence. See United States v.\nRobinson, 869 F.3d 933, 941 (9th Cir. 2017); see also United\nStates v. Vederoff, 914 F.3d 1238, 1244–46 (9th Cir. 2019).\nAccordingly, we vacate Door’s sentence and remand for\nresentencing.1\n\n FACTUAL AND PROCEDURAL BACKGROUND\n\n A search of Door’s home in 2011 led to the discovery of\ntwo handguns, some magazines loaded with ammunition, two\nmilitary grade ballistic vests (body armor), an explosive\ndevice known as a “seal bomb,” two digital scales, drug\npackaging materials, and two drug pipes containing\nmethamphetamine residue. These discoveries led to Door’s\nindictment in 2012; he was charged with being a Felon in\n\n\n 1\n Door also contends that he was improperly convicted of possessing\nbody armor, challenges the application of an enhancement for possessing\na firearm in connection with another felony, and argues that his case\nshould be reassigned if remanded. In a concurrently filed memorandum\ndisposition, we affirm the body armor conviction and sentencing\nenhancement, and decline to reassign the case on remand.\n\f UNITED STATES V. DOOR 5\n\nPossession of a Firearm, a Violent Felon in Possession of\nBody Armor, and a Felon in Possession of an Explosive. In\n2014, a jury convicted Door of each count. Door received a\n300-month sentence, but for reasons not pertinent to this\nappeal, we vacated Door’s sentence and remanded for\nresentencing.\n\n On remand, the probation officer recommended a base\noffense level of 24 because the officer concluded that Door’s\nprior Washington state convictions for second-degree assault\nand felony harassment constitute crimes of violence. See\nU.S.S.G. § 2K2.1(a)(2) (providing that a base offense level of\n24 applies “if the defendant committed any part of the instant\noffense subsequent to sustaining at least two felony\nconvictions of either a crime of violence or a controlled\nsubstance offense”). A total offense level of 32 and Door’s\ncriminal history category of VI yielded a guideline\nimprisonment range of 210 to 262 months. The probation\nofficer recommended 276 months of imprisonment due to\nDoor’s extensive criminal history. This represented the\nmaximum sentence for each count, served consecutively.\nDoor argued that his prior convictions for second-degree\nassault and felony harassment were not “crime[s] of violence”\nas defined in U.S.S.G. § 4B1.2, and thus his base offense\nlevel should only be 14.\n\n At the resentencing hearing held in 2017, the district court\nruled that Door’s prior convictions for second-degree assault\nand felony harassment qualified as crimes of violence\npursuant to U.S.S.G. § 4B1.2. The court concluded that Door\nhad a total offense level of 32 and imposed a sentence of\n276 months.\n\f6 UNITED STATES V. DOOR\n\n ANALYSIS\n\n We have jurisdiction pursuant to 28 U.S.C. § 1291. “We\nreview de novo whether a state-law crime constitutes a crime\nof violence under the [Federal Sentencing] Guidelines.”\nRobinson, 869 F.3d at 936.\n\n To determine whether a prior conviction qualifies as a\ncrime of violence, we apply the categorical approach first\noutlined in Taylor v. United States, 495 U.S. 575 (1990). The\ncategorical approach requires courts to compare the elements\nof the statute of conviction with the federal definition of\n“crime of violence” to determine whether the statute of\nconviction criminalizes a broader range of conduct than the\nfederal definition captures. United States v. Edling, 895 F.3d\n1153, 1155 (9th Cir. 2018). The 2013 Sentencing Guidelines\ndefine the term “crime of violence” as follows:\n\n [A]ny offense under federal or state law,\n punishable by imprisonment for a term\n exceeding one year, that—\n\n (1) has as an element the use, attempted\n use, or threatened use of physical\n force against the person of another\n [known as the force clause or the\n elements clause], or\n\n (2) is burglary of a dwelling, arson, or\n extortion, involves use of explosives\n [known as the enumerated offenses\n clause], or otherwise involves conduct\n that presents a serious potential risk of\n\f UNITED STATES V. DOOR 7\n\n physical injury to another [known as\n the residual clause].\n\nU.S.S.G. § 4B1.2(a) (2013).2\n\n When determining whether a prior conviction constitutes\na crime of violence, the precise inquiry differs depending on\nwhether the offense is alleged to qualify as a crime of\nviolence pursuant to the force clause, the enumerated offenses\nclause, or the residual clause. See, e.g., Edling, 895 F.3d at\n1156–58 (determining whether prior convictions qualified as\ncrimes of violence pursuant to the force clause and the\nenumerated offenses clause); United States v. Adkins,\n883 F.3d 1207, 1213–15 (9th Cir. 2018) (determining\nwhether prior convictions qualified as crimes of violence\npursuant to the residual clause). An offense constitutes a\n“crime of violence” if it qualifies under any one of the three\nclauses. See Edling, 895 F.3d at 1155. Throughout this\nanalysis, we “presume that the conviction ‘rested upon\n[nothing] more than the least of th[e] acts’ criminalized” by\nthe statute of conviction. Moncrieffe v. Holder, 569 U.S. 184,\n191–92 (2013) (alterations in original) (quoting Johnson v.\nUnited States, 559 U.S. 133, 137 (2010)).\n\n\n\n\n 2\n The relevant Sentencing Guidelines are those in effect on the date\nof the defendant’s first sentencing hearing. See 18 U.S.C. § 3742(g)(1).\nDoor’s first sentencing hearing occurred when the 2013 Guidelines were\nin effect. Although the residual clause was later omitted from the\nGuidelines’ definition of a crime of violence, we do not apply the\namendment retroactively because it substantively changed the Guidelines.\nUnited States v. Adkins, 883 F.3d 1207, 1211–12 (9th Cir. 2018).\n\f8 UNITED STATES V. DOOR\n\n A. Applying the Categorical Analysis to the\n Sentencing Guidelines’ Force Clause, Enumerated\n Offenses Clause, and Residual Clause\n\n To determine whether a prior conviction qualifies\npursuant to the force clause, the question is whether the crime\nof conviction “has as an element the use or threatened use of\nphysical force against the person of another, with ‘physical\nforce’ understood to mean in this context ‘violent force—that\nis, force capable of causing physical pain or injury to another\nperson.’” Edling, 895 F.3d at 1156 (quoting Johnson,\n559 U.S. at 140). If the crime of conviction necessarily\nentails the use or threatened use of violent physical force, it\nis considered a categorical match for a crime of violence\npursuant to the force clause of § 4B1.2(a)(1), and the inquiry\nends. See id.; see also Stokeling v. United States, 139 S. Ct.\n544, 554–55 (2019).\n\n A prior offense constitutes a crime of violence pursuant\nto the enumerated offenses clause if the elements of one of\nthe generic federal crimes listed in that clause fully subsume\nthe elements of the crime of conviction. See Mathis v. United\nStates, 136 S. Ct. 2243, 2248 (2016); United States v.\nPeterson, 902 F.3d 1016, 1021–22 (9th Cir. 2018). A generic\nfederal crime is defined by looking to the common law, the\nModel Penal Code, treatises, and the laws of other states. See\nVederoff, 914 F.3d at 1245; United States v. Esparza-\nHerrera, 557 F.3d 1019, 1022–23 (9th Cir. 2009). If the\ncrime of conviction falls within the generic federal\ndefinition—meaning it does not punish a broader range of\nconduct than the generic offense—the conviction qualifies as\na crime of violence. See, e.g., Mathis, 136 S. Ct. at 2248.\n\f UNITED STATES V. DOOR 9\n\n We use a two-part test to assess whether an offense is\ncategorically a crime of violence pursuant to the residual\nclause in § 4B1.2(a)(2). Adkins, 883 F.3d at 1213. “First, the\n‘conduct encompassed by the elements of the offense, in the\nordinary case, must present a serious potential risk of\nphysical injury to another.’” Id. (quoting United States v.\nPark, 649 F.3d 1175, 1177–78 (9th Cir. 2011)). “Second, the\nprior offense must be ‘roughly similar, in kind as well as in\ndegree of risk posed’” to the crimes listed in the enumerated\noffenses clause. Id. (quoting Park, 649 F.3d at 1178). To\ndetermine whether the offense is “similar in kind” to the\nlisted crimes, we consider “whether the predicate offense\ninvolves ‘purposeful, violent, and aggressive conduct.’” Id.\n(quoting Begay v. United States, 553 U.S. 137, 145 (2008)).\nBoth criteria must be satisfied for a prior offense to constitute\na crime of violence pursuant to the residual clause of\n§ 4B1.2(a)(2). Id.\n\n If the statute of conviction does not qualify as a\ncategorical match pursuant to the force clause, the\nenumerated offenses clause, or the residual clause, the court\nconsiders whether the statute of conviction’s elements are\ndivisible. See, e.g., Edling, 895 F.3d at 1156, 1159; Adkins,\n883 F.3d at 1215; Robinson, 869 F.3d at 938. A statute is\ndivisible if it “list[s] elements in the alternative, and thereby\ndefine[s] multiple crimes.” Mathis, 136 S. Ct. at 2249. If the\nstatute of conviction is not a categorical match and is\nindivisible, it is not a crime of violence. See id. at 2248–49.\nIf the statute of conviction is not a categorical match and is\ndivisible, then the court applies the modified categorical\napproach and “looks to a limited class of documents (for\nexample, the indictment, jury instructions, or plea agreement\nand colloquy) to determine what crime, with what elements,\na defendant was convicted of.” Id. at 2249 (citing Shepard v.\n\f10 UNITED STATES V. DOOR\n\nUnited States, 544 U.S. 13, 26 (2005)). We then apply the\nforce clause, the enumerated offenses clause, or the residual\nclause analysis to the specific crime of conviction to\ndetermine whether it is a crime of violence. See id.\n\n B. Door’s Prior Conviction for the Washington\n Crime of Felony Harassment Qualifies as a Crime\n of Violence Pursuant to the Force Clause.\n\n Door was convicted in 1997 of felony harassment in\nviolation of Wash. Rev. Code §§ 9A.46.020(1)(a)(i) and\n(2)(b) for “threatening to kill” a person.3 We previously\n\n\n\n\n 3\n At the time of Door’s conviction in 1997, the Washington\nharassment statute provided the following:\n\n (1) A person is guilty of harassment if:\n\n (a) Without lawful authority, the person knowingly\n threatens:\n\n (i) To cause bodily injury in the future to the\n person threatened or to any other person; or\n\n (ii) To cause physical damage to the property\n of a person other than the actor; or\n\n (iii) To subject the person threatened or any\n other person to physical confinement or\n restraint; or\n\n (iv) Maliciously to do any other act which is\n intended to substantially harm the person\n threatened or another with respect to his or her\n physical or mental health or safety; and\n\f UNITED STATES V. DOOR 11\n\nanalyzed this statute in Werle, and held that a conviction\npursuant to the “threat to kill” subsection of the statute\nconstitutes a crime of violence pursuant to the force clause of\n§ 4B1.2(a)(1).4 Werle, 877 F.3d at 883–84. Because a\nconviction pursuant to Wash. Rev. Code § 9A.46.020(2)(b)\nrequires that a defendant be found guilty of knowingly\nthreatening to kill a person, this offense necessarily entails\nthe threatened use of violent physical force against another\nperson, as required by § 4B1.2(a)(1). Id. Thus, Door’s\n\n\n\n (b) The person by words or conduct places the\n person threatened in reasonable fear that the threat\n will be carried out.\n\n (2) A person who harasses another is guilty of a gross\n misdemeanor punishable under chapter 9A.20 RCW,\n except that the person is guilty of a class C felony if\n either of the following applies:\n\n (a) The person has previously been convicted in\n this or any other state of any crime of harassment,\n as defined in RCW 9A.46.060, of the same victim\n or members of the victim’s family or household or\n any person specifically named in a no-contact or\n no-harassment order; or\n\n (b) the person harasses another person under subsection\n (1)(a)(i) of this section by threatening to kill the\n person threatened or any other person.\n\nWash. Rev. Code § 9A.46.020.\n 4\n In Werle, we assessed the amended version of Wash. Rev. Code\n§ 9A.46.020. 877 F.3d at 882. Although Door’s conviction occurred prior\nto the statutory amendment, the relevant language in the earlier version of\nthe felony harassment statute is substantially the same as the amended\nversion reviewed in Werle.\n\f12 UNITED STATES V. DOOR\n\nconviction for the Washington crime of felony harassment\nqualifies as a crime of violence for sentencing purposes.\n\n Door contends that we are not bound by Werle’s holding,\nobserving that in Werle, the defendant never argued his prior\nWashington felony harassment conviction failed to qualify as\na crime of violence on account of Washington’s version of\naiding and abetting, which is broader than the federal\ndefinition of aiding and abetting. As Door’s argument goes,\nbecause every Washington criminal statute incorporates\naiding and abetting, all Washington criminal statutes are\noverbroad, and therefore all Washington state convictions fail\nto qualify as crimes of violence—an argument inspired by\nUnited States v. Valdivia-Flores, 876 F.3d 1201 (9th Cir.\n2017). Door’s argument, however, overlooks the analytical\ndifference between the force clause and the enumerated\noffenses clause of U.S.S.G. § 4B1.2(a).\n\n In Valdivia-Flores, this court assessed whether a\nconviction for violating a Washington drug-trafficking law,\nWash. Rev. Code § 69.50.401, constituted an “aggravated\nfelony” for purposes of the Immigration and Nationality Act\n(INA). Valdivia-Flores, 876 F.3d at 1206. Applying\nTaylor’s categorical approach, we considered “whether the\nstate statute defining the crime of conviction categorically\nfit[] within the generic federal definition of a corresponding\naggravated felony.” Id. (quoting Roman-Suaste v. Holder,\n766 F.3d 1035, 1038 (9th Cir. 2014)). Valdivia-Flores’s\ncrime of conviction was overbroad because: (1) aiding and\nabetting is included in every Washington criminal statute;\n(2) Washington defines aiding and abetting more broadly than\nfederal law because it merely requires knowledge, whereas\nfederal law requires a mens rea of specific intent; and\n(3) accomplice and principal liability are indivisible under\n\f UNITED STATES V. DOOR 13\n\nWashington law. Id. at 1207–10. Concluding that Valdivia-\nFlores’s prior offense did not qualify as an “aggravated\nfelony,” we observed that it may be that “no Washington state\nconviction can serve as an aggravated felony at all.” Id. at\n1209–10.\n\n Door’s reliance on Valdivia-Flores is misplaced. The\ncategorical analysis in Valdivia-Flores involved comparing\nthe elements of the Washington drug trafficking crime with\nthe generic federal offense of drug trafficking because “drug\ntrafficking” is listed in the INA as an “aggravated felony.”\nSee id. at 1206–07. In other words, the categorical analysis\nemployed in Valdivia-Flores mirrors the inquiry under the\nenumerated offenses clause of U.S.S.G. § 4B1.2(a)(2).\nWerle, on the other hand, held that a prior conviction for\nWashington felony harassment constitutes a crime of violence\npursuant to the force clause of § 4B1.2(a)(1). Because a\nconviction for violating Wash. Rev. Code § 9A.46.020(2)(b)\nnecessarily entails the threatened use of violent physical\nforce, it qualifies as a crime of violence pursuant to the force\nclause, and our inquiry ends there. We need not compare the\nelements of the crime of conviction with the elements of the\ngeneric federal crime when analyzing whether an offense\nqualifies as a crime of violence pursuant to the force clause\nof § 4B1.2(a). Compare Peterson, 902 F.3d at 1016,\n1021–22, with Edling, 895 F.3d at 1156–58. Valdivia-Flores\nis consistent with Werle.\n\f14 UNITED STATES V. DOOR\n\n C. Door’s Prior Conviction for the Washington\n Crime of Second-Degree Assault Does Not Qualify\n as a Crime of Violence.\n\n Door was convicted in 2002 of second-degree assault in\nviolation of Wash. Rev. Code § 9A.36.021(1)(c).5 In\nRobinson, we held that Washington’s crime of second-degree\nassault did not qualify as a crime of violence pursuant to the\n\n\n 5\n At the time of Door’s conviction in 2002, the Washington second-\ndegree assault statute provided the following:\n\n (1) A person is guilty of assault in the second degree if\n he or she, under circumstances not amounting to assault\n in the first degree:\n\n (a) Intentionally assaults another and thereby\n recklessly inflicts substantial bodily harm; or\n\n (b) Intentionally and unlawfully causes substantial\n bodily harm to an unborn quick child by\n intentionally and unlawfully inflicting any injury\n upon the mother of such child; or\n\n (c) Assaults another with a deadly weapon; or\n\n (d) With intent to inflict bodily harm, administers\n to or causes to be taken by another, poison or any\n other destructive or noxious substance; or\n\n (e) With intent to commit a felony, assaults\n another; or\n\n (f) Knowingly inflicts bodily harm which by\n design causes such pain or agony as to be the\n equivalent of that produced by torture.\n\nWash. Rev. Code § 9A.36.021(1).\n\f UNITED STATES V. DOOR 15\n\nforce clause of U.S.S.G. § 4B1.2(a)(1).6 869 F.3d at 937–38.\nHowever, Robinson did not address whether second-degree\nassault constitutes a crime of violence pursuant to the residual\nclause of § 4B1.2(a)(2).7\n\n Applying the two-part residual clause test, we hold that\nWash. Rev. Code § 9A.36.021(1) is not a crime of violence\nbecause the offense, in the ordinary case, does not “present a\nserious potential risk of physical injury to another.” Adkins,\n883 F.3d at 1213 (quoting Park, 649 F.3d at 1177–78). The\nWashington second-degree assault statute encompasses\nassault with intent to commit a felony. Wash. Rev. Code\n§ 9A.36.021(1)(e). This includes the intent to commit any\nnon-violent felony offense. The “assault” may also be non-\nviolent because Washington defines assault broadly to\ninclude “an intentional touching . . . that is harmful or\noffensive regardless of whether any physical injury is done to\nthe person.” State v. Smith, 154 P.3d 873, 875 (Wash. 2007)\n(en banc); see also Robinson, 869 F.3d at 938 n.7. Thus, a\ndefendant may violate Wash. Rev. Code § 9A.36.021(1)(e) in\n\n\n\n\n 6\n In Robinson, we assessed the amended version of Wash. Rev. Code\n§ 9A.36.021(1). 869 F.3d at 937 n.6. The statute was amended after\nDoor’s conviction to include a seventh subsection, (1)(g) which prohibits\n“[a]ssaults . . . by strangulation or suffocation.” Id.\n 7\n Robinson also did not address the enumerated offenses clause. To\nthe extent the government argues that second-degree assault qualifies as\na crime of violence pursuant to the enumerated offenses clause, we decline\nto address this argument, raised for the first time on appeal. See United\nStates v. Fomichev, 899 F.3d 766, 770 (9th Cir. 2018), as amended by\n909 F.3d 1078 (9th Cir. 2018). Further, the argument is foreclosed by\nVederoff. See Vederoff, 914 F.3d at 1246.\n\f16 UNITED STATES V. DOOR\n\na way that poses no serious risk of physical injury to others.8\nBecause Door “could have been convicted on the basis of\nconduct that did not present a serious risk of physical injury\nto another,” a prior conviction for Washington second-degree\nassault does not qualify as a crime of violence. See United\nStates v. Simmons, 782 F.3d 510, 519 (9th Cir. 2015) (quoting\nUnited States v. Kelly, 422 F.3d 889, 893 (9th Cir. 2005)); see\nalso United States v. Lee, 821 F.3d 1124, 1128–29 (9th Cir.\n2016) (holding that California’s crime of battery committed\nagainst a custodial officer does not qualify as a crime of\nviolence pursuant to the residual clause because “the least\ntouching may constitute battery” (quoting People v. Mesce,\n60 Cal. Rptr. 2d 745, 756 (Cal. Ct. App. 1997)).\n\n Wash. Rev. Code § 9A.36.021(1) also fails to satisfy the\nsecond part of the residual clause test because it is not\n“similar in kind” to the crimes listed in the enumerated\noffenses clause, which typically involve “purposeful, violent,\nand aggressive conduct.” Adkins, 883 F.3d at 1213 (quoting\nBegay, 553 U.S. at 145). As discussed, the Washington crime\nof second-degree assault can be committed by offensive\ntouching, see Smith, 154 P.3d at 875, which would not\ninvolve “violent” or “aggressive” conduct.\n\n The Washington second-degree assault statute is\nindivisible, thus the modified categorical approach is\ninapplicable. Robinson, 869 F.3d at 941. The district court\nerred by concluding that Door’s prior second-degree assault\nconviction qualifies as a crime of violence pursuant to the\nresidual clause.\n\n\n 8\n For example, a defendant may be convicted of assault with intent to\ncommit a felony for touching a minor’s body in a sexual manner. See\nRobinson, 869 F.3d at 938 n.7.\n\f UNITED STATES V. DOOR 17\n\n D. Remand is Required\n\n The Sentencing Guidelines direct that a base offense level\nof 24 applies if the defendant has two or more felony\nconvictions that qualify as a “crime of violence.” U.S.S.G.\n§ 2K2.1(a)(2). A mistake in calculating the Guidelines\nsentencing range generally requires us to remand for\nresentencing. United States v. Bankston, 901 F.3d 1100, 1107\n(9th Cir. 2018). Door’s conviction for the Washington crime\nof felony harassment qualifies as a crime of violence, but the\ndistrict court erred in holding that Door’s conviction for the\nWashington crime of second-degree assault constitutes a\ncrime of violence. Because Door only had one conviction\nthat qualified as a crime of violence, his base offense level\nshould have been 20. See U.S.S.G. § 2K2.1(a)(4). The\ndistrict court therefore erred in calculating Door’s base\noffense level and remand is required.\n\n Sentence VACATED; REMANDED for resentencing.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. Kenneth Door\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-12\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-30165\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 3:12-cr-05126-RBL-1\\n\\nKENNETH RANDALE DOOR,\\n Defendant-Appellant. OPINION\\n\\n\\n Appeal from the United States District Court\\n for the Western District of Washington\\n Ronald B. Leighton, District Judge, Presiding\\n\\n Argued and Submitted December 7, 2018\\n Seattle, Washington\\n\\n Filed March 12, 2019\\n\\nBefore: Sidney R. Thomas, Chief Judge, and M. Margaret\\n McKeown and Morgan Christen, Circuit Judges.\\n\\n Opinion by Judge Christen\\n\\f2 UNITED STATES V. DOOR\\n\\n SUMMARY*\\n\\n\\n Criminal Law\\n\\n Vacating a sentence and remanding for resentencing, the\\npanel held that, in setting the defendant’s base offense level\\nunder U.S.S.G. § 2K2.1(a)(2), his prior Washington State\\nconviction for felony harassment qualified as a crime of\\nviolence but his prior conviction for second degree assault did\\nnot.\\n\\n Consistent with United States v. Werle, 877 F.3d 879 (9th\\nCir. 2017), the panel held that the defendant’s 1997\\nconviction for felony harassment, in violation of Wash. Rev.\\nCode §§ 9A.46.020(1)(a)(i) and (2)(B), qualified as a crime\\nof violence, as defined in U.S.S.G. § 4B1.2. Applying the\\ncategorical approach, the panel held that the conviction\\nqualified as a crime of violence under § 4B1.2(a)’s force\\nclause because it necessarily entailed the threatened use of\\nviolent physical force.\\n\\n The panel held that the district court erred in concluding\\nthat the defendant’s 2002 conviction for second degree\\nassault, in violation of Wash. Rev. Code § 9A.36.021(1)(c),\\nqualified as a crime of violence. Under United States v.\\nRobinson, 869 F.3d 933 (9th Cir. 2017), the conviction did\\nnot qualify under the force clause of § 4B1.2(a). The panel\\nheld that second degree assault also did not qualify as a crime\\nof violence under § 4B1.2(a)’s residual clause because the\\noffense, in the ordinary case, does not present a serious\\n\\n *\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f UNITED STATES V. DOOR 3\\n\\npotential risk of physical injury to another, and it is not\\nsimilar in kind to the crimes listed in the enumerated offenses\\nclause.\\n\\n The panel therefore vacated the sentence and remanded\\nfor resentencing. The panel addressed other issues in a\\nconcurrently-filed memorandum disposition.\\n\\n\\n COUNSEL\\n\\nCarlton Gunn (argued), Pasadena, California, for Defendant-\\nAppellant.\\n\\nHelen J. Brunner (argued), First Assistant United States\\nAttorney; Annette L. Hayes, United States Attorney; United\\nStates Attorney’s Office, Seattle, Washington; for Plaintiff-\\nAppellee.\\n\\n\\n OPINION\\n\\nCHRISTEN, Circuit Judge:\\n\\n Defendant Kenneth Randale Door appeals the sentence\\nthe district court imposed after he was convicted of several\\noffenses in 2014. At his 2017 sentencing hearing, the district\\ncourt determined that Door’s base offense level should be 24\\npursuant to United States Sentencing Guideline (U.S.S.G.)\\n§ 2K2.1(a)(2) after concluding that Door’s prior Washington\\nstate convictions for second-degree assault and felony\\nharassment qualify as crimes of violence. Door contends\\nthese offenses do not constitute crimes of violence and that\\nhis offense level was thus calculated incorrectly. Consistent\\n\\f4 UNITED STATES V. DOOR\\n\\nwith United States v. Werle, 877 F.3d 879, 884 (9th Cir.\\n2017), we hold that Door’s 1997 conviction for felony\\nharassment, in violation of the Revised Code of Washington\\n(Wash. Rev. Code) §§ 9A.46.020(1)(a)(i) and (2)(b), qualifies\\nas a crime of violence. Door’s argument to the contrary\\ndisregards that the framework for the “crime of violence”\\nanalysis differs depending on whether the prior offense is\\nalleged to constitute a crime of violence pursuant to the force\\nclause, the enumerated offenses clause, or the residual clause\\nof U.S.S.G. § 4B1.2(a). The district court did err, however,\\nin concluding that Door’s 2002 conviction for second-degree\\nassault, in violation of Wash. Rev. Code. § 9A.36.021(1)(c),\\nqualifies as a crime of violence. See United States v.\\nRobinson, 869 F.3d 933, 941 (9th Cir. 2017); see also United\\nStates v. Vederoff, 914 F.3d 1238, 1244–46 (9th Cir. 2019).\\nAccordingly, we vacate Door’s sentence and remand for\\nresentencing.1\\n\\n FACTUAL AND PROCEDURAL BACKGROUND\\n\\n A search of Door’s home in 2011 led to the discovery of\\ntwo handguns, some magazines loaded with ammunition, two\\nmilitary grade ballistic vests (body armor), an explosive\\ndevice known as a “seal bomb,” two digital scales, drug\\npackaging materials, and two drug pipes containing\\nmethamphetamine residue. These discoveries led to Door’s\\nindictment in 2012; he was charged with being a Felon in\\n\\n\\n 1\\n Door also contends that he was improperly convicted of possessing\\nbody armor, challenges the application of an enhancement for possessing\\na firearm in connection with another felony, and argues that his case\\nshould be reassigned if remanded. In a concurrently filed memorandum\\ndisposition, we affirm the body armor conviction and sentencing\\nenhancement, and decline to reassign the case on remand.\\n\\f UNITED STATES V. DOOR 5\\n\\nPossession of a Firearm, a Violent Felon in Possession of\\nBody Armor, and a Felon in Possession of an Explosive. In\\n2014, a jury convicted Door of each count. Door received a\\n300-month sentence, but for reasons not pertinent to this\\nappeal, we vacated Door’s sentence and remanded for\\nresentencing.\\n\\n On remand, the probation officer recommended a base\\noffense level of 24 because the officer concluded that Door’s\\nprior Washington state convictions for second-degree assault\\nand felony harassment constitute crimes of violence. See\\nU.S.S.G. § 2K2.1(a)(2) (providing that a base offense level of\\n24 applies “if the defendant committed any part of the instant\\noffense subsequent to sustaining at least two felony\\nconvictions of either a crime of violence or a controlled\\nsubstance offense”). A total offense level of 32 and Door’s\\ncriminal history category of VI yielded a guideline\\nimprisonment range of 210 to 262 months. The probation\\nofficer recommended 276 months of imprisonment due to\\nDoor’s extensive criminal history. This represented the\\nmaximum sentence for each count, served consecutively.\\nDoor argued that his prior convictions for second-degree\\nassault and felony harassment were not “crime[s] of violence”\\nas defined in U.S.S.G. § 4B1.2, and thus his base offense\\nlevel should only be 14.\\n\\n At the resentencing hearing held in 2017, the district court\\nruled that Door’s prior convictions for second-degree assault\\nand felony harassment qualified as crimes of violence\\npursuant to U.S.S.G. § 4B1.2. The court concluded that Door\\nhad a total offense level of 32 and imposed a sentence of\\n276 months.\\n\\f6 UNITED STATES V. DOOR\\n\\n ANALYSIS\\n\\n We have jurisdiction pursuant to 28 U.S.C. § 1291. “We\\nreview de novo whether a state-law crime constitutes a crime\\nof violence under the [Federal Sentencing] Guidelines.”\\nRobinson, 869 F.3d at 936.\\n\\n To determine whether a prior conviction qualifies as a\\ncrime of violence, we apply the categorical approach first\\noutlined in Taylor v. United States, 495 U.S. 575 (1990). The\\ncategorical approach requires courts to compare the elements\\nof the statute of conviction with the federal definition of\\n“crime of violence” to determine whether the statute of\\nconviction criminalizes a broader range of conduct than the\\nfederal definition captures. United States v. Edling, 895 F.3d\\n1153, 1155 (9th Cir. 2018). The 2013 Sentencing Guidelines\\ndefine the term “crime of violence” as follows:\\n\\n [A]ny offense under federal or state law,\\n punishable by imprisonment for a term\\n exceeding one year, that—\\n\\n (1) has as an element the use, attempted\\n use, or threatened use of physical\\n force against the person of another\\n [known as the force clause or the\\n elements clause], or\\n\\n (2) is burglary of a dwelling, arson, or\\n extortion, involves use of explosives\\n [known as the enumerated offenses\\n clause], or otherwise involves conduct\\n that presents a serious potential risk of\\n\\f UNITED STATES V. DOOR 7\\n\\n physical injury to another [known as\\n the residual clause].\\n\\nU.S.S.G. § 4B1.2(a) (2013).2\\n\\n When determining whether a prior conviction constitutes\\na crime of violence, the precise inquiry differs depending on\\nwhether the offense is alleged to qualify as a crime of\\nviolence pursuant to the force clause, the enumerated offenses\\nclause, or the residual clause. See, e.g., Edling, 895 F.3d at\\n1156–58 (determining whether prior convictions qualified as\\ncrimes of violence pursuant to the force clause and the\\nenumerated offenses clause); United States v. Adkins,\\n883 F.3d 1207, 1213–15 (9th Cir. 2018) (determining\\nwhether prior convictions qualified as crimes of violence\\npursuant to the residual clause). An offense constitutes a\\n“crime of violence” if it qualifies under any one of the three\\nclauses. See Edling, 895 F.3d at 1155. Throughout this\\nanalysis, we “presume that the conviction ‘rested upon\\n[nothing] more than the least of th[e] acts’ criminalized” by\\nthe statute of conviction. Moncrieffe v. Holder, 569 U.S. 184,\\n191–92 (2013) (alterations in original) (quoting Johnson v.\\nUnited States, 559 U.S. 133, 137 (2010)).\\n\\n\\n\\n\\n 2\\n The relevant Sentencing Guidelines are those in effect on the date\\nof the defendant’s first sentencing hearing. See 18 U.S.C. § 3742(g)(1).\\nDoor’s first sentencing hearing occurred when the 2013 Guidelines were\\nin effect. Although the residual clause was later omitted from the\\nGuidelines’ definition of a crime of violence, we do not apply the\\namendment retroactively because it substantively changed the Guidelines.\\nUnited States v. Adkins, 883 F.3d 1207, 1211–12 (9th Cir. 2018).\\n\\f8 UNITED STATES V. DOOR\\n\\n A. Applying the Categorical Analysis to the\\n Sentencing Guidelines’ Force Clause, Enumerated\\n Offenses Clause, and Residual Clause\\n\\n To determine whether a prior conviction qualifies\\npursuant to the force clause, the question is whether the crime\\nof conviction “has as an element the use or threatened use of\\nphysical force against the person of another, with ‘physical\\nforce’ understood to mean in this context ‘violent force—that\\nis, force capable of causing physical pain or injury to another\\nperson.’” Edling, 895 F.3d at 1156 (quoting Johnson,\\n559 U.S. at 140). If the crime of conviction necessarily\\nentails the use or threatened use of violent physical force, it\\nis considered a categorical match for a crime of violence\\npursuant to the force clause of § 4B1.2(a)(1), and the inquiry\\nends. See id.; see also Stokeling v. United States, 139 S. Ct.\\n544, 554–55 (2019).\\n\\n A prior offense constitutes a crime of violence pursuant\\nto the enumerated offenses clause if the elements of one of\\nthe generic federal crimes listed in that clause fully subsume\\nthe elements of the crime of conviction. See Mathis v. United\\nStates, 136 S. Ct. 2243, 2248 (2016); United States v.\\nPeterson, 902 F.3d 1016, 1021–22 (9th Cir. 2018). A generic\\nfederal crime is defined by looking to the common law, the\\nModel Penal Code, treatises, and the laws of other states. See\\nVederoff, 914 F.3d at 1245; United States v. Esparza-\\nHerrera, 557 F.3d 1019, 1022–23 (9th Cir. 2009). If the\\ncrime of conviction falls within the generic federal\\ndefinition—meaning it does not punish a broader range of\\nconduct than the generic offense—the conviction qualifies as\\na crime of violence. See, e.g., Mathis, 136 S. Ct. at 2248.\\n\\f UNITED STATES V. DOOR 9\\n\\n We use a two-part test to assess whether an offense is\\ncategorically a crime of violence pursuant to the residual\\nclause in § 4B1.2(a)(2). Adkins, 883 F.3d at 1213. “First, the\\n‘conduct encompassed by the elements of the offense, in the\\nordinary case, must present a serious potential risk of\\nphysical injury to another.’” Id. (quoting United States v.\\nPark, 649 F.3d 1175, 1177–78 (9th Cir. 2011)). “Second, the\\nprior offense must be ‘roughly similar, in kind as well as in\\ndegree of risk posed’” to the crimes listed in the enumerated\\noffenses clause. Id. (quoting Park, 649 F.3d at 1178). To\\ndetermine whether the offense is “similar in kind” to the\\nlisted crimes, we consider “whether the predicate offense\\ninvolves ‘purposeful, violent, and aggressive conduct.’” Id.\\n(quoting Begay v. United States, 553 U.S. 137, 145 (2008)).\\nBoth criteria must be satisfied for a prior offense to constitute\\na crime of violence pursuant to the residual clause of\\n§ 4B1.2(a)(2). Id.\\n\\n If the statute of conviction does not qualify as a\\ncategorical match pursuant to the force clause, the\\nenumerated offenses clause, or the residual clause, the court\\nconsiders whether the statute of conviction’s elements are\\ndivisible. See, e.g., Edling, 895 F.3d at 1156, 1159; Adkins,\\n883 F.3d at 1215; Robinson, 869 F.3d at 938. A statute is\\ndivisible if it “list[s] elements in the alternative, and thereby\\ndefine[s] multiple crimes.” Mathis, 136 S. Ct. at 2249. If the\\nstatute of conviction is not a categorical match and is\\nindivisible, it is not a crime of violence. See id. at 2248–49.\\nIf the statute of conviction is not a categorical match and is\\ndivisible, then the court applies the modified categorical\\napproach and “looks to a limited class of documents (for\\nexample, the indictment, jury instructions, or plea agreement\\nand colloquy) to determine what crime, with what elements,\\na defendant was convicted of.” Id. at 2249 (citing Shepard v.\\n\\f10 UNITED STATES V. DOOR\\n\\nUnited States, 544 U.S. 13, 26 (2005)). We then apply the\\nforce clause, the enumerated offenses clause, or the residual\\nclause analysis to the specific crime of conviction to\\ndetermine whether it is a crime of violence. See id.\\n\\n B. Door’s Prior Conviction for the Washington\\n Crime of Felony Harassment Qualifies as a Crime\\n of Violence Pursuant to the Force Clause.\\n\\n Door was convicted in 1997 of felony harassment in\\nviolation of Wash. Rev. Code §§ 9A.46.020(1)(a)(i) and\\n(2)(b) for “threatening to kill” a person.3 We previously\\n\\n\\n\\n\\n 3\\n At the time of Door’s conviction in 1997, the Washington\\nharassment statute provided the following:\\n\\n (1) A person is guilty of harassment if:\\n\\n (a) Without lawful authority, the person knowingly\\n threatens:\\n\\n (i) To cause bodily injury in the future to the\\n person threatened or to any other person; or\\n\\n (ii) To cause physical damage to the property\\n of a person other than the actor; or\\n\\n (iii) To subject the person threatened or any\\n other person to physical confinement or\\n restraint; or\\n\\n (iv) Maliciously to do any other act which is\\n intended to substantially harm the person\\n threatened or another with respect to his or her\\n physical or mental health or safety; and\\n\\f UNITED STATES V. DOOR 11\\n\\nanalyzed this statute in Werle, and held that a conviction\\npursuant to the “threat to kill” subsection of the statute\\nconstitutes a crime of violence pursuant to the force clause of\\n§ 4B1.2(a)(1).4 Werle, 877 F.3d at 883–84. Because a\\nconviction pursuant to Wash. Rev. Code § 9A.46.020(2)(b)\\nrequires that a defendant be found guilty of knowingly\\nthreatening to kill a person, this offense necessarily entails\\nthe threatened use of violent physical force against another\\nperson, as required by § 4B1.2(a)(1). Id. Thus, Door’s\\n\\n\\n\\n (b) The person by words or conduct places the\\n person threatened in reasonable fear that the threat\\n will be carried out.\\n\\n (2) A person who harasses another is guilty of a gross\\n misdemeanor punishable under chapter 9A.20 RCW,\\n except that the person is guilty of a class C felony if\\n either of the following applies:\\n\\n (a) The person has previously been convicted in\\n this or any other state of any crime of harassment,\\n as defined in RCW 9A.46.060, of the same victim\\n or members of the victim’s family or household or\\n any person specifically named in a no-contact or\\n no-harassment order; or\\n\\n (b) the person harasses another person under subsection\\n (1)(a)(i) of this section by threatening to kill the\\n person threatened or any other person.\\n\\nWash. Rev. Code § 9A.46.020.\\n 4\\n In Werle, we assessed the amended version of Wash. Rev. Code\\n§ 9A.46.020. 877 F.3d at 882. Although Door’s conviction occurred prior\\nto the statutory amendment, the relevant language in the earlier version of\\nthe felony harassment statute is substantially the same as the amended\\nversion reviewed in Werle.\\n\\f12 UNITED STATES V. DOOR\\n\\nconviction for the Washington crime of felony harassment\\nqualifies as a crime of violence for sentencing purposes.\\n\\n Door contends that we are not bound by Werle’s holding,\\nobserving that in Werle, the defendant never argued his prior\\nWashington felony harassment conviction failed to qualify as\\na crime of violence on account of Washington’s version of\\naiding and abetting, which is broader than the federal\\ndefinition of aiding and abetting. As Door’s argument goes,\\nbecause every Washington criminal statute incorporates\\naiding and abetting, all Washington criminal statutes are\\noverbroad, and therefore all Washington state convictions fail\\nto qualify as crimes of violence—an argument inspired by\\nUnited States v. Valdivia-Flores, 876 F.3d 1201 (9th Cir.\\n2017). Door’s argument, however, overlooks the analytical\\ndifference between the force clause and the enumerated\\noffenses clause of U.S.S.G. § 4B1.2(a).\\n\\n In Valdivia-Flores, this court assessed whether a\\nconviction for violating a Washington drug-trafficking law,\\nWash. Rev. Code § 69.50.401, constituted an “aggravated\\nfelony” for purposes of the Immigration and Nationality Act\\n(INA). Valdivia-Flores, 876 F.3d at 1206. Applying\\nTaylor’s categorical approach, we considered “whether the\\nstate statute defining the crime of conviction categorically\\nfit[] within the generic federal definition of a corresponding\\naggravated felony.” Id. (quoting Roman-Suaste v. Holder,\\n766 F.3d 1035, 1038 (9th Cir. 2014)). Valdivia-Flores’s\\ncrime of conviction was overbroad because: (1) aiding and\\nabetting is included in every Washington criminal statute;\\n(2) Washington defines aiding and abetting more broadly than\\nfederal law because it merely requires knowledge, whereas\\nfederal law requires a mens rea of specific intent; and\\n(3) accomplice and principal liability are indivisible under\\n\\f UNITED STATES V. DOOR 13\\n\\nWashington law. Id. at 1207–10. Concluding that Valdivia-\\nFlores’s prior offense did not qualify as an “aggravated\\nfelony,” we observed that it may be that “no Washington state\\nconviction can serve as an aggravated felony at all.” Id. at\\n1209–10.\\n\\n Door’s reliance on Valdivia-Flores is misplaced. The\\ncategorical analysis in Valdivia-Flores involved comparing\\nthe elements of the Washington drug trafficking crime with\\nthe generic federal offense of drug trafficking because “drug\\ntrafficking” is listed in the INA as an “aggravated felony.”\\nSee id. at 1206–07. In other words, the categorical analysis\\nemployed in Valdivia-Flores mirrors the inquiry under the\\nenumerated offenses clause of U.S.S.G. § 4B1.2(a)(2).\\nWerle, on the other hand, held that a prior conviction for\\nWashington felony harassment constitutes a crime of violence\\npursuant to the force clause of § 4B1.2(a)(1). Because a\\nconviction for violating Wash. Rev. Code § 9A.46.020(2)(b)\\nnecessarily entails the threatened use of violent physical\\nforce, it qualifies as a crime of violence pursuant to the force\\nclause, and our inquiry ends there. We need not compare the\\nelements of the crime of conviction with the elements of the\\ngeneric federal crime when analyzing whether an offense\\nqualifies as a crime of violence pursuant to the force clause\\nof § 4B1.2(a). Compare Peterson, 902 F.3d at 1016,\\n1021–22, with Edling, 895 F.3d at 1156–58. Valdivia-Flores\\nis consistent with Werle.\\n\\f14 UNITED STATES V. DOOR\\n\\n C. Door’s Prior Conviction for the Washington\\n Crime of Second-Degree Assault Does Not Qualify\\n as a Crime of Violence.\\n\\n Door was convicted in 2002 of second-degree assault in\\nviolation of Wash. Rev. Code § 9A.36.021(1)(c).5 In\\nRobinson, we held that Washington’s crime of second-degree\\nassault did not qualify as a crime of violence pursuant to the\\n\\n\\n 5\\n At the time of Door’s conviction in 2002, the Washington second-\\ndegree assault statute provided the following:\\n\\n (1) A person is guilty of assault in the second degree if\\n he or she, under circumstances not amounting to assault\\n in the first degree:\\n\\n (a) Intentionally assaults another and thereby\\n recklessly inflicts substantial bodily harm; or\\n\\n (b) Intentionally and unlawfully causes substantial\\n bodily harm to an unborn quick child by\\n intentionally and unlawfully inflicting any injury\\n upon the mother of such child; or\\n\\n (c) Assaults another with a deadly weapon; or\\n\\n (d) With intent to inflict bodily harm, administers\\n to or causes to be taken by another, poison or any\\n other destructive or noxious substance; or\\n\\n (e) With intent to commit a felony, assaults\\n another; or\\n\\n (f) Knowingly inflicts bodily harm which by\\n design causes such pain or agony as to be the\\n equivalent of that produced by torture.\\n\\nWash. Rev. Code § 9A.36.021(1).\\n\\f UNITED STATES V. DOOR 15\\n\\nforce clause of U.S.S.G. § 4B1.2(a)(1).6 869 F.3d at 937–38.\\nHowever, Robinson did not address whether second-degree\\nassault constitutes a crime of violence pursuant to the residual\\nclause of § 4B1.2(a)(2).7\\n\\n Applying the two-part residual clause test, we hold that\\nWash. Rev. Code § 9A.36.021(1) is not a crime of violence\\nbecause the offense, in the ordinary case, does not “present a\\nserious potential risk of physical injury to another.” Adkins,\\n883 F.3d at 1213 (quoting Park, 649 F.3d at 1177–78). The\\nWashington second-degree assault statute encompasses\\nassault with intent to commit a felony. Wash. Rev. Code\\n§ 9A.36.021(1)(e). This includes the intent to commit any\\nnon-violent felony offense. The “assault” may also be non-\\nviolent because Washington defines assault broadly to\\ninclude “an intentional touching . . . that is harmful or\\noffensive regardless of whether any physical injury is done to\\nthe person.” State v. Smith, 154 P.3d 873, 875 (Wash. 2007)\\n(en banc); see also Robinson, 869 F.3d at 938 n.7. Thus, a\\ndefendant may violate Wash. Rev. Code § 9A.36.021(1)(e) in\\n\\n\\n\\n\\n 6\\n In Robinson, we assessed the amended version of Wash. Rev. Code\\n§ 9A.36.021(1). 869 F.3d at 937 n.6. The statute was amended after\\nDoor’s conviction to include a seventh subsection, (1)(g) which prohibits\\n“[a]ssaults . . . by strangulation or suffocation.” Id.\\n 7\\n Robinson also did not address the enumerated offenses clause. To\\nthe extent the government argues that second-degree assault qualifies as\\na crime of violence pursuant to the enumerated offenses clause, we decline\\nto address this argument, raised for the first time on appeal. See United\\nStates v. Fomichev, 899 F.3d 766, 770 (9th Cir. 2018), as amended by\\n909 F.3d 1078 (9th Cir. 2018). Further, the argument is foreclosed by\\nVederoff. See Vederoff, 914 F.3d at 1246.\\n\\f16 UNITED STATES V. DOOR\\n\\na way that poses no serious risk of physical injury to others.8\\nBecause Door “could have been convicted on the basis of\\nconduct that did not present a serious risk of physical injury\\nto another,” a prior conviction for Washington second-degree\\nassault does not qualify as a crime of violence. See United\\nStates v. Simmons, 782 F.3d 510, 519 (9th Cir. 2015) (quoting\\nUnited States v. Kelly, 422 F.3d 889, 893 (9th Cir. 2005)); see\\nalso United States v. Lee, 821 F.3d 1124, 1128–29 (9th Cir.\\n2016) (holding that California’s crime of battery committed\\nagainst a custodial officer does not qualify as a crime of\\nviolence pursuant to the residual clause because “the least\\ntouching may constitute battery” (quoting People v. Mesce,\\n60 Cal. Rptr. 2d 745, 756 (Cal. Ct. App. 1997)).\\n\\n Wash. Rev. Code § 9A.36.021(1) also fails to satisfy the\\nsecond part of the residual clause test because it is not\\n“similar in kind” to the crimes listed in the enumerated\\noffenses clause, which typically involve “purposeful, violent,\\nand aggressive conduct.” Adkins, 883 F.3d at 1213 (quoting\\nBegay, 553 U.S. at 145). As discussed, the Washington crime\\nof second-degree assault can be committed by offensive\\ntouching, see Smith, 154 P.3d at 875, which would not\\ninvolve “violent” or “aggressive” conduct.\\n\\n The Washington second-degree assault statute is\\nindivisible, thus the modified categorical approach is\\ninapplicable. Robinson, 869 F.3d at 941. The district court\\nerred by concluding that Door’s prior second-degree assault\\nconviction qualifies as a crime of violence pursuant to the\\nresidual clause.\\n\\n\\n 8\\n For example, a defendant may be convicted of assault with intent to\\ncommit a felony for touching a minor’s body in a sexual manner. See\\nRobinson, 869 F.3d at 938 n.7.\\n\\f UNITED STATES V. DOOR 17\\n\\n D. Remand is Required\\n\\n The Sentencing Guidelines direct that a base offense level\\nof 24 applies if the defendant has two or more felony\\nconvictions that qualify as a “crime of violence.” U.S.S.G.\\n§ 2K2.1(a)(2). A mistake in calculating the Guidelines\\nsentencing range generally requires us to remand for\\nresentencing. United States v. Bankston, 901 F.3d 1100, 1107\\n(9th Cir. 2018). Door’s conviction for the Washington crime\\nof felony harassment qualifies as a crime of violence, but the\\ndistrict court erred in holding that Door’s conviction for the\\nWashington crime of second-degree assault constitutes a\\ncrime of violence. Because Door only had one conviction\\nthat qualified as a crime of violence, his base offense level\\nshould have been 20. See U.S.S.G. § 2K2.1(a)(4). The\\ndistrict court therefore erred in calculating Door’s base\\noffense level and remand is required.\\n\\n Sentence VACATED; REMANDED for resentencing.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1280,"cells":{"id":{"kind":"number","value":4598815,"string":"4,598,815"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. David Prien-Pinto"},"decision_date":{"kind":"string","value":"2019-03-12"},"docket_number":{"kind":"string","value":"18-30055"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: William A. Fletcher and Jay S. Bybee, Circuit Judges, and Larry A. Burns,** District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"BYBEE, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n UNITED STATES OF AMERICA, No. 18-30055\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 9:17-cr-00025-DLC-1\\n\\n DAVID PRIEN-PINTO,\\n Defendant-Appellant. OPINION\\n\\n\\n Appeal from the United States District Court\\n for the District of Montana\\n Dana L. Christensen, Chief Judge, Presiding\\n\\n Submitted December 7, 2018*\\n Seattle, Washington\\n\\n Filed March 12, 2019\\n\\n Before: William A. Fletcher and Jay S. Bybee, Circuit\\n Judges, and Larry A. Burns,** District Judge.\\n\\n Opinion by Judge Bybee\\n\\n\\n\\n *\\n The panel unanimously concludes this case is suitable for decision\\nwithout oral argument. See Fed. R. App. P. 34(a)(2).\\n **\\n The Honorable Larry A. Burns, United States District Judge for the\\nSouthern District of California, sitting by designation.\\n\\f2 UNITED STATES V. PRIEN-PINTO\\n\\n SUMMARY***\\n\\n\\n Criminal Law\\n\\n Affirming a sentence, the panel held that the strict-\\nliability enhancement of U.S.S.G. § 2K2.1(b)(4), for the\\ncommission of a crime with a stolen firearm, is constitutional.\\n\\n Joining ten other circuits, the panel reaffirmed the holding\\nof United States v. Goodell, 990 F.3d 497 (9th Cir. 1993), that\\nthe lack of a mens rea requirement in § 2K2.1(b)(4) does not\\nviolate due process. The panel wrote that subsequently-\\nissued Application Note 8(B), directing courts not to impose\\na mens rea requirement, served as confirmation of Goodell’s\\nanalysis of the plain language and legislative history of\\nthe enhancement. Further, the Supreme Court’s Sixth\\nAmendment jurisprudence requiring that all facts leading to\\na sentencing enhancement beyond the statutory maximum be\\nproven to a jury did not overrule the long-settled position that\\nthe Fifth Amendment permits a sentencing enhancement for\\npossession of a stolen firearm to apply on a strict-liability\\nbasis.\\n\\n\\n\\n\\n ***\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f UNITED STATES V. PRIEN-PINTO 3\\n\\n COUNSEL\\n\\nJohn Rhodes, Assistant Federal Defender; Anthony R.\\nGallagher, Federal Defender; Federal Defenders of Montana,\\nMissoula, Montana; for Defendant-Appellant.\\n\\nTimothy A. Tatarka, Assistant United States Attorney; Kurt\\nG. Alme, United States Attorney; United States Attorney’s\\nOffice, Billings, Montana; for Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nBYBEE, Circuit Judge:\\n\\n We are asked to review the vitality of our 1993 holding\\nthat the Sentencing Guidelines may constitutionally impose\\na strict-liability enhancement where a defendant committed\\na crime with a stolen firearm. See United States v. Goodell,\\n990 F.2d 497 (9th Cir. 1993). Since that time, the Supreme\\nCourt has issued a number of opinions recasting the role the\\nGuidelines play in a district court’s sentencing decision. We\\nconclude that none of these decisions affect Goodell. In\\nholding once more that the strict-liability enhancement of\\n§ 2K2.1(b)(4) of the Sentencing Guidelines is constitutional,\\nwe join all ten of the other regional circuit courts.\\n\\n I\\n\\n David Prien-Pinto was convicted in Montana state court\\nin 2014 of felony assault on a peace officer and burglary and\\nsentenced to a term in state prison. He was released on parole\\nin March 2016. After his release, a confidential source\\nalerted a joint task force of federal and local law enforcement\\n\\f4 UNITED STATES V. PRIEN-PINTO\\n\\nofficers that Prien-Pinto was selling narcotics out of his home\\nin Missoula. In September 2016, officers raided the home\\nand arrested Prien-Pinto on marijuana and methamphetamine\\ncharges. Shortly after, Prien-Pinto’s wife reported to local\\npolice that she had hidden a Taurus Model 94 .22 caliber\\nrevolver (“the firearm”) at Prien-Pinto’s instruction. Prien-\\nPinto admitted to possessing the firearm and told police that\\na friend had given him the firearm as payment for a marijuana\\ndebt.\\n\\n Police traced the firearm’s serial number and determined\\nthat it had been stolen the previous summer from its owner in\\nKalispell, about 120 miles north of Missoula. The owner\\nidentified the firearm and told police it had been taken from\\nthe glove compartment of his vehicle during a break-in. The\\nowner denied knowing Prien-Pinto.\\n\\n Montana authorities held Prien-Pinto on a parole\\nviolation. He has remained in state custody since his arrest,\\nserving a prison sentence on various state charges. In August\\n2017, a federal grand jury indicted Prien-Pinto on one count\\nof being a felon in possession of a firearm in violation of\\n18 U.S.C. § 922(g)(1). He pleaded guilty, without a plea\\nagreement, in November 2017.\\n\\n The district court sentenced Prien-Pinto to 36 months’\\nimprisonment: 18 months to be served concurrently to his\\nMontana state sentence, and 18 months to be served\\nconsecutively in federal custody. In calculating Prien-Pinto’s\\nGuidelines offense level, the district court applied a two-level\\nenhancement pursuant to U.S.S.G. § 2K2.1(b)(4) because\\nPrien-Pinto’s crime involved a stolen firearm. Prien-Pinto\\nobjected to this enhancement, arguing that nothing in the\\nrecord suggested he knew the firearm was stolen and that\\n\\f UNITED STATES V. PRIEN-PINTO 5\\n\\napplication of the enhancement without a mens rea would\\nviolate his Fifth Amendment rights, or alternatively, the\\nfederal statutes criminalizing possession of a stolen firearm.\\nThe district court noted that “the facts are pretty compelling\\nthat Mr. Prien-Pinto did not have any knowledge that this\\nfirearm had been stolen.” However, it held that Application\\nNote 8(B) in the Commentary following\\n§ 2K2.1(b)(4)—which provides that the enhancement applies\\n“regardless of whether the defendant knew or had reason to\\nbelieve that the firearm was stolen”—was “not ambiguous.”\\nIt thus applied the enhancement but invited us, on the record,\\nto review the constitutionality of § 2K2.1(b)(4) and the\\nApplication Note.\\n\\n II\\n\\n The constitutionality of U.S.S.G. § 2K2.1(b)(4)’s strict-\\nliability enhancement is the only issue before us on appeal.\\nWe review a claim that the Sentencing Guidelines are\\nunconstitutional de novo. United States v. Padilla-Diaz,\\n862 F.3d 856, 860 (9th Cir. 2017).\\n\\n A\\n\\n All sentencing proceedings begin with the district court’s\\ncalculations of the applicable Guidelines range. See Gall v.\\nUnited States, 552 U.S. 38, 49 (2007). Section 2K2.1\\nprovides the offense levels for crimes like Prien-Pinto’s,\\nwhich involve unlawful receipt, possession, or transportation\\nof firearms. Paragraph (b) provides a number of\\nenhancements that increase the base offense level for firearm\\ncrimes, including this provision in subparagraph (4): “If any\\nfirearm (A) was stolen, increase by 2 levels; or (B) had an\\naltered or obliterated serial number, increase by 4 levels.”\\n\\f6 UNITED STATES V. PRIEN-PINTO\\n\\nU.S.S.G. § 2K2.1(b)(4). In the Commentary section\\nfollowing § 2K2.1, Application Note 8(B) provides:\\n“Knowledge or Reason to Believe.—Subsection (b)(4)\\napplies regardless of whether the defendant knew or had\\nreason to believe that the firearm was stolen or had an altered\\nor obliterated serial number.” U.S.S.G § 2K2.1, App.\\nNote 8(B). In other words, this Application Note directs\\ncourts to apply § 2K2.1(b)(4)’s enhancements on a strict\\nliability basis and not to read a mens rea requirement into the\\ntext of the provision.\\n\\n Application Notes are not formally part of the Guidelines,\\nbut serve to “interpret[]” and “explain[]” the Guidelines for\\ndistrict courts. Stinson v. United States, 508 U.S. 36, 38\\n(1993). Like the Guidelines themselves, the Application\\nNotes are drafted by the Sentencing Commission and may be\\nrevised annually. See id. at 41, 45. However, Congress lacks\\nthe power to modify or disapprove of Application Notes, as\\nit may Guidelines. Id. at 40. The Supreme Court told us in\\nStinson that an Application Note “that interprets or explains\\na guideline is authoritative unless it violates the Constitution\\nor a federal statute, or is inconsistent with, or a plainly\\nerroneous reading of, that guideline.” Id.; see also id. at\\n42–43 (“Commentary which functions to interpret a guideline\\nor explain how it is to be applied controls, and if failure to\\nfollow, or a misreading of, such commentary results in a\\nsentence selected from the wrong guideline range, that\\nsentence would constitute an incorrect application of the\\nsentencing guidelines.” (internal citations, punctuation, and\\nalterations omitted)). The Court also analogized the\\nSentencing Commission’s use of Application Notes to an\\nagency’s interpretation of its own legislative rules, and thus\\ndirected courts to apply Application Notes with the same\\nforce as other such interpretations. Id. at 44–45. Thus, we\\n\\f UNITED STATES V. PRIEN-PINTO 7\\n\\nascribe somewhat less legal weight to the Application Notes\\nthan to the Guidelines proper: if the Guideline and\\nApplication Note are inconsistent, the Guideline prevails. See\\nid. at 38.\\n\\n After publishing Stinson in 1993, the Supreme Court\\nissued a series of opinions that have dramatically altered the\\nrole the Guidelines play in sentencing. Most significantly,\\nthe Court struck down the federal sentencing statute making\\nthe Guidelines mandatory, finding it inconsistent with the\\nSixth Amendment. United States v. Booker, 543 U.S. 220,\\n245 (2005). However, we and the Supreme Court have\\ncontinued to apply the holding of Stinson in the years\\nfollowing Booker. See, e.g., Freeman v. United States,\\n564 U.S. 522, 529 (2011) (citing Stinson, 508 U.S. at 38)\\n(“Guidelines commentary is authoritative[.]”); United States\\nv. Thornton, 444 F.3d 1163, 1165 n.3 (9th Cir. 2006)\\n(“[C]ommentary in the Guidelines Manual that interprets or\\nexplains a guideline is authoritative unless it violates the\\nConstitution or a federal statute, or is inconsistent with, or a\\nplainly erroneous reading of, that guideline.” (internal\\nquotation marks omitted)). We thus follow Stinson in\\nconsidering whether the district court properly applied\\n§ 2K2.1(b)(4) and Application Note 8(B) when calculating\\nPrien-Pinto’s sentence.\\n\\n B\\n\\n We considered and rejected a Fifth Amendment due\\nprocess challenge to the lack of a mens rea requirement in\\n§ 2K2.1(b)’s enhancement for a stolen firearm in Goodell. At\\nthat time, the stolen-firearm enhancement was codified at\\n§ 2K2.1(b)(2), and the Sentencing Commission had not yet\\ndrafted Application Note 8(B), which first appeared in the\\n\\f8 UNITED STATES V. PRIEN-PINTO\\n\\nGuidelines in 2007. Compare U.S.S.G. § 2K2.1 (1990) with\\nU.S.S.G. § 2K2.1 (2007).\\n\\n Working without the “authoritative,” Stinson, 508 U.S. at\\n38, weight of this Application Note, we reached the result it\\nnow prescribes through sound construction of the Guidelines.\\nWe first scrutinized“[t]he plain language” of § 2K2.1(b) and\\nfound that it “does not require [a mens rea.]” Goodell,\\n990 F.2d at 498. Second, we examined the drafting history to\\ndetermine whether the Guideline’s text was a “deliberate”\\nexception to the “rule of . . . Anglo-American criminal\\njurisprudence” that a penalty requires a mens rea. See id. at\\n498–99. Our review found that the Sentencing Commission\\n“promulgated [the enhancement] on the premise that ‘stolen\\nfirearms are used disproportionately in the commission of\\ncrimes.’” Id. at 499 (quoting United States v. Mobley,\\n956 F.2d 450, 454 (3d Cir. 1992)). Because the Fifth\\nAmendment permits a regulatory statute in the interest of\\npublic safety to provide for criminal liability in the absence\\nof mens rea, United States v. Freed, 401 U.S. 601, 607–09\\n(1971), and because an enhancement for possession of a\\nstolen firearm is rationally related to the goal of crime\\nprevention, Goodell, 990 F.2d at 499 (citing Mobley,\\n956 F.2d at 454) (noting that stolen firearms are\\ndisproportionately used in the commission of crimes), we\\nheld that there was no Fifth Amendment violation. Id. In the\\nend, strict liability “reflects the greater culpability of an ex-\\nfelon possessing a stolen weapon.” Id. at 501.\\n\\n Goodell’s analysis of the plain language and legislative\\nhistory of the enhancement seems to foreclose Prien-Pinto’s\\nargument that Application Note 8(B) is “a plainly erroneous\\nreading” of § 2K2.1(b)(4). See Stinson, 508 U.S. at 38.\\nThrough traditional techniques of construction, we had been\\n\\f UNITED STATES V. PRIEN-PINTO 9\\n\\nreading this enhancement to apply without a mens rea for\\nfourteen years before the Sentencing Commission began\\ndirecting us to do so. Application Note 8(B) simply serves as\\nconfirmation that Goodell’s reading has always been the\\ncorrect one.\\n\\n C\\n\\n Prien-Pinto raises several arguments that Goodell does\\nnot control this case. He first argues that the Supreme\\nCourt’s decision in Booker overruled Stinson,\\nand—apparently—all other pre-Booker decisions about the\\nconstitutionality of the Guidelines, including Goodell.\\nStinson’s holding that the Application Notes carry authority\\nakin to an agency’s interpretation of its legislative rules, he\\navers, is now incorrect, and courts would thus violate the\\nAdministrative Procedure Act by giving the Application\\nNotes the force of law because they are not promulgated\\nthrough notice-and-comment rulemaking.\\n\\n This argument misunderstands the case law. We relied\\nupon neither Stinson nor the Application Notes in deciding in\\nGoodell that the stolen-firearm enhancement does not carry\\na mens rea; rather, we based that holding on the text of\\n§ 2K2.1(b) alone. Thus, an overruling of Stinson would not\\nabrogate our holding in Goodell. Moreover, Stinson has not\\nbeen overruled. The Supreme Court, post-Booker, continues\\nto cite its holding regarding the Application Notes as positive\\nlaw—as do we. See, e.g., Freeman, 564 U.S. at 529;\\nThornton, 444 F.3d at 1165 n.3.\\n\\n Second, Prien-Pinto argues that Goodell is no longer good\\nlaw because that decision relied on McMillan v.\\nPennsylvania, 477 U.S. 79 (1986). McMillan established a\\n\\f10 UNITED STATES V. PRIEN-PINTO\\n\\nthree-factor test to determine whether a sentencing\\nenhancement factor violates the Fifth Amendment. Goodell,\\n990 F.2d at 499–500 (citing McMillan, 477 U.S. at 87–88).\\nFourteen years after McMillan, the Supreme Court interpreted\\nthe Sixth Amendment to require that sentencing\\nenhancements beyond the statutory maximum be based on\\nfacts proven to a jury beyond a reasonable doubt. Apprendi\\nv. New Jersey, 530 U.S. 466, 486–87, 490 (2000). The Court\\nsubsequently clarified that Apprendi did not disturb the\\nMcMillan due process standard in Harris v. United States,\\n536 U.S. 545, 563 (2002). However, eleven years later, the\\nCourt overruled Harris and questioned the “vitality, if not\\nvalidity” of McMillan. Alleyne v. United States, 570 U.S. 99,\\n106 (2013); see also id. at 118 (Sotomayor, J., concurring)\\n(expressing a view that McMillan was “wrongly decided”).\\nPrien-Pinto argues that if McMillan is no longer valid,\\nGoodell cannot be either.\\n\\n We need not reach the question of McMillan’s continued\\nvitality because, when we decided the portion of Goodell\\nwhich controls this appeal, we neither relied on nor engaged\\nwith McMillan. Only after determining that the stolen-\\nfirearm enhancement did not carry a mens rea and that this\\nstrict-liability enhancement did not violate the Fifth\\nAmendment did we turn to Mr. Goodell’s separate argument\\nthat the enhancement was invalid under the standard of\\nMcMillan. Goodell, 990 F.2d at 499. In fact, we even\\ntransitioned into the McMillan portion of our opinion with the\\nphrase “[t]he scienter issue aside . . . .” Id. Our subsequent\\nanalysis of whether the enhancement altered the statutory\\nmaximum penalty for Goodell’s crime is not relevant to\\nPrien-Pinto’s appeal.\\n\\f UNITED STATES V. PRIEN-PINTO 11\\n\\n Third, Prien-Pinto urges us to distinguish Goodell because\\nthere, unlike here, the government and the defendant\\nstipulated to the fact that the defendant did not know the\\nfirearm he possessed was stolen. Id. at 498. However,\\nnothing in our analysis or holding in Goodell relied on this\\nstipulation. Moreover, the district court here stated on the\\nrecord at sentencing that “the facts are pretty compelling that\\nMr. Prien-Pinto did not have any knowledge that this firearm\\nhad been stolen,” a finding we consider of equivalent force to\\na stipulation.\\n\\n Fourth, Prien-Pinto argues that § 2K2.1(b)(4)’s strict-\\nliability enhancement is in tension with the federal statute\\ncriminalizing possession of a stolen firearm, which does\\nrequire a mens rea. See 18 U.S.C. § 922(j) (making it a crime\\nto possess a stolen firearm or ammunition while “knowing or\\nhaving reasonable cause to believe that the firearm or\\nammunition was stolen”). Because Application Note 8(B)\\nmay not stand if it “violates . . . a federal statute,” Stinson,\\n508 U.S. at 38, Prien-Pinto avers that the statute’s\\nrequirement of a mens rea disallows the Guidelines from\\nimposing a strict-liability penalty for the same conduct. See\\nUnited States v. LaBonte, 520 U.S. 751, 757 (1997) (holding\\nthat where an Application Note and the plain language of a\\nfederal statute are “at odds,” the Application Note must “must\\ngive way”). However, we considered and rejected this\\nstatutory argument in United States v. Ellsworth, 456 F.3d\\n1146, 1150–51 (9th Cir. 2006), which we published after the\\nSupreme Court’s decision in Booker.\\n\\n In Ellsworth, the defendant argued, like Prien-Pinto, that\\n§ 2K2.1(b)(4) was “at odds with Congress’ plain language as\\nexpressed in 18 U.S.C. § 922(j).” Id. at 1151 (internal\\nquotation marks omitted). We, however, distinguished\\n\\f12 UNITED STATES V. PRIEN-PINTO\\n\\nbetween § 922(j) as “an independent basis for criminal\\nliability” and § 2K2.1(b)(4) as a mere “enhancement” applied\\nonly after liability has been established for the separate crime\\nof being a felon in possession of a firearm, describing the\\nprovisions as only “superficially similar.” See id. The\\nenhancement, we reasoned, does not establish culpability for\\npossession of a stolen firearm, but increases the penalty for a\\n“small[] and distinct category of persons” who “may not\\nlegitimately possess any firearm,” stolen or not: convicted\\nfelons. Id. (emphasis in original); see also Goodell, 990 F.2d\\nat 501 (describing the “greater culpability” of a felon who\\npossesses a stolen firearm). The provisions are thus\\nfundamentally different, and we cannot assume that Congress\\nintended to include an unwritten mens rea for the Guidelines\\nenhancement because it included a written one for the\\nstatutory offense. Guessing at Congress’s intent where the\\ntext is clear on its face is a dangerous game, and one that we\\nhave chosen not to play. Ellsworth thus has foreclosed Prien-\\nPinto’s statutory argument.\\n\\n Finally, Prien-Pinto argues that Booker represented a “sea\\nchange” which fundamentally altered due process\\njurisprudence on the Guidelines and compels a re-evaluation\\nof Goodell. He relies on a lone opinion from the Eastern\\nDistrict of New York, United States v. Handy, 570 F. Supp.\\n2d 437 (E.D.N.Y. 2008), the logic of which has been rejected\\nby the Second, Third, Sixth, and Eleventh Circuits.1 See\\nUnited States v. Thomas, 628 F.3d 64, 72 (2d Cir. 2010) (re-\\naffirming a pre-Booker decision holding the lack of a mens\\nrea requirement to § 2K2.1(b)(4) valid and noting “respectful\\ndisagreement” with Handy); see United States v. Stokes,\\n\\n 1\\n The government did not appeal Handy to the Second Circuit; thus,\\nthat court did not have an opportunity to directly review the decision.\\n\\f UNITED STATES V. PRIEN-PINTO 13\\n\\n621 F. App’x 316, 320 (6th Cir. 2015) (rejecting Handy);\\nUnited States v. Fields, 608 F. App’x 806, 812 (11th Cir.\\n2015) (same); United States v. Black, 386 F. App’x 238, 241\\n(3d Cir. 2010) (same).\\n\\n Further, although the First, Fourth, Fifth, Seventh, Eighth,\\nand Tenth Circuits have not addressed the Handy opinion,\\neach has continued to apply § 2K2.1(b)(4)’s strict-liability\\nenhancement in the years after Booker. See United States v.\\nGonzalez, 857 F.3d 46, 55–56 (1st Cir. 2017) (“[U]nlike a\\nmens-rea-less criminal statute, the Guidelines may compound\\nthe punishment for the offense, but they fall far short of\\ncriminalizing apparently innocent conduct. . . . The history of\\nthe Guideline itself confirms that the omission was\\nintentional[.]” (internal quotation marks and alteration\\nomitted)); United States v. Taylor, 659 F.3d 339, 344 (4th\\nCir. 2011) (“[I]t is not unusual to punish individuals for the\\nunintended consequences of their unlawful acts.” (quoting\\nDean v. United States, 556 U.S. 568, 575 (2009))); United\\nStates v. Perez, 585 F.3d 880, 883 (5th Cir. 2009) (“This\\ncourt has continually enforced the clear and unambiguous\\nlanguage of § 2K2.1(b)(4) and its strict liability standard.”);\\nUnited States v. Statham, 581 F.3d 548, 553 (7th Cir. 2009)\\n(applying § 2K2.1(b)(4)’s parallel strict-liability enhancement\\nfor use of a firearm with an altered or obliterated serial\\nnumber); United States v. Bates, 584 F.3d 1105, 1109 (8th\\nCir. 2009) (applying Application Note 8(B) to impose the\\nstrict-liability enhancement); United States v. Ellis, 351 F.\\nApp’x 307, 309 (10th Cir. 2009) (“We like other circuits have\\nlong and unanimously held that Section 2K2.1(b)(4) lacks a\\nscienter requirement.”).\\n\\f14 UNITED STATES V. PRIEN-PINTO\\n\\n III\\n\\n We join the unanimous ranks of these ten circuits and\\ndecline to reverse our longstanding precedent. We fail to\\nunderstand how the Supreme Court’s Sixth Amendment\\njurisprudence requiring that all facts leading to a sentence\\nenhancement beyond the statutory maximum be proven to a\\njury overrules the long-settled position that the Fifth\\nAmendment permits a sentencing enhancement for possession\\nof a stolen firearm to apply on a strict-liability basis. For the\\nforegoing reasons, the judgment of the district court is\\nAFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4376068/\", \"author_raw\": \"BYBEE, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"WILLIAM A FLETCHER"},"judge2":{"kind":"string","value":"JAY S BYBEE"},"judge3":{"kind":"string","value":"LARRY A BURNS"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4598815/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. David Prien-Pinto\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-12\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n UNITED STATES OF AMERICA, No. 18-30055\n Plaintiff-Appellee,\n D.C. No.\n v. 9:17-cr-00025-DLC-1\n\n DAVID PRIEN-PINTO,\n Defendant-Appellant. OPINION\n\n\n Appeal from the United States District Court\n for the District of Montana\n Dana L. Christensen, Chief Judge, Presiding\n\n Submitted December 7, 2018*\n Seattle, Washington\n\n Filed March 12, 2019\n\n Before: William A. Fletcher and Jay S. Bybee, Circuit\n Judges, and Larry A. Burns,** District Judge.\n\n Opinion by Judge Bybee\n\n\n\n *\n The panel unanimously concludes this case is suitable for decision\nwithout oral argument. See Fed. R. App. P. 34(a)(2).\n **\n The Honorable Larry A. Burns, United States District Judge for the\nSouthern District of California, sitting by designation.\n\f2 UNITED STATES V. PRIEN-PINTO\n\n SUMMARY***\n\n\n Criminal Law\n\n Affirming a sentence, the panel held that the strict-\nliability enhancement of U.S.S.G. § 2K2.1(b)(4), for the\ncommission of a crime with a stolen firearm, is constitutional.\n\n Joining ten other circuits, the panel reaffirmed the holding\nof United States v. Goodell, 990 F.3d 497 (9th Cir. 1993), that\nthe lack of a mens rea requirement in § 2K2.1(b)(4) does not\nviolate due process. The panel wrote that subsequently-\nissued Application Note 8(B), directing courts not to impose\na mens rea requirement, served as confirmation of Goodell’s\nanalysis of the plain language and legislative history of\nthe enhancement. Further, the Supreme Court’s Sixth\nAmendment jurisprudence requiring that all facts leading to\na sentencing enhancement beyond the statutory maximum be\nproven to a jury did not overrule the long-settled position that\nthe Fifth Amendment permits a sentencing enhancement for\npossession of a stolen firearm to apply on a strict-liability\nbasis.\n\n\n\n\n ***\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\f UNITED STATES V. PRIEN-PINTO 3\n\n COUNSEL\n\nJohn Rhodes, Assistant Federal Defender; Anthony R.\nGallagher, Federal Defender; Federal Defenders of Montana,\nMissoula, Montana; for Defendant-Appellant.\n\nTimothy A. Tatarka, Assistant United States Attorney; Kurt\nG. Alme, United States Attorney; United States Attorney’s\nOffice, Billings, Montana; for Plaintiff-Appellee.\n\n\n OPINION\n\nBYBEE, Circuit Judge:\n\n We are asked to review the vitality of our 1993 holding\nthat the Sentencing Guidelines may constitutionally impose\na strict-liability enhancement where a defendant committed\na crime with a stolen firearm. See United States v. Goodell,\n990 F.2d 497 (9th Cir. 1993). Since that time, the Supreme\nCourt has issued a number of opinions recasting the role the\nGuidelines play in a district court’s sentencing decision. We\nconclude that none of these decisions affect Goodell. In\nholding once more that the strict-liability enhancement of\n§ 2K2.1(b)(4) of the Sentencing Guidelines is constitutional,\nwe join all ten of the other regional circuit courts.\n\n I\n\n David Prien-Pinto was convicted in Montana state court\nin 2014 of felony assault on a peace officer and burglary and\nsentenced to a term in state prison. He was released on parole\nin March 2016. After his release, a confidential source\nalerted a joint task force of federal and local law enforcement\n\f4 UNITED STATES V. PRIEN-PINTO\n\nofficers that Prien-Pinto was selling narcotics out of his home\nin Missoula. In September 2016, officers raided the home\nand arrested Prien-Pinto on marijuana and methamphetamine\ncharges. Shortly after, Prien-Pinto’s wife reported to local\npolice that she had hidden a Taurus Model 94 .22 caliber\nrevolver (“the firearm”) at Prien-Pinto’s instruction. Prien-\nPinto admitted to possessing the firearm and told police that\na friend had given him the firearm as payment for a marijuana\ndebt.\n\n Police traced the firearm’s serial number and determined\nthat it had been stolen the previous summer from its owner in\nKalispell, about 120 miles north of Missoula. The owner\nidentified the firearm and told police it had been taken from\nthe glove compartment of his vehicle during a break-in. The\nowner denied knowing Prien-Pinto.\n\n Montana authorities held Prien-Pinto on a parole\nviolation. He has remained in state custody since his arrest,\nserving a prison sentence on various state charges. In August\n2017, a federal grand jury indicted Prien-Pinto on one count\nof being a felon in possession of a firearm in violation of\n18 U.S.C. § 922(g)(1). He pleaded guilty, without a plea\nagreement, in November 2017.\n\n The district court sentenced Prien-Pinto to 36 months’\nimprisonment: 18 months to be served concurrently to his\nMontana state sentence, and 18 months to be served\nconsecutively in federal custody. In calculating Prien-Pinto’s\nGuidelines offense level, the district court applied a two-level\nenhancement pursuant to U.S.S.G. § 2K2.1(b)(4) because\nPrien-Pinto’s crime involved a stolen firearm. Prien-Pinto\nobjected to this enhancement, arguing that nothing in the\nrecord suggested he knew the firearm was stolen and that\n\f UNITED STATES V. PRIEN-PINTO 5\n\napplication of the enhancement without a mens rea would\nviolate his Fifth Amendment rights, or alternatively, the\nfederal statutes criminalizing possession of a stolen firearm.\nThe district court noted that “the facts are pretty compelling\nthat Mr. Prien-Pinto did not have any knowledge that this\nfirearm had been stolen.” However, it held that Application\nNote 8(B) in the Commentary following\n§ 2K2.1(b)(4)—which provides that the enhancement applies\n“regardless of whether the defendant knew or had reason to\nbelieve that the firearm was stolen”—was “not ambiguous.”\nIt thus applied the enhancement but invited us, on the record,\nto review the constitutionality of § 2K2.1(b)(4) and the\nApplication Note.\n\n II\n\n The constitutionality of U.S.S.G. § 2K2.1(b)(4)’s strict-\nliability enhancement is the only issue before us on appeal.\nWe review a claim that the Sentencing Guidelines are\nunconstitutional de novo. United States v. Padilla-Diaz,\n862 F.3d 856, 860 (9th Cir. 2017).\n\n A\n\n All sentencing proceedings begin with the district court’s\ncalculations of the applicable Guidelines range. See Gall v.\nUnited States, 552 U.S. 38, 49 (2007). Section 2K2.1\nprovides the offense levels for crimes like Prien-Pinto’s,\nwhich involve unlawful receipt, possession, or transportation\nof firearms. Paragraph (b) provides a number of\nenhancements that increase the base offense level for firearm\ncrimes, including this provision in subparagraph (4): “If any\nfirearm (A) was stolen, increase by 2 levels; or (B) had an\naltered or obliterated serial number, increase by 4 levels.”\n\f6 UNITED STATES V. PRIEN-PINTO\n\nU.S.S.G. § 2K2.1(b)(4). In the Commentary section\nfollowing § 2K2.1, Application Note 8(B) provides:\n“Knowledge or Reason to Believe.—Subsection (b)(4)\napplies regardless of whether the defendant knew or had\nreason to believe that the firearm was stolen or had an altered\nor obliterated serial number.” U.S.S.G § 2K2.1, App.\nNote 8(B). In other words, this Application Note directs\ncourts to apply § 2K2.1(b)(4)’s enhancements on a strict\nliability basis and not to read a mens rea requirement into the\ntext of the provision.\n\n Application Notes are not formally part of the Guidelines,\nbut serve to “interpret[]” and “explain[]” the Guidelines for\ndistrict courts. Stinson v. United States, 508 U.S. 36, 38\n(1993). Like the Guidelines themselves, the Application\nNotes are drafted by the Sentencing Commission and may be\nrevised annually. See id. at 41, 45. However, Congress lacks\nthe power to modify or disapprove of Application Notes, as\nit may Guidelines. Id. at 40. The Supreme Court told us in\nStinson that an Application Note “that interprets or explains\na guideline is authoritative unless it violates the Constitution\nor a federal statute, or is inconsistent with, or a plainly\nerroneous reading of, that guideline.” Id.; see also id. at\n42–43 (“Commentary which functions to interpret a guideline\nor explain how it is to be applied controls, and if failure to\nfollow, or a misreading of, such commentary results in a\nsentence selected from the wrong guideline range, that\nsentence would constitute an incorrect application of the\nsentencing guidelines.” (internal citations, punctuation, and\nalterations omitted)). The Court also analogized the\nSentencing Commission’s use of Application Notes to an\nagency’s interpretation of its own legislative rules, and thus\ndirected courts to apply Application Notes with the same\nforce as other such interpretations. Id. at 44–45. Thus, we\n\f UNITED STATES V. PRIEN-PINTO 7\n\nascribe somewhat less legal weight to the Application Notes\nthan to the Guidelines proper: if the Guideline and\nApplication Note are inconsistent, the Guideline prevails. See\nid. at 38.\n\n After publishing Stinson in 1993, the Supreme Court\nissued a series of opinions that have dramatically altered the\nrole the Guidelines play in sentencing. Most significantly,\nthe Court struck down the federal sentencing statute making\nthe Guidelines mandatory, finding it inconsistent with the\nSixth Amendment. United States v. Booker, 543 U.S. 220,\n245 (2005). However, we and the Supreme Court have\ncontinued to apply the holding of Stinson in the years\nfollowing Booker. See, e.g., Freeman v. United States,\n564 U.S. 522, 529 (2011) (citing Stinson, 508 U.S. at 38)\n(“Guidelines commentary is authoritative[.]”); United States\nv. Thornton, 444 F.3d 1163, 1165 n.3 (9th Cir. 2006)\n(“[C]ommentary in the Guidelines Manual that interprets or\nexplains a guideline is authoritative unless it violates the\nConstitution or a federal statute, or is inconsistent with, or a\nplainly erroneous reading of, that guideline.” (internal\nquotation marks omitted)). We thus follow Stinson in\nconsidering whether the district court properly applied\n§ 2K2.1(b)(4) and Application Note 8(B) when calculating\nPrien-Pinto’s sentence.\n\n B\n\n We considered and rejected a Fifth Amendment due\nprocess challenge to the lack of a mens rea requirement in\n§ 2K2.1(b)’s enhancement for a stolen firearm in Goodell. At\nthat time, the stolen-firearm enhancement was codified at\n§ 2K2.1(b)(2), and the Sentencing Commission had not yet\ndrafted Application Note 8(B), which first appeared in the\n\f8 UNITED STATES V. PRIEN-PINTO\n\nGuidelines in 2007. Compare U.S.S.G. § 2K2.1 (1990) with\nU.S.S.G. § 2K2.1 (2007).\n\n Working without the “authoritative,” Stinson, 508 U.S. at\n38, weight of this Application Note, we reached the result it\nnow prescribes through sound construction of the Guidelines.\nWe first scrutinized“[t]he plain language” of § 2K2.1(b) and\nfound that it “does not require [a mens rea.]” Goodell,\n990 F.2d at 498. Second, we examined the drafting history to\ndetermine whether the Guideline’s text was a “deliberate”\nexception to the “rule of . . . Anglo-American criminal\njurisprudence” that a penalty requires a mens rea. See id. at\n498–99. Our review found that the Sentencing Commission\n“promulgated [the enhancement] on the premise that ‘stolen\nfirearms are used disproportionately in the commission of\ncrimes.’” Id. at 499 (quoting United States v. Mobley,\n956 F.2d 450, 454 (3d Cir. 1992)). Because the Fifth\nAmendment permits a regulatory statute in the interest of\npublic safety to provide for criminal liability in the absence\nof mens rea, United States v. Freed, 401 U.S. 601, 607–09\n(1971), and because an enhancement for possession of a\nstolen firearm is rationally related to the goal of crime\nprevention, Goodell, 990 F.2d at 499 (citing Mobley,\n956 F.2d at 454) (noting that stolen firearms are\ndisproportionately used in the commission of crimes), we\nheld that there was no Fifth Amendment violation. Id. In the\nend, strict liability “reflects the greater culpability of an ex-\nfelon possessing a stolen weapon.” Id. at 501.\n\n Goodell’s analysis of the plain language and legislative\nhistory of the enhancement seems to foreclose Prien-Pinto’s\nargument that Application Note 8(B) is “a plainly erroneous\nreading” of § 2K2.1(b)(4). See Stinson, 508 U.S. at 38.\nThrough traditional techniques of construction, we had been\n\f UNITED STATES V. PRIEN-PINTO 9\n\nreading this enhancement to apply without a mens rea for\nfourteen years before the Sentencing Commission began\ndirecting us to do so. Application Note 8(B) simply serves as\nconfirmation that Goodell’s reading has always been the\ncorrect one.\n\n C\n\n Prien-Pinto raises several arguments that Goodell does\nnot control this case. He first argues that the Supreme\nCourt’s decision in Booker overruled Stinson,\nand—apparently—all other pre-Booker decisions about the\nconstitutionality of the Guidelines, including Goodell.\nStinson’s holding that the Application Notes carry authority\nakin to an agency’s interpretation of its legislative rules, he\navers, is now incorrect, and courts would thus violate the\nAdministrative Procedure Act by giving the Application\nNotes the force of law because they are not promulgated\nthrough notice-and-comment rulemaking.\n\n This argument misunderstands the case law. We relied\nupon neither Stinson nor the Application Notes in deciding in\nGoodell that the stolen-firearm enhancement does not carry\na mens rea; rather, we based that holding on the text of\n§ 2K2.1(b) alone. Thus, an overruling of Stinson would not\nabrogate our holding in Goodell. Moreover, Stinson has not\nbeen overruled. The Supreme Court, post-Booker, continues\nto cite its holding regarding the Application Notes as positive\nlaw—as do we. See, e.g., Freeman, 564 U.S. at 529;\nThornton, 444 F.3d at 1165 n.3.\n\n Second, Prien-Pinto argues that Goodell is no longer good\nlaw because that decision relied on McMillan v.\nPennsylvania, 477 U.S. 79 (1986). McMillan established a\n\f10 UNITED STATES V. PRIEN-PINTO\n\nthree-factor test to determine whether a sentencing\nenhancement factor violates the Fifth Amendment. Goodell,\n990 F.2d at 499–500 (citing McMillan, 477 U.S. at 87–88).\nFourteen years after McMillan, the Supreme Court interpreted\nthe Sixth Amendment to require that sentencing\nenhancements beyond the statutory maximum be based on\nfacts proven to a jury beyond a reasonable doubt. Apprendi\nv. New Jersey, 530 U.S. 466, 486–87, 490 (2000). The Court\nsubsequently clarified that Apprendi did not disturb the\nMcMillan due process standard in Harris v. United States,\n536 U.S. 545, 563 (2002). However, eleven years later, the\nCourt overruled Harris and questioned the “vitality, if not\nvalidity” of McMillan. Alleyne v. United States, 570 U.S. 99,\n106 (2013); see also id. at 118 (Sotomayor, J., concurring)\n(expressing a view that McMillan was “wrongly decided”).\nPrien-Pinto argues that if McMillan is no longer valid,\nGoodell cannot be either.\n\n We need not reach the question of McMillan’s continued\nvitality because, when we decided the portion of Goodell\nwhich controls this appeal, we neither relied on nor engaged\nwith McMillan. Only after determining that the stolen-\nfirearm enhancement did not carry a mens rea and that this\nstrict-liability enhancement did not violate the Fifth\nAmendment did we turn to Mr. Goodell’s separate argument\nthat the enhancement was invalid under the standard of\nMcMillan. Goodell, 990 F.2d at 499. In fact, we even\ntransitioned into the McMillan portion of our opinion with the\nphrase “[t]he scienter issue aside . . . .” Id. Our subsequent\nanalysis of whether the enhancement altered the statutory\nmaximum penalty for Goodell’s crime is not relevant to\nPrien-Pinto’s appeal.\n\f UNITED STATES V. PRIEN-PINTO 11\n\n Third, Prien-Pinto urges us to distinguish Goodell because\nthere, unlike here, the government and the defendant\nstipulated to the fact that the defendant did not know the\nfirearm he possessed was stolen. Id. at 498. However,\nnothing in our analysis or holding in Goodell relied on this\nstipulation. Moreover, the district court here stated on the\nrecord at sentencing that “the facts are pretty compelling that\nMr. Prien-Pinto did not have any knowledge that this firearm\nhad been stolen,” a finding we consider of equivalent force to\na stipulation.\n\n Fourth, Prien-Pinto argues that § 2K2.1(b)(4)’s strict-\nliability enhancement is in tension with the federal statute\ncriminalizing possession of a stolen firearm, which does\nrequire a mens rea. See 18 U.S.C. § 922(j) (making it a crime\nto possess a stolen firearm or ammunition while “knowing or\nhaving reasonable cause to believe that the firearm or\nammunition was stolen”). Because Application Note 8(B)\nmay not stand if it “violates . . . a federal statute,” Stinson,\n508 U.S. at 38, Prien-Pinto avers that the statute’s\nrequirement of a mens rea disallows the Guidelines from\nimposing a strict-liability penalty for the same conduct. See\nUnited States v. LaBonte, 520 U.S. 751, 757 (1997) (holding\nthat where an Application Note and the plain language of a\nfederal statute are “at odds,” the Application Note must “must\ngive way”). However, we considered and rejected this\nstatutory argument in United States v. Ellsworth, 456 F.3d\n1146, 1150–51 (9th Cir. 2006), which we published after the\nSupreme Court’s decision in Booker.\n\n In Ellsworth, the defendant argued, like Prien-Pinto, that\n§ 2K2.1(b)(4) was “at odds with Congress’ plain language as\nexpressed in 18 U.S.C. § 922(j).” Id. at 1151 (internal\nquotation marks omitted). We, however, distinguished\n\f12 UNITED STATES V. PRIEN-PINTO\n\nbetween § 922(j) as “an independent basis for criminal\nliability” and § 2K2.1(b)(4) as a mere “enhancement” applied\nonly after liability has been established for the separate crime\nof being a felon in possession of a firearm, describing the\nprovisions as only “superficially similar.” See id. The\nenhancement, we reasoned, does not establish culpability for\npossession of a stolen firearm, but increases the penalty for a\n“small[] and distinct category of persons” who “may not\nlegitimately possess any firearm,” stolen or not: convicted\nfelons. Id. (emphasis in original); see also Goodell, 990 F.2d\nat 501 (describing the “greater culpability” of a felon who\npossesses a stolen firearm). The provisions are thus\nfundamentally different, and we cannot assume that Congress\nintended to include an unwritten mens rea for the Guidelines\nenhancement because it included a written one for the\nstatutory offense. Guessing at Congress’s intent where the\ntext is clear on its face is a dangerous game, and one that we\nhave chosen not to play. Ellsworth thus has foreclosed Prien-\nPinto’s statutory argument.\n\n Finally, Prien-Pinto argues that Booker represented a “sea\nchange” which fundamentally altered due process\njurisprudence on the Guidelines and compels a re-evaluation\nof Goodell. He relies on a lone opinion from the Eastern\nDistrict of New York, United States v. Handy, 570 F. Supp.\n2d 437 (E.D.N.Y. 2008), the logic of which has been rejected\nby the Second, Third, Sixth, and Eleventh Circuits.1 See\nUnited States v. Thomas, 628 F.3d 64, 72 (2d Cir. 2010) (re-\naffirming a pre-Booker decision holding the lack of a mens\nrea requirement to § 2K2.1(b)(4) valid and noting “respectful\ndisagreement” with Handy); see United States v. Stokes,\n\n 1\n The government did not appeal Handy to the Second Circuit; thus,\nthat court did not have an opportunity to directly review the decision.\n\f UNITED STATES V. PRIEN-PINTO 13\n\n621 F. App’x 316, 320 (6th Cir. 2015) (rejecting Handy);\nUnited States v. Fields, 608 F. App’x 806, 812 (11th Cir.\n2015) (same); United States v. Black, 386 F. App’x 238, 241\n(3d Cir. 2010) (same).\n\n Further, although the First, Fourth, Fifth, Seventh, Eighth,\nand Tenth Circuits have not addressed the Handy opinion,\neach has continued to apply § 2K2.1(b)(4)’s strict-liability\nenhancement in the years after Booker. See United States v.\nGonzalez, 857 F.3d 46, 55–56 (1st Cir. 2017) (“[U]nlike a\nmens-rea-less criminal statute, the Guidelines may compound\nthe punishment for the offense, but they fall far short of\ncriminalizing apparently innocent conduct. . . . The history of\nthe Guideline itself confirms that the omission was\nintentional[.]” (internal quotation marks and alteration\nomitted)); United States v. Taylor, 659 F.3d 339, 344 (4th\nCir. 2011) (“[I]t is not unusual to punish individuals for the\nunintended consequences of their unlawful acts.” (quoting\nDean v. United States, 556 U.S. 568, 575 (2009))); United\nStates v. Perez, 585 F.3d 880, 883 (5th Cir. 2009) (“This\ncourt has continually enforced the clear and unambiguous\nlanguage of § 2K2.1(b)(4) and its strict liability standard.”);\nUnited States v. Statham, 581 F.3d 548, 553 (7th Cir. 2009)\n(applying § 2K2.1(b)(4)’s parallel strict-liability enhancement\nfor use of a firearm with an altered or obliterated serial\nnumber); United States v. Bates, 584 F.3d 1105, 1109 (8th\nCir. 2009) (applying Application Note 8(B) to impose the\nstrict-liability enhancement); United States v. Ellis, 351 F.\nApp’x 307, 309 (10th Cir. 2009) (“We like other circuits have\nlong and unanimously held that Section 2K2.1(b)(4) lacks a\nscienter requirement.”).\n\f14 UNITED STATES V. PRIEN-PINTO\n\n III\n\n We join the unanimous ranks of these ten circuits and\ndecline to reverse our longstanding precedent. We fail to\nunderstand how the Supreme Court’s Sixth Amendment\njurisprudence requiring that all facts leading to a sentence\nenhancement beyond the statutory maximum be proven to a\njury overrules the long-settled position that the Fifth\nAmendment permits a sentencing enhancement for possession\nof a stolen firearm to apply on a strict-liability basis. For the\nforegoing reasons, the judgment of the district court is\nAFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. David Prien-Pinto\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-12\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n UNITED STATES OF AMERICA, No. 18-30055\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 9:17-cr-00025-DLC-1\\n\\n DAVID PRIEN-PINTO,\\n Defendant-Appellant. OPINION\\n\\n\\n Appeal from the United States District Court\\n for the District of Montana\\n Dana L. Christensen, Chief Judge, Presiding\\n\\n Submitted December 7, 2018*\\n Seattle, Washington\\n\\n Filed March 12, 2019\\n\\n Before: William A. Fletcher and Jay S. Bybee, Circuit\\n Judges, and Larry A. Burns,** District Judge.\\n\\n Opinion by Judge Bybee\\n\\n\\n\\n *\\n The panel unanimously concludes this case is suitable for decision\\nwithout oral argument. See Fed. R. App. P. 34(a)(2).\\n **\\n The Honorable Larry A. Burns, United States District Judge for the\\nSouthern District of California, sitting by designation.\\n\\f2 UNITED STATES V. PRIEN-PINTO\\n\\n SUMMARY***\\n\\n\\n Criminal Law\\n\\n Affirming a sentence, the panel held that the strict-\\nliability enhancement of U.S.S.G. § 2K2.1(b)(4), for the\\ncommission of a crime with a stolen firearm, is constitutional.\\n\\n Joining ten other circuits, the panel reaffirmed the holding\\nof United States v. Goodell, 990 F.3d 497 (9th Cir. 1993), that\\nthe lack of a mens rea requirement in § 2K2.1(b)(4) does not\\nviolate due process. The panel wrote that subsequently-\\nissued Application Note 8(B), directing courts not to impose\\na mens rea requirement, served as confirmation of Goodell’s\\nanalysis of the plain language and legislative history of\\nthe enhancement. Further, the Supreme Court’s Sixth\\nAmendment jurisprudence requiring that all facts leading to\\na sentencing enhancement beyond the statutory maximum be\\nproven to a jury did not overrule the long-settled position that\\nthe Fifth Amendment permits a sentencing enhancement for\\npossession of a stolen firearm to apply on a strict-liability\\nbasis.\\n\\n\\n\\n\\n ***\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f UNITED STATES V. PRIEN-PINTO 3\\n\\n COUNSEL\\n\\nJohn Rhodes, Assistant Federal Defender; Anthony R.\\nGallagher, Federal Defender; Federal Defenders of Montana,\\nMissoula, Montana; for Defendant-Appellant.\\n\\nTimothy A. Tatarka, Assistant United States Attorney; Kurt\\nG. Alme, United States Attorney; United States Attorney’s\\nOffice, Billings, Montana; for Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nBYBEE, Circuit Judge:\\n\\n We are asked to review the vitality of our 1993 holding\\nthat the Sentencing Guidelines may constitutionally impose\\na strict-liability enhancement where a defendant committed\\na crime with a stolen firearm. See United States v. Goodell,\\n990 F.2d 497 (9th Cir. 1993). Since that time, the Supreme\\nCourt has issued a number of opinions recasting the role the\\nGuidelines play in a district court’s sentencing decision. We\\nconclude that none of these decisions affect Goodell. In\\nholding once more that the strict-liability enhancement of\\n§ 2K2.1(b)(4) of the Sentencing Guidelines is constitutional,\\nwe join all ten of the other regional circuit courts.\\n\\n I\\n\\n David Prien-Pinto was convicted in Montana state court\\nin 2014 of felony assault on a peace officer and burglary and\\nsentenced to a term in state prison. He was released on parole\\nin March 2016. After his release, a confidential source\\nalerted a joint task force of federal and local law enforcement\\n\\f4 UNITED STATES V. PRIEN-PINTO\\n\\nofficers that Prien-Pinto was selling narcotics out of his home\\nin Missoula. In September 2016, officers raided the home\\nand arrested Prien-Pinto on marijuana and methamphetamine\\ncharges. Shortly after, Prien-Pinto’s wife reported to local\\npolice that she had hidden a Taurus Model 94 .22 caliber\\nrevolver (“the firearm”) at Prien-Pinto’s instruction. Prien-\\nPinto admitted to possessing the firearm and told police that\\na friend had given him the firearm as payment for a marijuana\\ndebt.\\n\\n Police traced the firearm’s serial number and determined\\nthat it had been stolen the previous summer from its owner in\\nKalispell, about 120 miles north of Missoula. The owner\\nidentified the firearm and told police it had been taken from\\nthe glove compartment of his vehicle during a break-in. The\\nowner denied knowing Prien-Pinto.\\n\\n Montana authorities held Prien-Pinto on a parole\\nviolation. He has remained in state custody since his arrest,\\nserving a prison sentence on various state charges. In August\\n2017, a federal grand jury indicted Prien-Pinto on one count\\nof being a felon in possession of a firearm in violation of\\n18 U.S.C. § 922(g)(1). He pleaded guilty, without a plea\\nagreement, in November 2017.\\n\\n The district court sentenced Prien-Pinto to 36 months’\\nimprisonment: 18 months to be served concurrently to his\\nMontana state sentence, and 18 months to be served\\nconsecutively in federal custody. In calculating Prien-Pinto’s\\nGuidelines offense level, the district court applied a two-level\\nenhancement pursuant to U.S.S.G. § 2K2.1(b)(4) because\\nPrien-Pinto’s crime involved a stolen firearm. Prien-Pinto\\nobjected to this enhancement, arguing that nothing in the\\nrecord suggested he knew the firearm was stolen and that\\n\\f UNITED STATES V. PRIEN-PINTO 5\\n\\napplication of the enhancement without a mens rea would\\nviolate his Fifth Amendment rights, or alternatively, the\\nfederal statutes criminalizing possession of a stolen firearm.\\nThe district court noted that “the facts are pretty compelling\\nthat Mr. Prien-Pinto did not have any knowledge that this\\nfirearm had been stolen.” However, it held that Application\\nNote 8(B) in the Commentary following\\n§ 2K2.1(b)(4)—which provides that the enhancement applies\\n“regardless of whether the defendant knew or had reason to\\nbelieve that the firearm was stolen”—was “not ambiguous.”\\nIt thus applied the enhancement but invited us, on the record,\\nto review the constitutionality of § 2K2.1(b)(4) and the\\nApplication Note.\\n\\n II\\n\\n The constitutionality of U.S.S.G. § 2K2.1(b)(4)’s strict-\\nliability enhancement is the only issue before us on appeal.\\nWe review a claim that the Sentencing Guidelines are\\nunconstitutional de novo. United States v. Padilla-Diaz,\\n862 F.3d 856, 860 (9th Cir. 2017).\\n\\n A\\n\\n All sentencing proceedings begin with the district court’s\\ncalculations of the applicable Guidelines range. See Gall v.\\nUnited States, 552 U.S. 38, 49 (2007). Section 2K2.1\\nprovides the offense levels for crimes like Prien-Pinto’s,\\nwhich involve unlawful receipt, possession, or transportation\\nof firearms. Paragraph (b) provides a number of\\nenhancements that increase the base offense level for firearm\\ncrimes, including this provision in subparagraph (4): “If any\\nfirearm (A) was stolen, increase by 2 levels; or (B) had an\\naltered or obliterated serial number, increase by 4 levels.”\\n\\f6 UNITED STATES V. PRIEN-PINTO\\n\\nU.S.S.G. § 2K2.1(b)(4). In the Commentary section\\nfollowing § 2K2.1, Application Note 8(B) provides:\\n“Knowledge or Reason to Believe.—Subsection (b)(4)\\napplies regardless of whether the defendant knew or had\\nreason to believe that the firearm was stolen or had an altered\\nor obliterated serial number.” U.S.S.G § 2K2.1, App.\\nNote 8(B). In other words, this Application Note directs\\ncourts to apply § 2K2.1(b)(4)’s enhancements on a strict\\nliability basis and not to read a mens rea requirement into the\\ntext of the provision.\\n\\n Application Notes are not formally part of the Guidelines,\\nbut serve to “interpret[]” and “explain[]” the Guidelines for\\ndistrict courts. Stinson v. United States, 508 U.S. 36, 38\\n(1993). Like the Guidelines themselves, the Application\\nNotes are drafted by the Sentencing Commission and may be\\nrevised annually. See id. at 41, 45. However, Congress lacks\\nthe power to modify or disapprove of Application Notes, as\\nit may Guidelines. Id. at 40. The Supreme Court told us in\\nStinson that an Application Note “that interprets or explains\\na guideline is authoritative unless it violates the Constitution\\nor a federal statute, or is inconsistent with, or a plainly\\nerroneous reading of, that guideline.” Id.; see also id. at\\n42–43 (“Commentary which functions to interpret a guideline\\nor explain how it is to be applied controls, and if failure to\\nfollow, or a misreading of, such commentary results in a\\nsentence selected from the wrong guideline range, that\\nsentence would constitute an incorrect application of the\\nsentencing guidelines.” (internal citations, punctuation, and\\nalterations omitted)). The Court also analogized the\\nSentencing Commission’s use of Application Notes to an\\nagency’s interpretation of its own legislative rules, and thus\\ndirected courts to apply Application Notes with the same\\nforce as other such interpretations. Id. at 44–45. Thus, we\\n\\f UNITED STATES V. PRIEN-PINTO 7\\n\\nascribe somewhat less legal weight to the Application Notes\\nthan to the Guidelines proper: if the Guideline and\\nApplication Note are inconsistent, the Guideline prevails. See\\nid. at 38.\\n\\n After publishing Stinson in 1993, the Supreme Court\\nissued a series of opinions that have dramatically altered the\\nrole the Guidelines play in sentencing. Most significantly,\\nthe Court struck down the federal sentencing statute making\\nthe Guidelines mandatory, finding it inconsistent with the\\nSixth Amendment. United States v. Booker, 543 U.S. 220,\\n245 (2005). However, we and the Supreme Court have\\ncontinued to apply the holding of Stinson in the years\\nfollowing Booker. See, e.g., Freeman v. United States,\\n564 U.S. 522, 529 (2011) (citing Stinson, 508 U.S. at 38)\\n(“Guidelines commentary is authoritative[.]”); United States\\nv. Thornton, 444 F.3d 1163, 1165 n.3 (9th Cir. 2006)\\n(“[C]ommentary in the Guidelines Manual that interprets or\\nexplains a guideline is authoritative unless it violates the\\nConstitution or a federal statute, or is inconsistent with, or a\\nplainly erroneous reading of, that guideline.” (internal\\nquotation marks omitted)). We thus follow Stinson in\\nconsidering whether the district court properly applied\\n§ 2K2.1(b)(4) and Application Note 8(B) when calculating\\nPrien-Pinto’s sentence.\\n\\n B\\n\\n We considered and rejected a Fifth Amendment due\\nprocess challenge to the lack of a mens rea requirement in\\n§ 2K2.1(b)’s enhancement for a stolen firearm in Goodell. At\\nthat time, the stolen-firearm enhancement was codified at\\n§ 2K2.1(b)(2), and the Sentencing Commission had not yet\\ndrafted Application Note 8(B), which first appeared in the\\n\\f8 UNITED STATES V. PRIEN-PINTO\\n\\nGuidelines in 2007. Compare U.S.S.G. § 2K2.1 (1990) with\\nU.S.S.G. § 2K2.1 (2007).\\n\\n Working without the “authoritative,” Stinson, 508 U.S. at\\n38, weight of this Application Note, we reached the result it\\nnow prescribes through sound construction of the Guidelines.\\nWe first scrutinized“[t]he plain language” of § 2K2.1(b) and\\nfound that it “does not require [a mens rea.]” Goodell,\\n990 F.2d at 498. Second, we examined the drafting history to\\ndetermine whether the Guideline’s text was a “deliberate”\\nexception to the “rule of . . . Anglo-American criminal\\njurisprudence” that a penalty requires a mens rea. See id. at\\n498–99. Our review found that the Sentencing Commission\\n“promulgated [the enhancement] on the premise that ‘stolen\\nfirearms are used disproportionately in the commission of\\ncrimes.’” Id. at 499 (quoting United States v. Mobley,\\n956 F.2d 450, 454 (3d Cir. 1992)). Because the Fifth\\nAmendment permits a regulatory statute in the interest of\\npublic safety to provide for criminal liability in the absence\\nof mens rea, United States v. Freed, 401 U.S. 601, 607–09\\n(1971), and because an enhancement for possession of a\\nstolen firearm is rationally related to the goal of crime\\nprevention, Goodell, 990 F.2d at 499 (citing Mobley,\\n956 F.2d at 454) (noting that stolen firearms are\\ndisproportionately used in the commission of crimes), we\\nheld that there was no Fifth Amendment violation. Id. In the\\nend, strict liability “reflects the greater culpability of an ex-\\nfelon possessing a stolen weapon.” Id. at 501.\\n\\n Goodell’s analysis of the plain language and legislative\\nhistory of the enhancement seems to foreclose Prien-Pinto’s\\nargument that Application Note 8(B) is “a plainly erroneous\\nreading” of § 2K2.1(b)(4). See Stinson, 508 U.S. at 38.\\nThrough traditional techniques of construction, we had been\\n\\f UNITED STATES V. PRIEN-PINTO 9\\n\\nreading this enhancement to apply without a mens rea for\\nfourteen years before the Sentencing Commission began\\ndirecting us to do so. Application Note 8(B) simply serves as\\nconfirmation that Goodell’s reading has always been the\\ncorrect one.\\n\\n C\\n\\n Prien-Pinto raises several arguments that Goodell does\\nnot control this case. He first argues that the Supreme\\nCourt’s decision in Booker overruled Stinson,\\nand—apparently—all other pre-Booker decisions about the\\nconstitutionality of the Guidelines, including Goodell.\\nStinson’s holding that the Application Notes carry authority\\nakin to an agency’s interpretation of its legislative rules, he\\navers, is now incorrect, and courts would thus violate the\\nAdministrative Procedure Act by giving the Application\\nNotes the force of law because they are not promulgated\\nthrough notice-and-comment rulemaking.\\n\\n This argument misunderstands the case law. We relied\\nupon neither Stinson nor the Application Notes in deciding in\\nGoodell that the stolen-firearm enhancement does not carry\\na mens rea; rather, we based that holding on the text of\\n§ 2K2.1(b) alone. Thus, an overruling of Stinson would not\\nabrogate our holding in Goodell. Moreover, Stinson has not\\nbeen overruled. The Supreme Court, post-Booker, continues\\nto cite its holding regarding the Application Notes as positive\\nlaw—as do we. See, e.g., Freeman, 564 U.S. at 529;\\nThornton, 444 F.3d at 1165 n.3.\\n\\n Second, Prien-Pinto argues that Goodell is no longer good\\nlaw because that decision relied on McMillan v.\\nPennsylvania, 477 U.S. 79 (1986). McMillan established a\\n\\f10 UNITED STATES V. PRIEN-PINTO\\n\\nthree-factor test to determine whether a sentencing\\nenhancement factor violates the Fifth Amendment. Goodell,\\n990 F.2d at 499–500 (citing McMillan, 477 U.S. at 87–88).\\nFourteen years after McMillan, the Supreme Court interpreted\\nthe Sixth Amendment to require that sentencing\\nenhancements beyond the statutory maximum be based on\\nfacts proven to a jury beyond a reasonable doubt. Apprendi\\nv. New Jersey, 530 U.S. 466, 486–87, 490 (2000). The Court\\nsubsequently clarified that Apprendi did not disturb the\\nMcMillan due process standard in Harris v. United States,\\n536 U.S. 545, 563 (2002). However, eleven years later, the\\nCourt overruled Harris and questioned the “vitality, if not\\nvalidity” of McMillan. Alleyne v. United States, 570 U.S. 99,\\n106 (2013); see also id. at 118 (Sotomayor, J., concurring)\\n(expressing a view that McMillan was “wrongly decided”).\\nPrien-Pinto argues that if McMillan is no longer valid,\\nGoodell cannot be either.\\n\\n We need not reach the question of McMillan’s continued\\nvitality because, when we decided the portion of Goodell\\nwhich controls this appeal, we neither relied on nor engaged\\nwith McMillan. Only after determining that the stolen-\\nfirearm enhancement did not carry a mens rea and that this\\nstrict-liability enhancement did not violate the Fifth\\nAmendment did we turn to Mr. Goodell’s separate argument\\nthat the enhancement was invalid under the standard of\\nMcMillan. Goodell, 990 F.2d at 499. In fact, we even\\ntransitioned into the McMillan portion of our opinion with the\\nphrase “[t]he scienter issue aside . . . .” Id. Our subsequent\\nanalysis of whether the enhancement altered the statutory\\nmaximum penalty for Goodell’s crime is not relevant to\\nPrien-Pinto’s appeal.\\n\\f UNITED STATES V. PRIEN-PINTO 11\\n\\n Third, Prien-Pinto urges us to distinguish Goodell because\\nthere, unlike here, the government and the defendant\\nstipulated to the fact that the defendant did not know the\\nfirearm he possessed was stolen. Id. at 498. However,\\nnothing in our analysis or holding in Goodell relied on this\\nstipulation. Moreover, the district court here stated on the\\nrecord at sentencing that “the facts are pretty compelling that\\nMr. Prien-Pinto did not have any knowledge that this firearm\\nhad been stolen,” a finding we consider of equivalent force to\\na stipulation.\\n\\n Fourth, Prien-Pinto argues that § 2K2.1(b)(4)’s strict-\\nliability enhancement is in tension with the federal statute\\ncriminalizing possession of a stolen firearm, which does\\nrequire a mens rea. See 18 U.S.C. § 922(j) (making it a crime\\nto possess a stolen firearm or ammunition while “knowing or\\nhaving reasonable cause to believe that the firearm or\\nammunition was stolen”). Because Application Note 8(B)\\nmay not stand if it “violates . . . a federal statute,” Stinson,\\n508 U.S. at 38, Prien-Pinto avers that the statute’s\\nrequirement of a mens rea disallows the Guidelines from\\nimposing a strict-liability penalty for the same conduct. See\\nUnited States v. LaBonte, 520 U.S. 751, 757 (1997) (holding\\nthat where an Application Note and the plain language of a\\nfederal statute are “at odds,” the Application Note must “must\\ngive way”). However, we considered and rejected this\\nstatutory argument in United States v. Ellsworth, 456 F.3d\\n1146, 1150–51 (9th Cir. 2006), which we published after the\\nSupreme Court’s decision in Booker.\\n\\n In Ellsworth, the defendant argued, like Prien-Pinto, that\\n§ 2K2.1(b)(4) was “at odds with Congress’ plain language as\\nexpressed in 18 U.S.C. § 922(j).” Id. at 1151 (internal\\nquotation marks omitted). We, however, distinguished\\n\\f12 UNITED STATES V. PRIEN-PINTO\\n\\nbetween § 922(j) as “an independent basis for criminal\\nliability” and § 2K2.1(b)(4) as a mere “enhancement” applied\\nonly after liability has been established for the separate crime\\nof being a felon in possession of a firearm, describing the\\nprovisions as only “superficially similar.” See id. The\\nenhancement, we reasoned, does not establish culpability for\\npossession of a stolen firearm, but increases the penalty for a\\n“small[] and distinct category of persons” who “may not\\nlegitimately possess any firearm,” stolen or not: convicted\\nfelons. Id. (emphasis in original); see also Goodell, 990 F.2d\\nat 501 (describing the “greater culpability” of a felon who\\npossesses a stolen firearm). The provisions are thus\\nfundamentally different, and we cannot assume that Congress\\nintended to include an unwritten mens rea for the Guidelines\\nenhancement because it included a written one for the\\nstatutory offense. Guessing at Congress’s intent where the\\ntext is clear on its face is a dangerous game, and one that we\\nhave chosen not to play. Ellsworth thus has foreclosed Prien-\\nPinto’s statutory argument.\\n\\n Finally, Prien-Pinto argues that Booker represented a “sea\\nchange” which fundamentally altered due process\\njurisprudence on the Guidelines and compels a re-evaluation\\nof Goodell. He relies on a lone opinion from the Eastern\\nDistrict of New York, United States v. Handy, 570 F. Supp.\\n2d 437 (E.D.N.Y. 2008), the logic of which has been rejected\\nby the Second, Third, Sixth, and Eleventh Circuits.1 See\\nUnited States v. Thomas, 628 F.3d 64, 72 (2d Cir. 2010) (re-\\naffirming a pre-Booker decision holding the lack of a mens\\nrea requirement to § 2K2.1(b)(4) valid and noting “respectful\\ndisagreement” with Handy); see United States v. Stokes,\\n\\n 1\\n The government did not appeal Handy to the Second Circuit; thus,\\nthat court did not have an opportunity to directly review the decision.\\n\\f UNITED STATES V. PRIEN-PINTO 13\\n\\n621 F. App’x 316, 320 (6th Cir. 2015) (rejecting Handy);\\nUnited States v. Fields, 608 F. App’x 806, 812 (11th Cir.\\n2015) (same); United States v. Black, 386 F. App’x 238, 241\\n(3d Cir. 2010) (same).\\n\\n Further, although the First, Fourth, Fifth, Seventh, Eighth,\\nand Tenth Circuits have not addressed the Handy opinion,\\neach has continued to apply § 2K2.1(b)(4)’s strict-liability\\nenhancement in the years after Booker. See United States v.\\nGonzalez, 857 F.3d 46, 55–56 (1st Cir. 2017) (“[U]nlike a\\nmens-rea-less criminal statute, the Guidelines may compound\\nthe punishment for the offense, but they fall far short of\\ncriminalizing apparently innocent conduct. . . . The history of\\nthe Guideline itself confirms that the omission was\\nintentional[.]” (internal quotation marks and alteration\\nomitted)); United States v. Taylor, 659 F.3d 339, 344 (4th\\nCir. 2011) (“[I]t is not unusual to punish individuals for the\\nunintended consequences of their unlawful acts.” (quoting\\nDean v. United States, 556 U.S. 568, 575 (2009))); United\\nStates v. Perez, 585 F.3d 880, 883 (5th Cir. 2009) (“This\\ncourt has continually enforced the clear and unambiguous\\nlanguage of § 2K2.1(b)(4) and its strict liability standard.”);\\nUnited States v. Statham, 581 F.3d 548, 553 (7th Cir. 2009)\\n(applying § 2K2.1(b)(4)’s parallel strict-liability enhancement\\nfor use of a firearm with an altered or obliterated serial\\nnumber); United States v. Bates, 584 F.3d 1105, 1109 (8th\\nCir. 2009) (applying Application Note 8(B) to impose the\\nstrict-liability enhancement); United States v. Ellis, 351 F.\\nApp’x 307, 309 (10th Cir. 2009) (“We like other circuits have\\nlong and unanimously held that Section 2K2.1(b)(4) lacks a\\nscienter requirement.”).\\n\\f14 UNITED STATES V. PRIEN-PINTO\\n\\n III\\n\\n We join the unanimous ranks of these ten circuits and\\ndecline to reverse our longstanding precedent. We fail to\\nunderstand how the Supreme Court’s Sixth Amendment\\njurisprudence requiring that all facts leading to a sentence\\nenhancement beyond the statutory maximum be proven to a\\njury overrules the long-settled position that the Fifth\\nAmendment permits a sentencing enhancement for possession\\nof a stolen firearm to apply on a strict-liability basis. For the\\nforegoing reasons, the judgment of the district court is\\nAFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1281,"cells":{"id":{"kind":"number","value":4598816,"string":"4,598,816"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Franks Landing Indian Cmty. v. National Indian Gaming Comm'r."},"decision_date":{"kind":"string","value":"2019-03-12"},"docket_number":{"kind":"string","value":"17-35368"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: N. Randy Smith and Morgan Christen, Circuit Judges, and Robert E. Payne,* District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"CHRISTEN, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nFRANK’S LANDING INDIAN No. 17-35368\\nCOMMUNITY, a federally-\\nrecognized self-governing D.C. No.\\ndependent Indian community, 3:15-cv-05828-BHS\\n Plaintiff-Appellant,\\n\\n v. OPINION\\n\\nNATIONAL INDIAN GAMING\\nCOMMISSION; UNITED STATES\\nDEPARTMENT OF THE\\nINTERIOR; JONODEV\\nCHAUDHURI, in his official\\ncapacity as Chairman of the\\nNational Indian Gaming\\nCommission; TARA KATUK\\nMACLEAN SWEENEY, in her\\nofficial capacity as Assistant\\nSecretary of the Interior\\nIndian Affairs, United States\\nDepartment of the Interior;\\nDAVID BERNHARDT, in his\\nofficial capacity as Acting\\nSecretary of the Interior,\\n Defendants-Appellees.\\n\\f2 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n Appeal from the United States District Court\\n for the Western District of Washington\\n Benjamin H. Settle, District Judge, Presiding\\n\\n Argued and Submitted October 11, 2018\\n Seattle, Washington\\n\\n Filed March 12, 2019\\n\\n Before: N. Randy Smith and Morgan Christen, Circuit\\n Judges, and Robert E. Payne,* District Judge.\\n\\n Opinion by Judge Christen\\n\\n\\n SUMMARY**\\n\\n\\n Tribal Matters\\n\\n The panel affirmed the district court’s summary judgment\\nentered in favor of the U.S. Department of the Interior, its\\nSecretary, and Assistant Secretary of Indian Affairs, in an\\naction challenging the Department of the Interior’s\\ndetermination that Frank’s Landing Indian Community is\\nineligible for gaming for purposes of the Indian Gaming\\nRegulatory Act (“IGRA”).\\n\\n\\n *\\n The Honorable Robert E. Payne, United States District Judge for the\\nEastern District of Virginia, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f FRANKS LANDING INDIAN COMTY. V. 3\\n NAT’L INDIAN GAMING COMM’N\\n\\n The panel held that IGRA clearly and unambiguously\\nrequired federal recognition by the Secretary of the Interior\\nbefore a tribe may qualify to participate in Indian gaming.\\nThe panel also held that when Congress amended the Frank’s\\nLanding Act in 1994, it did not separately authorize the non-\\nfederally recognized Frank’s Landing Indian Community to\\nengage in Class II gaming. The panel held that the\\nDepartment of the Interior correctly concluded that the\\nCommunity was not eligible to engage in IGRA gaming.\\n\\n\\n COUNSEL\\n\\nScott David Crowell (argued), Crowell Law Office-Tribal\\nAdvocacy Group, Sedona, Arizona; Bryan T. Newland,\\nFletcher Law PLLC, East Lansing, Michigan; Alan S.\\nMiddleton, Law Offices of Alan S. Middleton PLLC, Lake\\nForest Park, Washington; for Plaintiff-Appellant.\\n\\nKevin W. McArdle (argued), Mary Gabrielle Sprague, and\\nDevon Lehman McCune, Environment and Natural Resources\\nDivision; Eric Grant, Deputy Assistant Attorney General;\\nJeffrey H. Wood, Acting Assistant Attorney General; United\\nStates Department of Justice, Washington, D.C.; Samuel E.\\nEnnis, Office of the Solicitor, Department of the Interior,\\nWashington, D.C.; for Defendants-Appellees.\\n\\f4 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n OPINION\\n\\nCHRISTEN, Circuit Judge:\\n\\n Frank’s Landing Indian Community (Community)\\nappeals the district court’s order granting summary judgment\\nin favor of appellees Department of the Interior, its Secretary,\\nand the Assistant Secretary–Indian Affairs. The Community,\\nwhich is not a federally recognized tribe, challenges Interior’s\\ndetermination that it is ineligible for gaming for purposes of\\nthe Indian Gaming Regulatory Act, (IGRA), 25 U.S.C.\\n§§ 2701–2721. The Community argues that its unique status,\\nrecognized and defined in the Frank’s Landing Act1 and the\\n1994 amendments to that Act,2 renders it eligible to engage in\\nclass II gaming.\\n\\n We have jurisdiction pursuant to 28 U.S.C. § 1291 and we\\naffirm the district court’s judgment. Reviewing Interior’s\\ndecision under the Administrative Procedure Act (APA), we\\nconclude the agency’s determination was correct. First, we\\nhold that IGRA clearly and unambiguously requires federal\\nrecognition by the Secretary of the Department of the Interior\\nbefore a tribe may qualify to participate in Indian gaming.\\nSecond, we hold that the Frank’s Landing Act does not\\nauthorize the Community to engage in class II gaming.\\n\\n\\n\\n\\n 1\\n Pub. L. No. 100-153, § 10, 101 Stat. 886, 889.\\n 2\\n Pub. L. No. 103-435, § 8, 108 Stat. 4566, 4569–70.\\n\\f FRANKS LANDING INDIAN COMTY. V. 5\\n NAT’L INDIAN GAMING COMM’N\\n\\n BACKGROUND\\n\\n A. Statutory Background\\n\\n Adopted in 1987, the Frank’s Landing Act originally\\nprovided as follows:\\n\\n The Frank’s Landing Indian Community in\\n the State of Washington is hereby recognized\\n as eligible for the special programs and\\n services provided by the United States to\\n Indians because of their status as Indians and\\n is recognized as eligible to contract, and to\\n receive grants, under the Indian Self-\\n Determination and Education Assistance Act\\n for such services, but the proviso in section\\n 4(c) of such Act (25 U.S.C. 450b(c)) shall not\\n apply with respect to grants awarded to, and\\n contracts entered into with, such Community.\\n\\nPub. L. No. 100-153, § 10, 101 Stat. 889.\\n\\n In 1988, Congress enacted IGRA to provide “a statutory\\nbasis for the operation and regulation” of Indian gaming.\\nSeminole Tribe of Fla. v. Florida, 517 U.S. 44, 48 (1996).\\nIGRA created the National Indian Gaming Commission\\n(Gaming Commission), the federal agency that regulates three\\nclasses of gaming on Indian lands. See 25 U.S.C.\\n§§ 2703(6)–(8), 2704, 2710. To conduct class II gaming, an\\n“Indian tribe” must adopt a “tribal ordinance” concerning the\\nregulation of class II gaming, and submit the ordinance to the\\nChair of the Gaming Commission for approval. 25 U.S.C.\\n§ 2710(b). IGRA defines “Indian tribe” as follows:\\n\\f6 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n [A]ny Indian tribe, band, nation, or other\\n organized group or community of Indians\\n which—\\n\\n (A) is recognized as eligible by the\\n Secretary for the special programs and\\n services provided by the United States to\\n Indians because of their status as Indians,\\n and\\n\\n (B) is recognized as possessing powers of\\n self-government.\\n\\nId. § 2703(5).\\n\\n In 1994, Congress amended the Frank’s Landing Act so\\nthat it now provides:\\n\\n (a) Subject to subsection (b), the Frank’s\\n Landing Indian Community in the State of\\n Washington is hereby recognized—\\n\\n (1) as eligible for the special programs\\n and services provided by the United States\\n to Indians because of their status as\\n Indians and is recognized as eligible to\\n contract, and to receive grants, under the\\n Indian Self-Determination and Education\\n Assistance Act for such services, but the\\n proviso in section 4(c) of such Act\\n (25 U.S.C. 450b(c)) shall not apply with\\n respect to grants awarded to, and contracts\\n entered into with, such Community; and\\n\\f FRANKS LANDING INDIAN COMTY. V. 7\\n NAT’L INDIAN GAMING COMM’N\\n\\n (2) as a self-governing dependent Indian\\n community that is not subject to the\\n jurisdiction of any federally recognized\\n tribe.\\n\\n (b)(1) Nothing in this section may be\\n construed to alter or affect the jurisdiction of\\n the State of Washington under section 1162 of\\n title 18, United States Code.\\n\\n (2) Nothing in this section may be\\n construed to constitute the recognition by\\n the United States that the Frank’s Landing\\n Indian Community is a federally\\n recognized Indian tribe.\\n\\n (3) Notwithstanding any other provision\\n of law, the Frank’s Landing Indian\\n Community shall not engage in any class\\n III gaming activity (as defined in section\\n 3(8) of the Indian Gaming Regulatory Act\\n of 1988 (25 U.S.C. 2703(8))).\\n\\nPub. L. No. 103-435, 108 Stat. 4566, 4569–70 (emphasis\\nadded).\\n\\n When Congress enacted the Frank’s Landing Act in 1987,\\nit had not yet enacted IGRA and thus it could not have\\nconsidered the Community’s tribal status or gaming rights for\\npurposes of IGRA. By the time the Frank’s Landing Act was\\namended in 1994, IGRA was well established, as was the\\nprocess for identifying which groups qualify as federally\\nrecognized tribes.\\n\\f8 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n B. Federal Recognition\\n\\n To provide context for our decision that the Community\\nis ineligible to participate in Indian gaming, we explain what\\nit means to be a federally recognized tribe.3 “Federal\\nrecognition” of an Indian tribe is a legal term of art meaning\\nthat the federal government acknowledges as a matter of law\\nthat a particular Indian group has tribal status. See H.R. Rep.\\nNo. 103-781, at 2 (1994), as reprinted in 1994 U.S.C.C.A.N.\\n3768; Felix Cohen, Cohen’s Handbook of Federal Indian\\nLaw § 3.02[3], at 134 (Nell Jessup Newton ed., 2012)\\n(hereinafter Cohen’s Handbook). Federal recognition\\nestablishes a government-to-government relationship between\\nthe United States and the recognized tribe as a “domestic\\ndependent nation,” and requires the Secretary to provide “a\\npanoply of benefits and services” to the tribe and its\\nmembers. See Cohen’s Handbook, § 3.02[3], at 134 (internal\\nquotation marks omitted).\\n\\n A group of Indians may achieve federal recognition in\\nthree ways: (1) by Congressional act; (2) by Secretarial\\nacknowledgment4 ; or (3) by a decision of a United States\\ncourt. See Federally Recognized Indian Tribe List Act of\\n1994, Pub. L. No. 103-454; United States v. Zepeda, 792 F.3d\\n1103, 1114 (9th Cir. 2015). As required by the Federally\\n\\n\\n 3\\n The term “federal recognition” is relevant to understanding the\\nsignificance of: (1) IGRA’s requirement that an Indian tribe must be\\n“recognized as eligible by the Secretary” for special programs and\\nservices; and (2) the Frank’s Landing Act’s provision that the Community\\nis not a federally recognized Indian tribe.\\n 4\\n “The terms ‘recognize’ and ‘acknowledge’ are often used\\ninterchangeably.” Cohen’s Handbook, § 3.02[3], at 134 n.21.\\n\\f FRANKS LANDING INDIAN COMTY. V. 9\\n NAT’L INDIAN GAMING COMM’N\\n\\nRecognized Indian Tribe List Act of 1994 (List Act),\\n25 U.S.C. § 5131, the Secretary of the Interior publishes an\\nannual list of all federally recognized tribes that have\\nobtained recognition through any one of these three means.\\nSee Zepeda, 792 F.3d 1103 at 1114. The Community has\\nnever petitioned the Secretary for inclusion on the annual list,\\nand it concedes that it has not been federally recognized\\nunder any of these three pathways.\\n\\n C. The Community’s Efforts to Engage in Class II\\n Gaming\\n\\n The Community submitted a class II gaming ordinance to\\nthe Gaming Commission for review and approval in\\nDecember of 2014. In March of 2015, the Assistant\\nSecretary–Indian Affairs, acting on behalf of Interior, issued\\na memorandum to the Chair of the Gaming Commission\\nconcluding that the Community did not qualify as an Indian\\ntribe for purposes of IGRA because it is not federally\\nrecognized and does not appear on the annual list of\\nrecognized tribes the Secretary publishes pursuant to the List\\nAct. The Assistant Secretary opined that the Gaming\\nCommission could rely on this annual list to determine\\nwhether an entity is a federally recognized Indian tribe\\nbecause the list is intended to be exhaustive, because it offers\\ntransparency surrounding which groups of Indians are eligible\\nfor gaming pursuant to IGRA, and because it provides a\\n“bright line” rule that preserves government resources. The\\nSolicitor of the Department of the Interior also concluded that\\nthe Community was not an Indian tribe for purposes of\\nIGRA.\\n\\f10 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n Relying on Interior’s determination, the Chair of the\\nGaming Commission concluded that the Commission was\\nwithout authority to review the Community’s gaming\\nordinance. The Community’s requests for reconsideration\\nwere denied, and the Community filed suit in district court.\\nThe district court dismissed the claims against the Gaming\\nCommission and its Chair for failure to state a claim.\\nThereafter, the court granted summary judgment in favor of\\nthe remaining defendants. The district court reasoned that\\nIGRA and the Frank’s Landing Act, read together, are\\nambiguous regarding whether Congress intended to authorize\\nthe Community to engage in class II gaming, but it\\nconcluded, pursuant to Chevron, U.S.A., Inc. v. Natural\\nResources Defense Council, Inc., 467 U.S. 837, 843–44\\n(1984), that defendants reasonably interpreted these statutes.\\nThis appeal of the summary judgment order followed.\\n\\n STANDARD OF REVIEW\\n\\n “We review de novo the district court’s summary\\njudgment ruling[ ], ‘thus reviewing directly the agency’s\\naction under the [APA’s] arbitrary and capricious standard.’”\\nCty. of Amador v. United States Dep’t of the Interior,\\n872 F.3d 1012, 1020 (9th Cir. 2017) (quoting Alaska\\nWilderness League v. Jewell, 788 F.3d 1212, 1217 (9th Cir.\\n2015)), cert. denied, 139 S. Ct. 64 (2018).\\n\\n DISCUSSION\\n\\n Interior determined that the Community is not entitled to\\nconduct Indian gaming for purposes of IGRA because it is not\\nfederally recognized and does not appear on the annual list of\\nrecognized tribes the Secretary publishes pursuant to the List\\n\\f FRANKS LANDING INDIAN COMTY. V. 11\\n NAT’L INDIAN GAMING COMM’N\\n\\nAct. Accordingly, the dispositive questions in this case are\\nnarrow: (1) whether an Indian group must be recognized by\\nthe Secretary in order to qualify as an “Indian tribe” for\\npurposes of IGRA; and, if so, (2) whether the Frank’s\\nLanding Act nevertheless authorizes the Community to\\nengage in class II gaming pursuant to IGRA. To resolve\\nthese questions, we start with the definition of “Indian tribe”\\nin IGRA, § 2703(5). Because the Frank’s Landing Act\\ngranted the Community some form of special status before\\nCongress adopted IGRA, we also consider the Frank’s\\nLanding Act, both as originally adopted and as amended in\\n1994.\\n\\n “We interpret [each] federal statute by ascertaining the\\nintent of Congress and by giving effect to its legislative will.”\\nArtichoke Joe’s California Grand Casino v. Norton, 353 F.3d\\n712, 720 (9th Cir. 2003) (internal quotation marks omitted).\\nTo resolve this appeal, “we need not decide whether Chevron\\ndeference (or any other level of deference) is appropriate,\\nbecause we reach the same conclusion as Interior when we\\nreview the [interpretive questions] de novo.” Cty. of Amador,\\n872 F.3d at 1021–22.\\n\\n We hold that Congress intended federal recognition by the\\nSecretary to be a prerequisite for inclusion in IGRA’s\\ndefinition of Indian tribe, and that when it amended the\\nFrank’s Landing Act in 1994, Congress did not separately\\nauthorize the non-federally recognized Community to engage\\nin class II gaming.\\n\\f12 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n A. Congress Made Federal Recognition a Prerequisite\\n for Indian Gaming.\\n\\n IGRA states that an “Indian tribe . . . recognized as\\neligible by the Secretary for the special programs and\\nservices” may engage in class II gaming. 25 U.S.C.\\n§ 2703(5)(A); see id. § 2710(b)(1). Our court has never\\nsquarely addressed the meaning of IGRA’s Secretarial\\nrecognition requirement, though we have suggested that only\\nfederally recognized tribes may engage in Indian gaming.\\nSee Artichoke Joe’s, 353 F.3d at 734 (stating that “only\\nfederally recognized tribes are covered” by IGRA, not\\nindividual Indians); see also Timbisha Shoshone Tribe v. U.S.\\nDep’t of Interior, 824 F.3d 807, 809 (9th Cir. 2016)\\n(observing that “only federally recognized tribes may operate\\ngambling facilities under the federal [IGRA].”)\\n\\n IGRA’s definition of “Indian tribe” includes\\n“communit[ies] of Indians” that are “recognized as\\npossessing powers of self-government.” 25 U.S.C.\\n§ 2703(5)(B). Interior does not challenge the Community’s\\nassertion that it is a self-governing community of Indians, but\\nthe parties part ways over how to interpret IGRA’s\\nrequirement that an Indian tribe must be “recognized as\\neligible by the Secretary for the special programs and services\\nprovided by the United States to Indians because of their\\nstatus as Indians[.]” Id. § 2703(5)(A).\\n\\n Interior points to IGRA’s requirement that a tribe be\\nrecognized as eligible by the Secretary, and argues that no\\nsuch recognition has been made for the Community. The\\nCommunity isolates IGRA’s phrase “special programs and\\nservices provided by the United States to Indians because of\\n\\f FRANKS LANDING INDIAN COMTY. V. 13\\n NAT’L INDIAN GAMING COMM’N\\n\\ntheir status as Indians,” and contends that it unambiguously\\nincludes the Community, because the Frank’s Landing Act\\nprovides that it is eligible to participate in “special programs\\nand services” specifically targeted to Indians. The\\nCommunity argues that the phrase “recognized as eligible by\\nthe Secretary” is irrelevant to the task of determining whether\\nthe Community may engage in class II gaming because\\nCongress has plenary power to legislate concerning Indian\\ntribes, and the Secretary must accede to what the Community\\nargues is a congressional directive in the Frank’s Landing\\nAct—that the Community is eligible for “special programs\\nand services.” The Community urges us to conclude that it is\\ntherefore eligible to participate in class II gaming.\\n\\n We agree with the Community that the phrase “special\\nprograms and services provided by the United States to\\nIndians because of their status as Indians” refers to eligibility\\nto participate in federal programs specifically targeted to\\nIndians. See Hoopa Valley Indian Tribe v. Ryan, 415 F.3d\\n986, 990–91 (9th Cir. 2005) (describing programs and\\nservices provided to Indians “because of their status as\\nIndians” as “specifically targeted to Indians” (internal\\nquotation marks omitted)); Navajo Nation v. Dep’t of Health\\n& Human Servs., 325 F.3d 1133, 1138 (9th Cir. 2003) (en\\nbanc) (same). However, IGRA’s additional phrase\\n“recognized as eligible by the Secretary” is a key qualifier.\\nOn its face, this phrase means that the Secretary must\\nrecognize an Indian tribe as eligible for special programs and\\nservices. And, as we explain infra, the significance of\\nSecretarial recognition in the context of federal Indian law\\nmeans that a tribe is federally recognized and that it appears\\non the Secretary’s annual list. See generally F.A.A. v.\\nCooper, 566 U.S. 284, 291–92 (2012) (observing the\\n\\f14 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n“cardinal rule of statutory construction” that when “Congress\\nemploys a term of art, it presumably knows and adopts the\\ncluster of ideas that were attached to each borrowed word in\\nthe body of learning from which it was taken.” (internal\\nquotation marks omitted)); Dolan v. U.S.P.S., 546 U.S. 481,\\n486 (2006) (“Interpretation of a word or phrase depends upon\\nreading the whole statutory text, considering the purpose and\\ncontext of the statute, and consulting any precedents or\\nauthorities that inform the analysis.”).\\n\\n The administrative process by which the Secretary\\nfederally recognizes an Indian tribe was set out in Interior’s\\nPart 83 regulations in 1978, long before the adoption of\\nIGRA. See 43 Fed. Reg. 39,361 (Sept. 5, 1978). The Part 83\\nregulations established a “procedure and policy for\\nacknowledging that certain American Indian tribes exist.\\nSuch acknowledgment of tribal existence by the Department\\n[of the Interior] is a prerequisite to the protection, services,\\nand benefits from the Federal Government available to Indian\\ntribes.” 43 Fed. Reg. at 39,362. The Part 83 regulations\\nrequired the Secretary to publish “a list of all Indian tribes\\nwhich are recognized and receiving services from the Bureau\\nof Indian Affairs.” Id. Inclusion on the Secretary’s list\\nsignifies that a tribe has been federally recognized and is\\neligible to receive services from the Bureau. Id.\\n\\n We assume Congress is knowledgeable about existing law\\npertinent to the legislation it enacts, see South Dakota v.\\nYankton Sioux Tribe, 522 U.S. 329, 351 (1998), and the\\nPart 83 regulations were longstanding by the time Congress\\nenacted IGRA in 1988. Thus, we have no doubt about what\\nCongress intended when it used the phrase “recognized as\\neligible by the Secretary” in IGRA; this phrase is\\n\\f FRANKS LANDING INDIAN COMTY. V. 15\\n NAT’L INDIAN GAMING COMM’N\\n\\nsynonymous with federal recognition. See Pit River Home &\\nAgric. Coop. Ass’n v. United States, 30 F.3d 1088, 1094–96\\n(9th Cir. 1994) (concluding a tribe was not “recognized by\\nthe Secretary” because the tribe was not federally recognized\\nin any statute or treaty, and was not listed pursuant to Part 83\\nregulations). Use of “recognized as eligible by the Secretary”\\nin IGRA’s definition of “Indian tribe” reflects Congressional\\nintent to require formal Secretarial recognition as a\\nprerequisite for gaming.5\\n\\n Our conclusion that IGRA only applies to tribes\\nrecognized on the Secretary’s annual list is consistent with\\n\\n\\n 5\\n The Community argues that Native Village of Noatak v. Hoffman\\nsupports its position. 896 F.2d 1157, 1160 (9th Cir. 1990), rev’d on other\\ngrounds by Blatchford v. Native Village of Noatak and Circle Village,\\n501 U.S. 775 (1991). There, in a different context, we said that “[i]f\\nCongress has recognized the tribe, a fortiori the tribe is entitled to\\nrecognition and is in fact recognized by the Secretary of the Interior.” Id.\\nBut the issue in Noatak was whether two Native Villages were entitled to\\nbring suit in federal court pursuant to 28 U.S.C. § 1362; i.e., whether the\\nvillages were “Indian tribe[s] or band[s] with [] governing bod[ies] duly\\nrecognized by the Secretary of the Interior” Id. One of the villages had\\na governing body approved by the Secretary pursuant to 25 U.S.C. § 476,\\nid., and when the Native Villages filed their complaint in 1985, they were\\nalready included on the Secretary’s annual List. See 48 Fed. Reg.\\n56,865–66 (Dec. 23, 1983). In contrast, the question here is whether the\\nCommunity qualified as a tribe for purposes of IGRA. IGRA\\nunambiguously requires recognition by the Secretary, and the Community\\nconcedes that it is not a federally recognized tribe. Further, Congress\\ndeclared the Community eligible for “special programs and services”\\nbefore IGRA was enacted, so it could not have intended the original\\nFrank’s Landing Act to bestow IGRA’s benefits upon the Community.\\nCongress did not take the opportunity to expand the Community’s ability\\nto participate in IGRA gaming when it amended the Frank’s Landing Act\\nin 1994.\\n\\f16 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\nthe observations of other federal courts,6 and we know of no\\ncourt that has ruled otherwise. It is unsurprising that the\\nparticular interpretive question we address here has not been\\nexamined in our earlier caselaw; IGRA’s plain language\\nprovides that an Indian tribe is one that is “recognized as\\neligible by the Secretary” for “special programs and\\nservices,” and this type of recognition is typically determined\\nby simply confirming whether an Indian group appears on the\\nSecretary’s annual list. In this case, the Community’s\\nstatus—as set forth in the Frank’s Landing Act—is unique.\\nWe do not know of any other group with a comparable status,\\nand neither does the Community.\\n\\n Having reviewed the applicable statutory text and\\nconsidering the sequence in which the related legislation was\\nenacted, we conclude that Congress clearly and\\nunambiguously intended federal recognition by the Secretary\\nto be a prerequisite for inclusion in IGRA’s definition of\\n\\n\\n\\n 6\\n See Wisconsin v. Ho-Chunk Nation, 784 F.3d 1076, 1079 (7th Cir.\\n2015) (“IGRA divides all Indian gaming (that is, gambling run by\\nfederally recognized tribes) into three classes.”); Narragansett Indian\\nTribe v. Nat’l Indian Gaming Comm’n, 158 F.3d 1335, 1337 (D.C. Cir.\\n1998) (noting that IGRA “permits federally recognized tribes to apply for\\nCommission approval of gaming proposals.”); Passamaquoddy Tribe v.\\nMaine, 75 F.3d 784, 792 n.4 (1st Cir. 1996) (observing that “[IGRA] has\\nno application to tribes that do not seek and attain formal federal\\nrecognition.”); Carruthers v. Flaum, 365 F. Supp. 2d 448, 451, 466–67\\n(S.D.N.Y. 2005) (concluding that because the Unkechaug are not federally\\nrecognized, IGRA provides no exception to New York’s general\\nprohibition on gambling); First Am. Casino Corp. v. E. Pequot Nation,\\n175 F. Supp. 2d 205, 208 (D. Conn. 2000) (“Unless and until defendant\\nobtains federal acknowledgment, its activities are not regulated by\\nIGRA.”).\\n\\f FRANKS LANDING INDIAN COMTY. V. 17\\n NAT’L INDIAN GAMING COMM’N\\n\\n“Indian tribe.” IGRA does not authorize the non-federally\\nrecognized Community to engage in class II gaming.\\n\\n B. The Frank’s Landing Act Does Not Authorize the\\n Community to Engage in Indian Gaming.\\n\\n Despite IGRA’s requirement that an Indian tribe must be\\nrecognized by the Secretary, the Community argues that the\\nFrank’s Landing Act separately granted it permission to\\nengage in class II gaming. Specifically, the Community\\nargues that because Congress defined “Indian tribe” in IGRA\\nwith the “precise language” used in the Frank’s Landing Act,\\nCongress intended to authorize the Community to engage in\\nclass II gaming.\\n\\n Both the 1987 Frank’s Landing Act and the 1994\\namended Act provide that the Community is:\\n\\n recognized as eligible for the special\\n programs and services provided by the United\\n States to Indians because of their status as\\n Indians and is recognized as eligible to\\n contract, and to receive grants, under the\\n Indian Self-Determination and Education\\n Assistance Act for such services. . . .\\n\\nPub. L. No. 100-153, 101 Stat. 886, 889; § 10, Pub. L. No.\\n103-435, 108 Stat. 4566, 4569. Plainly, this language\\nindicates that Congress intended the Community to be\\neligible for special programs and services by virtue of their\\nstatus as Indians, and Congress made clear that it intended the\\nCommunity would contract for, and receive grants pursuant\\nto, the Indian Self-Determination and Education Assistance\\n\\f18 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\nAct. See Indian Self-Determination and Education Assistance\\nAct, 25 U.S.C. § 5321(a)(1)(E) (providing that upon request,\\nthe Secretary shall enter into self-determination contracts that\\ninclude programs “for the benefit of Indians because of their\\nstatus as Indians[.]”). But neither the 1987 Frank’s Landing\\nAct nor the 1994 amendments contain IGRA’s qualifying\\nphrase signaling federal recognition, “recognized as eligible\\nby the Secretary[,]” 25 U.S.C. § 2703(5) (emphasis added).\\nFurther, Congress underscored that the 1994 amendments\\nwere not intended to federally recognize the Community by\\nspecifying that “[n]othing in this section may be construed to\\nconstitute the recognition by the United States that the\\n[Community] is a federally recognized Indian tribe.” Pub. L.\\nNo. 103-435, § 8, 108 Stat. 4569.\\n\\n The Community argues that we should infer that Congress\\nintended to allow it to engage in class I or II gaming because\\nthe amended Act only expressly prohibits the Community\\nfrom engaging in class III gaming. To support this argument,\\nthe Community invokes the rule that a statute should be\\nconstrued so that no clause, sentence, or word is rendered\\nsuperfluous, and argues that there would have been no need\\nto expressly prohibit it from engaging in class III gaming if\\nlack of federal recognition rendered it ineligible to participate\\nin all IGRA gaming. Interior responds that Congress would\\nnot choose such an indirect route to convey an easily\\nexpressed message, and that there is no indication that this\\nclass III provision means anything other than what it\\nsays—that notwithstanding any other provision of law, i.e.,\\nnotwithstanding federal recognition at some future time, the\\nCommunity is prohibited from engaging in class III gaming.\\n\\f FRANKS LANDING INDIAN COMTY. V. 19\\n NAT’L INDIAN GAMING COMM’N\\n\\n Like the district court, we read the amended Act’s\\nprohibition of class III gaming as an express limitation on the\\nCommunity’s class III gaming rights if it is ever federally\\nrecognized. The words Congress used make this clear.\\nSubsections (b)(1) and (b)(2) provide that “[n]othing in this\\nsection may be construed to” alter or affect the criminal\\njurisdiction of Washington under Public Law 2807 or to\\nconstitute federal recognition of the Community. Congress\\nemployed language with a much broader sweep when it\\ndrafted subsection (b)(3). That subsection states\\n“[n]otwithstanding any other provision of law” the\\nCommunity shall not engage in class III gaming. The\\nSupreme Court recently observed that “[t]he ordinary\\nmeaning of ‘notwithstanding’ is ‘in spite of,’ or ‘without\\nprevention or obstruction from or by.’” N.L.R.B. v. SW Gen.,\\nInc., — U.S. —, 137 S. Ct. 929, 939 (2017) (quoting\\nWebster’s Third New International Dictionary 1545 (1986);\\nBlack’s Law Dictionary 1091 (7th ed. 1999)). The Court\\nexplained that the word “notwithstanding” demonstrates\\n“which provision prevails in the event of a clash,” id.\\n(quoting A. Scalia & B. Garner, Reading Law: The\\nInterpretation of Legal Texts 126–27 (2012)), and that a\\n“notwithstanding” clause “confirms rather than constrains\\nbreadth.” Id. at 940.\\n\\n Accordingly, we read “notwithstanding” in\\nsubsection (b)(3) as a signal that the amended Act’s broad\\nprohibition on class III gaming shall prevail in the event of a\\nconflict with “any other provision of law.” In contrast, the\\npreface to subsections (b)(1) and (b)(2)—“[n]othing in this\\nsection”—constrains the reach of the amended Act with\\n\\n 7\\n 18 U.S.C. § 1162.\\n\\f20 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\nrespect to Public Law 280 and federal recognition. This\\nvariation in language reflects a deliberate choice, see S.E.C.\\nv. McCarthy, 322 F.3d 650, 656 (9th Cir. 2003) (“the use of\\ndifferent words or terms within a statute demonstrates that\\nCongress intended to convey a different meaning for those\\nwords.”).\\n\\n The historical context of the Frank’s Landing Act also\\nconfirms that the Community is not authorized to engage in\\ngaming. On the same day Congress amended the Frank’s\\nLanding Act, November 2, 1994, see 108 Stat. 4566, 4569, it\\ncodified Interior’s longstanding Part 83 practice of publishing\\na list of recognized tribes, see 108 Stat. 4791. As explained,\\nthe List Act directed the Secretary to annually publish “a list\\nof all Indian tribes which the Secretary recognizes to be\\neligible for the special programs and services provided by the\\nUnited States to Indians because of their status as Indians.”\\n25 U.S.C. § 5131. Assuming, as we must, that Congress was\\nknowledgeable about Interior’s longstanding Part 83\\nregulations, IGRA’s requirement that tribes be recognized by\\nthe Secretary, and the List Act’s codification of the\\nSecretary’s annual list, it is conspicuous that Congress did not\\ntake the opportunity when it amended the Frank’s Landing\\nAct to specifically authorize the Community to engage in\\nclass II gaming. Indeed, Congress did just the opposite by\\nunequivocally stating that the Frank’s Landing Act\\namendments were not intended to confer federal recognition.\\nSee Pub. Law No. 103-435, 108 Stat. 4566, 4569.\\n\\n In light of the express statement in the Frank’s Landing\\nAct that the Community is not a federally recognized tribe,\\nthe amended Act’s broad limitation on class III gaming\\nrights, and the context in which the Act was amended, we\\n\\f FRANKS LANDING INDIAN COMTY. V. 21\\n NAT’L INDIAN GAMING COMM’N\\n\\nconclude it is not susceptible to more than one reasonable\\ninterpretation. The Frank’s Landing Act does not separately\\nauthorize the Community to engage in Indian gaming.\\n\\n CONCLUSION\\n\\n Congress clearly and unambiguously intended federal\\nrecognition by the Secretary to be a prerequisite for\\nparticipation in IGRA gaming. Congress did not separately\\nauthorize the non-federally recognized Community to engage\\nin class II gaming by amending the Frank’s Landing Act.\\nTherefore, Interior correctly concluded that the Community\\nwas not eligible to engage in IGRA gaming and we affirm the\\ndistrict court’s order granting summary judgment in favor of\\nthe appellees.\\n\\n AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4376069/\", \"author_raw\": \"CHRISTEN, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"N RANDY SMITH"},"judge2":{"kind":"string","value":"MORGAN CHRISTEN"},"judge3":{"kind":"string","value":"ROBERT E PAYNE"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4598816/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Franks Landing Indian Cmty. v. National Indian Gaming Comm'r.\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-12\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nFRANK’S LANDING INDIAN No. 17-35368\nCOMMUNITY, a federally-\nrecognized self-governing D.C. No.\ndependent Indian community, 3:15-cv-05828-BHS\n Plaintiff-Appellant,\n\n v. OPINION\n\nNATIONAL INDIAN GAMING\nCOMMISSION; UNITED STATES\nDEPARTMENT OF THE\nINTERIOR; JONODEV\nCHAUDHURI, in his official\ncapacity as Chairman of the\nNational Indian Gaming\nCommission; TARA KATUK\nMACLEAN SWEENEY, in her\nofficial capacity as Assistant\nSecretary of the Interior\nIndian Affairs, United States\nDepartment of the Interior;\nDAVID BERNHARDT, in his\nofficial capacity as Acting\nSecretary of the Interior,\n Defendants-Appellees.\n\f2 FRANKS LANDING INDIAN COMTY. V.\n NAT’L INDIAN GAMING COMM’N\n\n Appeal from the United States District Court\n for the Western District of Washington\n Benjamin H. Settle, District Judge, Presiding\n\n Argued and Submitted October 11, 2018\n Seattle, Washington\n\n Filed March 12, 2019\n\n Before: N. Randy Smith and Morgan Christen, Circuit\n Judges, and Robert E. Payne,* District Judge.\n\n Opinion by Judge Christen\n\n\n SUMMARY**\n\n\n Tribal Matters\n\n The panel affirmed the district court’s summary judgment\nentered in favor of the U.S. Department of the Interior, its\nSecretary, and Assistant Secretary of Indian Affairs, in an\naction challenging the Department of the Interior’s\ndetermination that Frank’s Landing Indian Community is\nineligible for gaming for purposes of the Indian Gaming\nRegulatory Act (“IGRA”).\n\n\n *\n The Honorable Robert E. Payne, United States District Judge for the\nEastern District of Virginia, sitting by designation.\n **\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\f FRANKS LANDING INDIAN COMTY. V. 3\n NAT’L INDIAN GAMING COMM’N\n\n The panel held that IGRA clearly and unambiguously\nrequired federal recognition by the Secretary of the Interior\nbefore a tribe may qualify to participate in Indian gaming.\nThe panel also held that when Congress amended the Frank’s\nLanding Act in 1994, it did not separately authorize the non-\nfederally recognized Frank’s Landing Indian Community to\nengage in Class II gaming. The panel held that the\nDepartment of the Interior correctly concluded that the\nCommunity was not eligible to engage in IGRA gaming.\n\n\n COUNSEL\n\nScott David Crowell (argued), Crowell Law Office-Tribal\nAdvocacy Group, Sedona, Arizona; Bryan T. Newland,\nFletcher Law PLLC, East Lansing, Michigan; Alan S.\nMiddleton, Law Offices of Alan S. Middleton PLLC, Lake\nForest Park, Washington; for Plaintiff-Appellant.\n\nKevin W. McArdle (argued), Mary Gabrielle Sprague, and\nDevon Lehman McCune, Environment and Natural Resources\nDivision; Eric Grant, Deputy Assistant Attorney General;\nJeffrey H. Wood, Acting Assistant Attorney General; United\nStates Department of Justice, Washington, D.C.; Samuel E.\nEnnis, Office of the Solicitor, Department of the Interior,\nWashington, D.C.; for Defendants-Appellees.\n\f4 FRANKS LANDING INDIAN COMTY. V.\n NAT’L INDIAN GAMING COMM’N\n\n OPINION\n\nCHRISTEN, Circuit Judge:\n\n Frank’s Landing Indian Community (Community)\nappeals the district court’s order granting summary judgment\nin favor of appellees Department of the Interior, its Secretary,\nand the Assistant Secretary–Indian Affairs. The Community,\nwhich is not a federally recognized tribe, challenges Interior’s\ndetermination that it is ineligible for gaming for purposes of\nthe Indian Gaming Regulatory Act, (IGRA), 25 U.S.C.\n§§ 2701–2721. The Community argues that its unique status,\nrecognized and defined in the Frank’s Landing Act1 and the\n1994 amendments to that Act,2 renders it eligible to engage in\nclass II gaming.\n\n We have jurisdiction pursuant to 28 U.S.C. § 1291 and we\naffirm the district court’s judgment. Reviewing Interior’s\ndecision under the Administrative Procedure Act (APA), we\nconclude the agency’s determination was correct. First, we\nhold that IGRA clearly and unambiguously requires federal\nrecognition by the Secretary of the Department of the Interior\nbefore a tribe may qualify to participate in Indian gaming.\nSecond, we hold that the Frank’s Landing Act does not\nauthorize the Community to engage in class II gaming.\n\n\n\n\n 1\n Pub. L. No. 100-153, § 10, 101 Stat. 886, 889.\n 2\n Pub. L. No. 103-435, § 8, 108 Stat. 4566, 4569–70.\n\f FRANKS LANDING INDIAN COMTY. V. 5\n NAT’L INDIAN GAMING COMM’N\n\n BACKGROUND\n\n A. Statutory Background\n\n Adopted in 1987, the Frank’s Landing Act originally\nprovided as follows:\n\n The Frank’s Landing Indian Community in\n the State of Washington is hereby recognized\n as eligible for the special programs and\n services provided by the United States to\n Indians because of their status as Indians and\n is recognized as eligible to contract, and to\n receive grants, under the Indian Self-\n Determination and Education Assistance Act\n for such services, but the proviso in section\n 4(c) of such Act (25 U.S.C. 450b(c)) shall not\n apply with respect to grants awarded to, and\n contracts entered into with, such Community.\n\nPub. L. No. 100-153, § 10, 101 Stat. 889.\n\n In 1988, Congress enacted IGRA to provide “a statutory\nbasis for the operation and regulation” of Indian gaming.\nSeminole Tribe of Fla. v. Florida, 517 U.S. 44, 48 (1996).\nIGRA created the National Indian Gaming Commission\n(Gaming Commission), the federal agency that regulates three\nclasses of gaming on Indian lands. See 25 U.S.C.\n§§ 2703(6)–(8), 2704, 2710. To conduct class II gaming, an\n“Indian tribe” must adopt a “tribal ordinance” concerning the\nregulation of class II gaming, and submit the ordinance to the\nChair of the Gaming Commission for approval. 25 U.S.C.\n§ 2710(b). IGRA defines “Indian tribe” as follows:\n\f6 FRANKS LANDING INDIAN COMTY. V.\n NAT’L INDIAN GAMING COMM’N\n\n [A]ny Indian tribe, band, nation, or other\n organized group or community of Indians\n which—\n\n (A) is recognized as eligible by the\n Secretary for the special programs and\n services provided by the United States to\n Indians because of their status as Indians,\n and\n\n (B) is recognized as possessing powers of\n self-government.\n\nId. § 2703(5).\n\n In 1994, Congress amended the Frank’s Landing Act so\nthat it now provides:\n\n (a) Subject to subsection (b), the Frank’s\n Landing Indian Community in the State of\n Washington is hereby recognized—\n\n (1) as eligible for the special programs\n and services provided by the United States\n to Indians because of their status as\n Indians and is recognized as eligible to\n contract, and to receive grants, under the\n Indian Self-Determination and Education\n Assistance Act for such services, but the\n proviso in section 4(c) of such Act\n (25 U.S.C. 450b(c)) shall not apply with\n respect to grants awarded to, and contracts\n entered into with, such Community; and\n\f FRANKS LANDING INDIAN COMTY. V. 7\n NAT’L INDIAN GAMING COMM’N\n\n (2) as a self-governing dependent Indian\n community that is not subject to the\n jurisdiction of any federally recognized\n tribe.\n\n (b)(1) Nothing in this section may be\n construed to alter or affect the jurisdiction of\n the State of Washington under section 1162 of\n title 18, United States Code.\n\n (2) Nothing in this section may be\n construed to constitute the recognition by\n the United States that the Frank’s Landing\n Indian Community is a federally\n recognized Indian tribe.\n\n (3) Notwithstanding any other provision\n of law, the Frank’s Landing Indian\n Community shall not engage in any class\n III gaming activity (as defined in section\n 3(8) of the Indian Gaming Regulatory Act\n of 1988 (25 U.S.C. 2703(8))).\n\nPub. L. No. 103-435, 108 Stat. 4566, 4569–70 (emphasis\nadded).\n\n When Congress enacted the Frank’s Landing Act in 1987,\nit had not yet enacted IGRA and thus it could not have\nconsidered the Community’s tribal status or gaming rights for\npurposes of IGRA. By the time the Frank’s Landing Act was\namended in 1994, IGRA was well established, as was the\nprocess for identifying which groups qualify as federally\nrecognized tribes.\n\f8 FRANKS LANDING INDIAN COMTY. V.\n NAT’L INDIAN GAMING COMM’N\n\n B. Federal Recognition\n\n To provide context for our decision that the Community\nis ineligible to participate in Indian gaming, we explain what\nit means to be a federally recognized tribe.3 “Federal\nrecognition” of an Indian tribe is a legal term of art meaning\nthat the federal government acknowledges as a matter of law\nthat a particular Indian group has tribal status. See H.R. Rep.\nNo. 103-781, at 2 (1994), as reprinted in 1994 U.S.C.C.A.N.\n3768; Felix Cohen, Cohen’s Handbook of Federal Indian\nLaw § 3.02[3], at 134 (Nell Jessup Newton ed., 2012)\n(hereinafter Cohen’s Handbook). Federal recognition\nestablishes a government-to-government relationship between\nthe United States and the recognized tribe as a “domestic\ndependent nation,” and requires the Secretary to provide “a\npanoply of benefits and services” to the tribe and its\nmembers. See Cohen’s Handbook, § 3.02[3], at 134 (internal\nquotation marks omitted).\n\n A group of Indians may achieve federal recognition in\nthree ways: (1) by Congressional act; (2) by Secretarial\nacknowledgment4 ; or (3) by a decision of a United States\ncourt. See Federally Recognized Indian Tribe List Act of\n1994, Pub. L. No. 103-454; United States v. Zepeda, 792 F.3d\n1103, 1114 (9th Cir. 2015). As required by the Federally\n\n\n 3\n The term “federal recognition” is relevant to understanding the\nsignificance of: (1) IGRA’s requirement that an Indian tribe must be\n“recognized as eligible by the Secretary” for special programs and\nservices; and (2) the Frank’s Landing Act’s provision that the Community\nis not a federally recognized Indian tribe.\n 4\n “The terms ‘recognize’ and ‘acknowledge’ are often used\ninterchangeably.” Cohen’s Handbook, § 3.02[3], at 134 n.21.\n\f FRANKS LANDING INDIAN COMTY. V. 9\n NAT’L INDIAN GAMING COMM’N\n\nRecognized Indian Tribe List Act of 1994 (List Act),\n25 U.S.C. § 5131, the Secretary of the Interior publishes an\nannual list of all federally recognized tribes that have\nobtained recognition through any one of these three means.\nSee Zepeda, 792 F.3d 1103 at 1114. The Community has\nnever petitioned the Secretary for inclusion on the annual list,\nand it concedes that it has not been federally recognized\nunder any of these three pathways.\n\n C. The Community’s Efforts to Engage in Class II\n Gaming\n\n The Community submitted a class II gaming ordinance to\nthe Gaming Commission for review and approval in\nDecember of 2014. In March of 2015, the Assistant\nSecretary–Indian Affairs, acting on behalf of Interior, issued\na memorandum to the Chair of the Gaming Commission\nconcluding that the Community did not qualify as an Indian\ntribe for purposes of IGRA because it is not federally\nrecognized and does not appear on the annual list of\nrecognized tribes the Secretary publishes pursuant to the List\nAct. The Assistant Secretary opined that the Gaming\nCommission could rely on this annual list to determine\nwhether an entity is a federally recognized Indian tribe\nbecause the list is intended to be exhaustive, because it offers\ntransparency surrounding which groups of Indians are eligible\nfor gaming pursuant to IGRA, and because it provides a\n“bright line” rule that preserves government resources. The\nSolicitor of the Department of the Interior also concluded that\nthe Community was not an Indian tribe for purposes of\nIGRA.\n\f10 FRANKS LANDING INDIAN COMTY. V.\n NAT’L INDIAN GAMING COMM’N\n\n Relying on Interior’s determination, the Chair of the\nGaming Commission concluded that the Commission was\nwithout authority to review the Community’s gaming\nordinance. The Community’s requests for reconsideration\nwere denied, and the Community filed suit in district court.\nThe district court dismissed the claims against the Gaming\nCommission and its Chair for failure to state a claim.\nThereafter, the court granted summary judgment in favor of\nthe remaining defendants. The district court reasoned that\nIGRA and the Frank’s Landing Act, read together, are\nambiguous regarding whether Congress intended to authorize\nthe Community to engage in class II gaming, but it\nconcluded, pursuant to Chevron, U.S.A., Inc. v. Natural\nResources Defense Council, Inc., 467 U.S. 837, 843–44\n(1984), that defendants reasonably interpreted these statutes.\nThis appeal of the summary judgment order followed.\n\n STANDARD OF REVIEW\n\n “We review de novo the district court’s summary\njudgment ruling[ ], ‘thus reviewing directly the agency’s\naction under the [APA’s] arbitrary and capricious standard.’”\nCty. of Amador v. United States Dep’t of the Interior,\n872 F.3d 1012, 1020 (9th Cir. 2017) (quoting Alaska\nWilderness League v. Jewell, 788 F.3d 1212, 1217 (9th Cir.\n2015)), cert. denied, 139 S. Ct. 64 (2018).\n\n DISCUSSION\n\n Interior determined that the Community is not entitled to\nconduct Indian gaming for purposes of IGRA because it is not\nfederally recognized and does not appear on the annual list of\nrecognized tribes the Secretary publishes pursuant to the List\n\f FRANKS LANDING INDIAN COMTY. V. 11\n NAT’L INDIAN GAMING COMM’N\n\nAct. Accordingly, the dispositive questions in this case are\nnarrow: (1) whether an Indian group must be recognized by\nthe Secretary in order to qualify as an “Indian tribe” for\npurposes of IGRA; and, if so, (2) whether the Frank’s\nLanding Act nevertheless authorizes the Community to\nengage in class II gaming pursuant to IGRA. To resolve\nthese questions, we start with the definition of “Indian tribe”\nin IGRA, § 2703(5). Because the Frank’s Landing Act\ngranted the Community some form of special status before\nCongress adopted IGRA, we also consider the Frank’s\nLanding Act, both as originally adopted and as amended in\n1994.\n\n “We interpret [each] federal statute by ascertaining the\nintent of Congress and by giving effect to its legislative will.”\nArtichoke Joe’s California Grand Casino v. Norton, 353 F.3d\n712, 720 (9th Cir. 2003) (internal quotation marks omitted).\nTo resolve this appeal, “we need not decide whether Chevron\ndeference (or any other level of deference) is appropriate,\nbecause we reach the same conclusion as Interior when we\nreview the [interpretive questions] de novo.” Cty. of Amador,\n872 F.3d at 1021–22.\n\n We hold that Congress intended federal recognition by the\nSecretary to be a prerequisite for inclusion in IGRA’s\ndefinition of Indian tribe, and that when it amended the\nFrank’s Landing Act in 1994, Congress did not separately\nauthorize the non-federally recognized Community to engage\nin class II gaming.\n\f12 FRANKS LANDING INDIAN COMTY. V.\n NAT’L INDIAN GAMING COMM’N\n\n A. Congress Made Federal Recognition a Prerequisite\n for Indian Gaming.\n\n IGRA states that an “Indian tribe . . . recognized as\neligible by the Secretary for the special programs and\nservices” may engage in class II gaming. 25 U.S.C.\n§ 2703(5)(A); see id. § 2710(b)(1). Our court has never\nsquarely addressed the meaning of IGRA’s Secretarial\nrecognition requirement, though we have suggested that only\nfederally recognized tribes may engage in Indian gaming.\nSee Artichoke Joe’s, 353 F.3d at 734 (stating that “only\nfederally recognized tribes are covered” by IGRA, not\nindividual Indians); see also Timbisha Shoshone Tribe v. U.S.\nDep’t of Interior, 824 F.3d 807, 809 (9th Cir. 2016)\n(observing that “only federally recognized tribes may operate\ngambling facilities under the federal [IGRA].”)\n\n IGRA’s definition of “Indian tribe” includes\n“communit[ies] of Indians” that are “recognized as\npossessing powers of self-government.” 25 U.S.C.\n§ 2703(5)(B). Interior does not challenge the Community’s\nassertion that it is a self-governing community of Indians, but\nthe parties part ways over how to interpret IGRA’s\nrequirement that an Indian tribe must be “recognized as\neligible by the Secretary for the special programs and services\nprovided by the United States to Indians because of their\nstatus as Indians[.]” Id. § 2703(5)(A).\n\n Interior points to IGRA’s requirement that a tribe be\nrecognized as eligible by the Secretary, and argues that no\nsuch recognition has been made for the Community. The\nCommunity isolates IGRA’s phrase “special programs and\nservices provided by the United States to Indians because of\n\f FRANKS LANDING INDIAN COMTY. V. 13\n NAT’L INDIAN GAMING COMM’N\n\ntheir status as Indians,” and contends that it unambiguously\nincludes the Community, because the Frank’s Landing Act\nprovides that it is eligible to participate in “special programs\nand services” specifically targeted to Indians. The\nCommunity argues that the phrase “recognized as eligible by\nthe Secretary” is irrelevant to the task of determining whether\nthe Community may engage in class II gaming because\nCongress has plenary power to legislate concerning Indian\ntribes, and the Secretary must accede to what the Community\nargues is a congressional directive in the Frank’s Landing\nAct—that the Community is eligible for “special programs\nand services.” The Community urges us to conclude that it is\ntherefore eligible to participate in class II gaming.\n\n We agree with the Community that the phrase “special\nprograms and services provided by the United States to\nIndians because of their status as Indians” refers to eligibility\nto participate in federal programs specifically targeted to\nIndians. See Hoopa Valley Indian Tribe v. Ryan, 415 F.3d\n986, 990–91 (9th Cir. 2005) (describing programs and\nservices provided to Indians “because of their status as\nIndians” as “specifically targeted to Indians” (internal\nquotation marks omitted)); Navajo Nation v. Dep’t of Health\n& Human Servs., 325 F.3d 1133, 1138 (9th Cir. 2003) (en\nbanc) (same). However, IGRA’s additional phrase\n“recognized as eligible by the Secretary” is a key qualifier.\nOn its face, this phrase means that the Secretary must\nrecognize an Indian tribe as eligible for special programs and\nservices. And, as we explain infra, the significance of\nSecretarial recognition in the context of federal Indian law\nmeans that a tribe is federally recognized and that it appears\non the Secretary’s annual list. See generally F.A.A. v.\nCooper, 566 U.S. 284, 291–92 (2012) (observing the\n\f14 FRANKS LANDING INDIAN COMTY. V.\n NAT’L INDIAN GAMING COMM’N\n\n“cardinal rule of statutory construction” that when “Congress\nemploys a term of art, it presumably knows and adopts the\ncluster of ideas that were attached to each borrowed word in\nthe body of learning from which it was taken.” (internal\nquotation marks omitted)); Dolan v. U.S.P.S., 546 U.S. 481,\n486 (2006) (“Interpretation of a word or phrase depends upon\nreading the whole statutory text, considering the purpose and\ncontext of the statute, and consulting any precedents or\nauthorities that inform the analysis.”).\n\n The administrative process by which the Secretary\nfederally recognizes an Indian tribe was set out in Interior’s\nPart 83 regulations in 1978, long before the adoption of\nIGRA. See 43 Fed. Reg. 39,361 (Sept. 5, 1978). The Part 83\nregulations established a “procedure and policy for\nacknowledging that certain American Indian tribes exist.\nSuch acknowledgment of tribal existence by the Department\n[of the Interior] is a prerequisite to the protection, services,\nand benefits from the Federal Government available to Indian\ntribes.” 43 Fed. Reg. at 39,362. The Part 83 regulations\nrequired the Secretary to publish “a list of all Indian tribes\nwhich are recognized and receiving services from the Bureau\nof Indian Affairs.” Id. Inclusion on the Secretary’s list\nsignifies that a tribe has been federally recognized and is\neligible to receive services from the Bureau. Id.\n\n We assume Congress is knowledgeable about existing law\npertinent to the legislation it enacts, see South Dakota v.\nYankton Sioux Tribe, 522 U.S. 329, 351 (1998), and the\nPart 83 regulations were longstanding by the time Congress\nenacted IGRA in 1988. Thus, we have no doubt about what\nCongress intended when it used the phrase “recognized as\neligible by the Secretary” in IGRA; this phrase is\n\f FRANKS LANDING INDIAN COMTY. V. 15\n NAT’L INDIAN GAMING COMM’N\n\nsynonymous with federal recognition. See Pit River Home &\nAgric. Coop. Ass’n v. United States, 30 F.3d 1088, 1094–96\n(9th Cir. 1994) (concluding a tribe was not “recognized by\nthe Secretary” because the tribe was not federally recognized\nin any statute or treaty, and was not listed pursuant to Part 83\nregulations). Use of “recognized as eligible by the Secretary”\nin IGRA’s definition of “Indian tribe” reflects Congressional\nintent to require formal Secretarial recognition as a\nprerequisite for gaming.5\n\n Our conclusion that IGRA only applies to tribes\nrecognized on the Secretary’s annual list is consistent with\n\n\n 5\n The Community argues that Native Village of Noatak v. Hoffman\nsupports its position. 896 F.2d 1157, 1160 (9th Cir. 1990), rev’d on other\ngrounds by Blatchford v. Native Village of Noatak and Circle Village,\n501 U.S. 775 (1991). There, in a different context, we said that “[i]f\nCongress has recognized the tribe, a fortiori the tribe is entitled to\nrecognition and is in fact recognized by the Secretary of the Interior.” Id.\nBut the issue in Noatak was whether two Native Villages were entitled to\nbring suit in federal court pursuant to 28 U.S.C. § 1362; i.e., whether the\nvillages were “Indian tribe[s] or band[s] with [] governing bod[ies] duly\nrecognized by the Secretary of the Interior” Id. One of the villages had\na governing body approved by the Secretary pursuant to 25 U.S.C. § 476,\nid., and when the Native Villages filed their complaint in 1985, they were\nalready included on the Secretary’s annual List. See 48 Fed. Reg.\n56,865–66 (Dec. 23, 1983). In contrast, the question here is whether the\nCommunity qualified as a tribe for purposes of IGRA. IGRA\nunambiguously requires recognition by the Secretary, and the Community\nconcedes that it is not a federally recognized tribe. Further, Congress\ndeclared the Community eligible for “special programs and services”\nbefore IGRA was enacted, so it could not have intended the original\nFrank’s Landing Act to bestow IGRA’s benefits upon the Community.\nCongress did not take the opportunity to expand the Community’s ability\nto participate in IGRA gaming when it amended the Frank’s Landing Act\nin 1994.\n\f16 FRANKS LANDING INDIAN COMTY. V.\n NAT’L INDIAN GAMING COMM’N\n\nthe observations of other federal courts,6 and we know of no\ncourt that has ruled otherwise. It is unsurprising that the\nparticular interpretive question we address here has not been\nexamined in our earlier caselaw; IGRA’s plain language\nprovides that an Indian tribe is one that is “recognized as\neligible by the Secretary” for “special programs and\nservices,” and this type of recognition is typically determined\nby simply confirming whether an Indian group appears on the\nSecretary’s annual list. In this case, the Community’s\nstatus—as set forth in the Frank’s Landing Act—is unique.\nWe do not know of any other group with a comparable status,\nand neither does the Community.\n\n Having reviewed the applicable statutory text and\nconsidering the sequence in which the related legislation was\nenacted, we conclude that Congress clearly and\nunambiguously intended federal recognition by the Secretary\nto be a prerequisite for inclusion in IGRA’s definition of\n\n\n\n 6\n See Wisconsin v. Ho-Chunk Nation, 784 F.3d 1076, 1079 (7th Cir.\n2015) (“IGRA divides all Indian gaming (that is, gambling run by\nfederally recognized tribes) into three classes.”); Narragansett Indian\nTribe v. Nat’l Indian Gaming Comm’n, 158 F.3d 1335, 1337 (D.C. Cir.\n1998) (noting that IGRA “permits federally recognized tribes to apply for\nCommission approval of gaming proposals.”); Passamaquoddy Tribe v.\nMaine, 75 F.3d 784, 792 n.4 (1st Cir. 1996) (observing that “[IGRA] has\nno application to tribes that do not seek and attain formal federal\nrecognition.”); Carruthers v. Flaum, 365 F. Supp. 2d 448, 451, 466–67\n(S.D.N.Y. 2005) (concluding that because the Unkechaug are not federally\nrecognized, IGRA provides no exception to New York’s general\nprohibition on gambling); First Am. Casino Corp. v. E. Pequot Nation,\n175 F. Supp. 2d 205, 208 (D. Conn. 2000) (“Unless and until defendant\nobtains federal acknowledgment, its activities are not regulated by\nIGRA.”).\n\f FRANKS LANDING INDIAN COMTY. V. 17\n NAT’L INDIAN GAMING COMM’N\n\n“Indian tribe.” IGRA does not authorize the non-federally\nrecognized Community to engage in class II gaming.\n\n B. The Frank’s Landing Act Does Not Authorize the\n Community to Engage in Indian Gaming.\n\n Despite IGRA’s requirement that an Indian tribe must be\nrecognized by the Secretary, the Community argues that the\nFrank’s Landing Act separately granted it permission to\nengage in class II gaming. Specifically, the Community\nargues that because Congress defined “Indian tribe” in IGRA\nwith the “precise language” used in the Frank’s Landing Act,\nCongress intended to authorize the Community to engage in\nclass II gaming.\n\n Both the 1987 Frank’s Landing Act and the 1994\namended Act provide that the Community is:\n\n recognized as eligible for the special\n programs and services provided by the United\n States to Indians because of their status as\n Indians and is recognized as eligible to\n contract, and to receive grants, under the\n Indian Self-Determination and Education\n Assistance Act for such services. . . .\n\nPub. L. No. 100-153, 101 Stat. 886, 889; § 10, Pub. L. No.\n103-435, 108 Stat. 4566, 4569. Plainly, this language\nindicates that Congress intended the Community to be\neligible for special programs and services by virtue of their\nstatus as Indians, and Congress made clear that it intended the\nCommunity would contract for, and receive grants pursuant\nto, the Indian Self-Determination and Education Assistance\n\f18 FRANKS LANDING INDIAN COMTY. V.\n NAT’L INDIAN GAMING COMM’N\n\nAct. See Indian Self-Determination and Education Assistance\nAct, 25 U.S.C. § 5321(a)(1)(E) (providing that upon request,\nthe Secretary shall enter into self-determination contracts that\ninclude programs “for the benefit of Indians because of their\nstatus as Indians[.]”). But neither the 1987 Frank’s Landing\nAct nor the 1994 amendments contain IGRA’s qualifying\nphrase signaling federal recognition, “recognized as eligible\nby the Secretary[,]” 25 U.S.C. § 2703(5) (emphasis added).\nFurther, Congress underscored that the 1994 amendments\nwere not intended to federally recognize the Community by\nspecifying that “[n]othing in this section may be construed to\nconstitute the recognition by the United States that the\n[Community] is a federally recognized Indian tribe.” Pub. L.\nNo. 103-435, § 8, 108 Stat. 4569.\n\n The Community argues that we should infer that Congress\nintended to allow it to engage in class I or II gaming because\nthe amended Act only expressly prohibits the Community\nfrom engaging in class III gaming. To support this argument,\nthe Community invokes the rule that a statute should be\nconstrued so that no clause, sentence, or word is rendered\nsuperfluous, and argues that there would have been no need\nto expressly prohibit it from engaging in class III gaming if\nlack of federal recognition rendered it ineligible to participate\nin all IGRA gaming. Interior responds that Congress would\nnot choose such an indirect route to convey an easily\nexpressed message, and that there is no indication that this\nclass III provision means anything other than what it\nsays—that notwithstanding any other provision of law, i.e.,\nnotwithstanding federal recognition at some future time, the\nCommunity is prohibited from engaging in class III gaming.\n\f FRANKS LANDING INDIAN COMTY. V. 19\n NAT’L INDIAN GAMING COMM’N\n\n Like the district court, we read the amended Act’s\nprohibition of class III gaming as an express limitation on the\nCommunity’s class III gaming rights if it is ever federally\nrecognized. The words Congress used make this clear.\nSubsections (b)(1) and (b)(2) provide that “[n]othing in this\nsection may be construed to” alter or affect the criminal\njurisdiction of Washington under Public Law 2807 or to\nconstitute federal recognition of the Community. Congress\nemployed language with a much broader sweep when it\ndrafted subsection (b)(3). That subsection states\n“[n]otwithstanding any other provision of law” the\nCommunity shall not engage in class III gaming. The\nSupreme Court recently observed that “[t]he ordinary\nmeaning of ‘notwithstanding’ is ‘in spite of,’ or ‘without\nprevention or obstruction from or by.’” N.L.R.B. v. SW Gen.,\nInc., — U.S. —, 137 S. Ct. 929, 939 (2017) (quoting\nWebster’s Third New International Dictionary 1545 (1986);\nBlack’s Law Dictionary 1091 (7th ed. 1999)). The Court\nexplained that the word “notwithstanding” demonstrates\n“which provision prevails in the event of a clash,” id.\n(quoting A. Scalia & B. Garner, Reading Law: The\nInterpretation of Legal Texts 126–27 (2012)), and that a\n“notwithstanding” clause “confirms rather than constrains\nbreadth.” Id. at 940.\n\n Accordingly, we read “notwithstanding” in\nsubsection (b)(3) as a signal that the amended Act’s broad\nprohibition on class III gaming shall prevail in the event of a\nconflict with “any other provision of law.” In contrast, the\npreface to subsections (b)(1) and (b)(2)—“[n]othing in this\nsection”—constrains the reach of the amended Act with\n\n 7\n 18 U.S.C. § 1162.\n\f20 FRANKS LANDING INDIAN COMTY. V.\n NAT’L INDIAN GAMING COMM’N\n\nrespect to Public Law 280 and federal recognition. This\nvariation in language reflects a deliberate choice, see S.E.C.\nv. McCarthy, 322 F.3d 650, 656 (9th Cir. 2003) (“the use of\ndifferent words or terms within a statute demonstrates that\nCongress intended to convey a different meaning for those\nwords.”).\n\n The historical context of the Frank’s Landing Act also\nconfirms that the Community is not authorized to engage in\ngaming. On the same day Congress amended the Frank’s\nLanding Act, November 2, 1994, see 108 Stat. 4566, 4569, it\ncodified Interior’s longstanding Part 83 practice of publishing\na list of recognized tribes, see 108 Stat. 4791. As explained,\nthe List Act directed the Secretary to annually publish “a list\nof all Indian tribes which the Secretary recognizes to be\neligible for the special programs and services provided by the\nUnited States to Indians because of their status as Indians.”\n25 U.S.C. § 5131. Assuming, as we must, that Congress was\nknowledgeable about Interior’s longstanding Part 83\nregulations, IGRA’s requirement that tribes be recognized by\nthe Secretary, and the List Act’s codification of the\nSecretary’s annual list, it is conspicuous that Congress did not\ntake the opportunity when it amended the Frank’s Landing\nAct to specifically authorize the Community to engage in\nclass II gaming. Indeed, Congress did just the opposite by\nunequivocally stating that the Frank’s Landing Act\namendments were not intended to confer federal recognition.\nSee Pub. Law No. 103-435, 108 Stat. 4566, 4569.\n\n In light of the express statement in the Frank’s Landing\nAct that the Community is not a federally recognized tribe,\nthe amended Act’s broad limitation on class III gaming\nrights, and the context in which the Act was amended, we\n\f FRANKS LANDING INDIAN COMTY. V. 21\n NAT’L INDIAN GAMING COMM’N\n\nconclude it is not susceptible to more than one reasonable\ninterpretation. The Frank’s Landing Act does not separately\nauthorize the Community to engage in Indian gaming.\n\n CONCLUSION\n\n Congress clearly and unambiguously intended federal\nrecognition by the Secretary to be a prerequisite for\nparticipation in IGRA gaming. Congress did not separately\nauthorize the non-federally recognized Community to engage\nin class II gaming by amending the Frank’s Landing Act.\nTherefore, Interior correctly concluded that the Community\nwas not eligible to engage in IGRA gaming and we affirm the\ndistrict court’s order granting summary judgment in favor of\nthe appellees.\n\n AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Franks Landing Indian Cmty. v. National Indian Gaming Comm'r.\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-12\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nFRANK’S LANDING INDIAN No. 17-35368\\nCOMMUNITY, a federally-\\nrecognized self-governing D.C. No.\\ndependent Indian community, 3:15-cv-05828-BHS\\n Plaintiff-Appellant,\\n\\n v. OPINION\\n\\nNATIONAL INDIAN GAMING\\nCOMMISSION; UNITED STATES\\nDEPARTMENT OF THE\\nINTERIOR; JONODEV\\nCHAUDHURI, in his official\\ncapacity as Chairman of the\\nNational Indian Gaming\\nCommission; TARA KATUK\\nMACLEAN SWEENEY, in her\\nofficial capacity as Assistant\\nSecretary of the Interior\\nIndian Affairs, United States\\nDepartment of the Interior;\\nDAVID BERNHARDT, in his\\nofficial capacity as Acting\\nSecretary of the Interior,\\n Defendants-Appellees.\\n\\f2 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n Appeal from the United States District Court\\n for the Western District of Washington\\n Benjamin H. Settle, District Judge, Presiding\\n\\n Argued and Submitted October 11, 2018\\n Seattle, Washington\\n\\n Filed March 12, 2019\\n\\n Before: N. Randy Smith and Morgan Christen, Circuit\\n Judges, and Robert E. Payne,* District Judge.\\n\\n Opinion by Judge Christen\\n\\n\\n SUMMARY**\\n\\n\\n Tribal Matters\\n\\n The panel affirmed the district court’s summary judgment\\nentered in favor of the U.S. Department of the Interior, its\\nSecretary, and Assistant Secretary of Indian Affairs, in an\\naction challenging the Department of the Interior’s\\ndetermination that Frank’s Landing Indian Community is\\nineligible for gaming for purposes of the Indian Gaming\\nRegulatory Act (“IGRA”).\\n\\n\\n *\\n The Honorable Robert E. Payne, United States District Judge for the\\nEastern District of Virginia, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f FRANKS LANDING INDIAN COMTY. V. 3\\n NAT’L INDIAN GAMING COMM’N\\n\\n The panel held that IGRA clearly and unambiguously\\nrequired federal recognition by the Secretary of the Interior\\nbefore a tribe may qualify to participate in Indian gaming.\\nThe panel also held that when Congress amended the Frank’s\\nLanding Act in 1994, it did not separately authorize the non-\\nfederally recognized Frank’s Landing Indian Community to\\nengage in Class II gaming. The panel held that the\\nDepartment of the Interior correctly concluded that the\\nCommunity was not eligible to engage in IGRA gaming.\\n\\n\\n COUNSEL\\n\\nScott David Crowell (argued), Crowell Law Office-Tribal\\nAdvocacy Group, Sedona, Arizona; Bryan T. Newland,\\nFletcher Law PLLC, East Lansing, Michigan; Alan S.\\nMiddleton, Law Offices of Alan S. Middleton PLLC, Lake\\nForest Park, Washington; for Plaintiff-Appellant.\\n\\nKevin W. McArdle (argued), Mary Gabrielle Sprague, and\\nDevon Lehman McCune, Environment and Natural Resources\\nDivision; Eric Grant, Deputy Assistant Attorney General;\\nJeffrey H. Wood, Acting Assistant Attorney General; United\\nStates Department of Justice, Washington, D.C.; Samuel E.\\nEnnis, Office of the Solicitor, Department of the Interior,\\nWashington, D.C.; for Defendants-Appellees.\\n\\f4 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n OPINION\\n\\nCHRISTEN, Circuit Judge:\\n\\n Frank’s Landing Indian Community (Community)\\nappeals the district court’s order granting summary judgment\\nin favor of appellees Department of the Interior, its Secretary,\\nand the Assistant Secretary–Indian Affairs. The Community,\\nwhich is not a federally recognized tribe, challenges Interior’s\\ndetermination that it is ineligible for gaming for purposes of\\nthe Indian Gaming Regulatory Act, (IGRA), 25 U.S.C.\\n§§ 2701–2721. The Community argues that its unique status,\\nrecognized and defined in the Frank’s Landing Act1 and the\\n1994 amendments to that Act,2 renders it eligible to engage in\\nclass II gaming.\\n\\n We have jurisdiction pursuant to 28 U.S.C. § 1291 and we\\naffirm the district court’s judgment. Reviewing Interior’s\\ndecision under the Administrative Procedure Act (APA), we\\nconclude the agency’s determination was correct. First, we\\nhold that IGRA clearly and unambiguously requires federal\\nrecognition by the Secretary of the Department of the Interior\\nbefore a tribe may qualify to participate in Indian gaming.\\nSecond, we hold that the Frank’s Landing Act does not\\nauthorize the Community to engage in class II gaming.\\n\\n\\n\\n\\n 1\\n Pub. L. No. 100-153, § 10, 101 Stat. 886, 889.\\n 2\\n Pub. L. No. 103-435, § 8, 108 Stat. 4566, 4569–70.\\n\\f FRANKS LANDING INDIAN COMTY. V. 5\\n NAT’L INDIAN GAMING COMM’N\\n\\n BACKGROUND\\n\\n A. Statutory Background\\n\\n Adopted in 1987, the Frank’s Landing Act originally\\nprovided as follows:\\n\\n The Frank’s Landing Indian Community in\\n the State of Washington is hereby recognized\\n as eligible for the special programs and\\n services provided by the United States to\\n Indians because of their status as Indians and\\n is recognized as eligible to contract, and to\\n receive grants, under the Indian Self-\\n Determination and Education Assistance Act\\n for such services, but the proviso in section\\n 4(c) of such Act (25 U.S.C. 450b(c)) shall not\\n apply with respect to grants awarded to, and\\n contracts entered into with, such Community.\\n\\nPub. L. No. 100-153, § 10, 101 Stat. 889.\\n\\n In 1988, Congress enacted IGRA to provide “a statutory\\nbasis for the operation and regulation” of Indian gaming.\\nSeminole Tribe of Fla. v. Florida, 517 U.S. 44, 48 (1996).\\nIGRA created the National Indian Gaming Commission\\n(Gaming Commission), the federal agency that regulates three\\nclasses of gaming on Indian lands. See 25 U.S.C.\\n§§ 2703(6)–(8), 2704, 2710. To conduct class II gaming, an\\n“Indian tribe” must adopt a “tribal ordinance” concerning the\\nregulation of class II gaming, and submit the ordinance to the\\nChair of the Gaming Commission for approval. 25 U.S.C.\\n§ 2710(b). IGRA defines “Indian tribe” as follows:\\n\\f6 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n [A]ny Indian tribe, band, nation, or other\\n organized group or community of Indians\\n which—\\n\\n (A) is recognized as eligible by the\\n Secretary for the special programs and\\n services provided by the United States to\\n Indians because of their status as Indians,\\n and\\n\\n (B) is recognized as possessing powers of\\n self-government.\\n\\nId. § 2703(5).\\n\\n In 1994, Congress amended the Frank’s Landing Act so\\nthat it now provides:\\n\\n (a) Subject to subsection (b), the Frank’s\\n Landing Indian Community in the State of\\n Washington is hereby recognized—\\n\\n (1) as eligible for the special programs\\n and services provided by the United States\\n to Indians because of their status as\\n Indians and is recognized as eligible to\\n contract, and to receive grants, under the\\n Indian Self-Determination and Education\\n Assistance Act for such services, but the\\n proviso in section 4(c) of such Act\\n (25 U.S.C. 450b(c)) shall not apply with\\n respect to grants awarded to, and contracts\\n entered into with, such Community; and\\n\\f FRANKS LANDING INDIAN COMTY. V. 7\\n NAT’L INDIAN GAMING COMM’N\\n\\n (2) as a self-governing dependent Indian\\n community that is not subject to the\\n jurisdiction of any federally recognized\\n tribe.\\n\\n (b)(1) Nothing in this section may be\\n construed to alter or affect the jurisdiction of\\n the State of Washington under section 1162 of\\n title 18, United States Code.\\n\\n (2) Nothing in this section may be\\n construed to constitute the recognition by\\n the United States that the Frank’s Landing\\n Indian Community is a federally\\n recognized Indian tribe.\\n\\n (3) Notwithstanding any other provision\\n of law, the Frank’s Landing Indian\\n Community shall not engage in any class\\n III gaming activity (as defined in section\\n 3(8) of the Indian Gaming Regulatory Act\\n of 1988 (25 U.S.C. 2703(8))).\\n\\nPub. L. No. 103-435, 108 Stat. 4566, 4569–70 (emphasis\\nadded).\\n\\n When Congress enacted the Frank’s Landing Act in 1987,\\nit had not yet enacted IGRA and thus it could not have\\nconsidered the Community’s tribal status or gaming rights for\\npurposes of IGRA. By the time the Frank’s Landing Act was\\namended in 1994, IGRA was well established, as was the\\nprocess for identifying which groups qualify as federally\\nrecognized tribes.\\n\\f8 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n B. Federal Recognition\\n\\n To provide context for our decision that the Community\\nis ineligible to participate in Indian gaming, we explain what\\nit means to be a federally recognized tribe.3 “Federal\\nrecognition” of an Indian tribe is a legal term of art meaning\\nthat the federal government acknowledges as a matter of law\\nthat a particular Indian group has tribal status. See H.R. Rep.\\nNo. 103-781, at 2 (1994), as reprinted in 1994 U.S.C.C.A.N.\\n3768; Felix Cohen, Cohen’s Handbook of Federal Indian\\nLaw § 3.02[3], at 134 (Nell Jessup Newton ed., 2012)\\n(hereinafter Cohen’s Handbook). Federal recognition\\nestablishes a government-to-government relationship between\\nthe United States and the recognized tribe as a “domestic\\ndependent nation,” and requires the Secretary to provide “a\\npanoply of benefits and services” to the tribe and its\\nmembers. See Cohen’s Handbook, § 3.02[3], at 134 (internal\\nquotation marks omitted).\\n\\n A group of Indians may achieve federal recognition in\\nthree ways: (1) by Congressional act; (2) by Secretarial\\nacknowledgment4 ; or (3) by a decision of a United States\\ncourt. See Federally Recognized Indian Tribe List Act of\\n1994, Pub. L. No. 103-454; United States v. Zepeda, 792 F.3d\\n1103, 1114 (9th Cir. 2015). As required by the Federally\\n\\n\\n 3\\n The term “federal recognition” is relevant to understanding the\\nsignificance of: (1) IGRA’s requirement that an Indian tribe must be\\n“recognized as eligible by the Secretary” for special programs and\\nservices; and (2) the Frank’s Landing Act’s provision that the Community\\nis not a federally recognized Indian tribe.\\n 4\\n “The terms ‘recognize’ and ‘acknowledge’ are often used\\ninterchangeably.” Cohen’s Handbook, § 3.02[3], at 134 n.21.\\n\\f FRANKS LANDING INDIAN COMTY. V. 9\\n NAT’L INDIAN GAMING COMM’N\\n\\nRecognized Indian Tribe List Act of 1994 (List Act),\\n25 U.S.C. § 5131, the Secretary of the Interior publishes an\\nannual list of all federally recognized tribes that have\\nobtained recognition through any one of these three means.\\nSee Zepeda, 792 F.3d 1103 at 1114. The Community has\\nnever petitioned the Secretary for inclusion on the annual list,\\nand it concedes that it has not been federally recognized\\nunder any of these three pathways.\\n\\n C. The Community’s Efforts to Engage in Class II\\n Gaming\\n\\n The Community submitted a class II gaming ordinance to\\nthe Gaming Commission for review and approval in\\nDecember of 2014. In March of 2015, the Assistant\\nSecretary–Indian Affairs, acting on behalf of Interior, issued\\na memorandum to the Chair of the Gaming Commission\\nconcluding that the Community did not qualify as an Indian\\ntribe for purposes of IGRA because it is not federally\\nrecognized and does not appear on the annual list of\\nrecognized tribes the Secretary publishes pursuant to the List\\nAct. The Assistant Secretary opined that the Gaming\\nCommission could rely on this annual list to determine\\nwhether an entity is a federally recognized Indian tribe\\nbecause the list is intended to be exhaustive, because it offers\\ntransparency surrounding which groups of Indians are eligible\\nfor gaming pursuant to IGRA, and because it provides a\\n“bright line” rule that preserves government resources. The\\nSolicitor of the Department of the Interior also concluded that\\nthe Community was not an Indian tribe for purposes of\\nIGRA.\\n\\f10 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n Relying on Interior’s determination, the Chair of the\\nGaming Commission concluded that the Commission was\\nwithout authority to review the Community’s gaming\\nordinance. The Community’s requests for reconsideration\\nwere denied, and the Community filed suit in district court.\\nThe district court dismissed the claims against the Gaming\\nCommission and its Chair for failure to state a claim.\\nThereafter, the court granted summary judgment in favor of\\nthe remaining defendants. The district court reasoned that\\nIGRA and the Frank’s Landing Act, read together, are\\nambiguous regarding whether Congress intended to authorize\\nthe Community to engage in class II gaming, but it\\nconcluded, pursuant to Chevron, U.S.A., Inc. v. Natural\\nResources Defense Council, Inc., 467 U.S. 837, 843–44\\n(1984), that defendants reasonably interpreted these statutes.\\nThis appeal of the summary judgment order followed.\\n\\n STANDARD OF REVIEW\\n\\n “We review de novo the district court’s summary\\njudgment ruling[ ], ‘thus reviewing directly the agency’s\\naction under the [APA’s] arbitrary and capricious standard.’”\\nCty. of Amador v. United States Dep’t of the Interior,\\n872 F.3d 1012, 1020 (9th Cir. 2017) (quoting Alaska\\nWilderness League v. Jewell, 788 F.3d 1212, 1217 (9th Cir.\\n2015)), cert. denied, 139 S. Ct. 64 (2018).\\n\\n DISCUSSION\\n\\n Interior determined that the Community is not entitled to\\nconduct Indian gaming for purposes of IGRA because it is not\\nfederally recognized and does not appear on the annual list of\\nrecognized tribes the Secretary publishes pursuant to the List\\n\\f FRANKS LANDING INDIAN COMTY. V. 11\\n NAT’L INDIAN GAMING COMM’N\\n\\nAct. Accordingly, the dispositive questions in this case are\\nnarrow: (1) whether an Indian group must be recognized by\\nthe Secretary in order to qualify as an “Indian tribe” for\\npurposes of IGRA; and, if so, (2) whether the Frank’s\\nLanding Act nevertheless authorizes the Community to\\nengage in class II gaming pursuant to IGRA. To resolve\\nthese questions, we start with the definition of “Indian tribe”\\nin IGRA, § 2703(5). Because the Frank’s Landing Act\\ngranted the Community some form of special status before\\nCongress adopted IGRA, we also consider the Frank’s\\nLanding Act, both as originally adopted and as amended in\\n1994.\\n\\n “We interpret [each] federal statute by ascertaining the\\nintent of Congress and by giving effect to its legislative will.”\\nArtichoke Joe’s California Grand Casino v. Norton, 353 F.3d\\n712, 720 (9th Cir. 2003) (internal quotation marks omitted).\\nTo resolve this appeal, “we need not decide whether Chevron\\ndeference (or any other level of deference) is appropriate,\\nbecause we reach the same conclusion as Interior when we\\nreview the [interpretive questions] de novo.” Cty. of Amador,\\n872 F.3d at 1021–22.\\n\\n We hold that Congress intended federal recognition by the\\nSecretary to be a prerequisite for inclusion in IGRA’s\\ndefinition of Indian tribe, and that when it amended the\\nFrank’s Landing Act in 1994, Congress did not separately\\nauthorize the non-federally recognized Community to engage\\nin class II gaming.\\n\\f12 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n A. Congress Made Federal Recognition a Prerequisite\\n for Indian Gaming.\\n\\n IGRA states that an “Indian tribe . . . recognized as\\neligible by the Secretary for the special programs and\\nservices” may engage in class II gaming. 25 U.S.C.\\n§ 2703(5)(A); see id. § 2710(b)(1). Our court has never\\nsquarely addressed the meaning of IGRA’s Secretarial\\nrecognition requirement, though we have suggested that only\\nfederally recognized tribes may engage in Indian gaming.\\nSee Artichoke Joe’s, 353 F.3d at 734 (stating that “only\\nfederally recognized tribes are covered” by IGRA, not\\nindividual Indians); see also Timbisha Shoshone Tribe v. U.S.\\nDep’t of Interior, 824 F.3d 807, 809 (9th Cir. 2016)\\n(observing that “only federally recognized tribes may operate\\ngambling facilities under the federal [IGRA].”)\\n\\n IGRA’s definition of “Indian tribe” includes\\n“communit[ies] of Indians” that are “recognized as\\npossessing powers of self-government.” 25 U.S.C.\\n§ 2703(5)(B). Interior does not challenge the Community’s\\nassertion that it is a self-governing community of Indians, but\\nthe parties part ways over how to interpret IGRA’s\\nrequirement that an Indian tribe must be “recognized as\\neligible by the Secretary for the special programs and services\\nprovided by the United States to Indians because of their\\nstatus as Indians[.]” Id. § 2703(5)(A).\\n\\n Interior points to IGRA’s requirement that a tribe be\\nrecognized as eligible by the Secretary, and argues that no\\nsuch recognition has been made for the Community. The\\nCommunity isolates IGRA’s phrase “special programs and\\nservices provided by the United States to Indians because of\\n\\f FRANKS LANDING INDIAN COMTY. V. 13\\n NAT’L INDIAN GAMING COMM’N\\n\\ntheir status as Indians,” and contends that it unambiguously\\nincludes the Community, because the Frank’s Landing Act\\nprovides that it is eligible to participate in “special programs\\nand services” specifically targeted to Indians. The\\nCommunity argues that the phrase “recognized as eligible by\\nthe Secretary” is irrelevant to the task of determining whether\\nthe Community may engage in class II gaming because\\nCongress has plenary power to legislate concerning Indian\\ntribes, and the Secretary must accede to what the Community\\nargues is a congressional directive in the Frank’s Landing\\nAct—that the Community is eligible for “special programs\\nand services.” The Community urges us to conclude that it is\\ntherefore eligible to participate in class II gaming.\\n\\n We agree with the Community that the phrase “special\\nprograms and services provided by the United States to\\nIndians because of their status as Indians” refers to eligibility\\nto participate in federal programs specifically targeted to\\nIndians. See Hoopa Valley Indian Tribe v. Ryan, 415 F.3d\\n986, 990–91 (9th Cir. 2005) (describing programs and\\nservices provided to Indians “because of their status as\\nIndians” as “specifically targeted to Indians” (internal\\nquotation marks omitted)); Navajo Nation v. Dep’t of Health\\n& Human Servs., 325 F.3d 1133, 1138 (9th Cir. 2003) (en\\nbanc) (same). However, IGRA’s additional phrase\\n“recognized as eligible by the Secretary” is a key qualifier.\\nOn its face, this phrase means that the Secretary must\\nrecognize an Indian tribe as eligible for special programs and\\nservices. And, as we explain infra, the significance of\\nSecretarial recognition in the context of federal Indian law\\nmeans that a tribe is federally recognized and that it appears\\non the Secretary’s annual list. See generally F.A.A. v.\\nCooper, 566 U.S. 284, 291–92 (2012) (observing the\\n\\f14 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\n“cardinal rule of statutory construction” that when “Congress\\nemploys a term of art, it presumably knows and adopts the\\ncluster of ideas that were attached to each borrowed word in\\nthe body of learning from which it was taken.” (internal\\nquotation marks omitted)); Dolan v. U.S.P.S., 546 U.S. 481,\\n486 (2006) (“Interpretation of a word or phrase depends upon\\nreading the whole statutory text, considering the purpose and\\ncontext of the statute, and consulting any precedents or\\nauthorities that inform the analysis.”).\\n\\n The administrative process by which the Secretary\\nfederally recognizes an Indian tribe was set out in Interior’s\\nPart 83 regulations in 1978, long before the adoption of\\nIGRA. See 43 Fed. Reg. 39,361 (Sept. 5, 1978). The Part 83\\nregulations established a “procedure and policy for\\nacknowledging that certain American Indian tribes exist.\\nSuch acknowledgment of tribal existence by the Department\\n[of the Interior] is a prerequisite to the protection, services,\\nand benefits from the Federal Government available to Indian\\ntribes.” 43 Fed. Reg. at 39,362. The Part 83 regulations\\nrequired the Secretary to publish “a list of all Indian tribes\\nwhich are recognized and receiving services from the Bureau\\nof Indian Affairs.” Id. Inclusion on the Secretary’s list\\nsignifies that a tribe has been federally recognized and is\\neligible to receive services from the Bureau. Id.\\n\\n We assume Congress is knowledgeable about existing law\\npertinent to the legislation it enacts, see South Dakota v.\\nYankton Sioux Tribe, 522 U.S. 329, 351 (1998), and the\\nPart 83 regulations were longstanding by the time Congress\\nenacted IGRA in 1988. Thus, we have no doubt about what\\nCongress intended when it used the phrase “recognized as\\neligible by the Secretary” in IGRA; this phrase is\\n\\f FRANKS LANDING INDIAN COMTY. V. 15\\n NAT’L INDIAN GAMING COMM’N\\n\\nsynonymous with federal recognition. See Pit River Home &\\nAgric. Coop. Ass’n v. United States, 30 F.3d 1088, 1094–96\\n(9th Cir. 1994) (concluding a tribe was not “recognized by\\nthe Secretary” because the tribe was not federally recognized\\nin any statute or treaty, and was not listed pursuant to Part 83\\nregulations). Use of “recognized as eligible by the Secretary”\\nin IGRA’s definition of “Indian tribe” reflects Congressional\\nintent to require formal Secretarial recognition as a\\nprerequisite for gaming.5\\n\\n Our conclusion that IGRA only applies to tribes\\nrecognized on the Secretary’s annual list is consistent with\\n\\n\\n 5\\n The Community argues that Native Village of Noatak v. Hoffman\\nsupports its position. 896 F.2d 1157, 1160 (9th Cir. 1990), rev’d on other\\ngrounds by Blatchford v. Native Village of Noatak and Circle Village,\\n501 U.S. 775 (1991). There, in a different context, we said that “[i]f\\nCongress has recognized the tribe, a fortiori the tribe is entitled to\\nrecognition and is in fact recognized by the Secretary of the Interior.” Id.\\nBut the issue in Noatak was whether two Native Villages were entitled to\\nbring suit in federal court pursuant to 28 U.S.C. § 1362; i.e., whether the\\nvillages were “Indian tribe[s] or band[s] with [] governing bod[ies] duly\\nrecognized by the Secretary of the Interior” Id. One of the villages had\\na governing body approved by the Secretary pursuant to 25 U.S.C. § 476,\\nid., and when the Native Villages filed their complaint in 1985, they were\\nalready included on the Secretary’s annual List. See 48 Fed. Reg.\\n56,865–66 (Dec. 23, 1983). In contrast, the question here is whether the\\nCommunity qualified as a tribe for purposes of IGRA. IGRA\\nunambiguously requires recognition by the Secretary, and the Community\\nconcedes that it is not a federally recognized tribe. Further, Congress\\ndeclared the Community eligible for “special programs and services”\\nbefore IGRA was enacted, so it could not have intended the original\\nFrank’s Landing Act to bestow IGRA’s benefits upon the Community.\\nCongress did not take the opportunity to expand the Community’s ability\\nto participate in IGRA gaming when it amended the Frank’s Landing Act\\nin 1994.\\n\\f16 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\nthe observations of other federal courts,6 and we know of no\\ncourt that has ruled otherwise. It is unsurprising that the\\nparticular interpretive question we address here has not been\\nexamined in our earlier caselaw; IGRA’s plain language\\nprovides that an Indian tribe is one that is “recognized as\\neligible by the Secretary” for “special programs and\\nservices,” and this type of recognition is typically determined\\nby simply confirming whether an Indian group appears on the\\nSecretary’s annual list. In this case, the Community’s\\nstatus—as set forth in the Frank’s Landing Act—is unique.\\nWe do not know of any other group with a comparable status,\\nand neither does the Community.\\n\\n Having reviewed the applicable statutory text and\\nconsidering the sequence in which the related legislation was\\nenacted, we conclude that Congress clearly and\\nunambiguously intended federal recognition by the Secretary\\nto be a prerequisite for inclusion in IGRA’s definition of\\n\\n\\n\\n 6\\n See Wisconsin v. Ho-Chunk Nation, 784 F.3d 1076, 1079 (7th Cir.\\n2015) (“IGRA divides all Indian gaming (that is, gambling run by\\nfederally recognized tribes) into three classes.”); Narragansett Indian\\nTribe v. Nat’l Indian Gaming Comm’n, 158 F.3d 1335, 1337 (D.C. Cir.\\n1998) (noting that IGRA “permits federally recognized tribes to apply for\\nCommission approval of gaming proposals.”); Passamaquoddy Tribe v.\\nMaine, 75 F.3d 784, 792 n.4 (1st Cir. 1996) (observing that “[IGRA] has\\nno application to tribes that do not seek and attain formal federal\\nrecognition.”); Carruthers v. Flaum, 365 F. Supp. 2d 448, 451, 466–67\\n(S.D.N.Y. 2005) (concluding that because the Unkechaug are not federally\\nrecognized, IGRA provides no exception to New York’s general\\nprohibition on gambling); First Am. Casino Corp. v. E. Pequot Nation,\\n175 F. Supp. 2d 205, 208 (D. Conn. 2000) (“Unless and until defendant\\nobtains federal acknowledgment, its activities are not regulated by\\nIGRA.”).\\n\\f FRANKS LANDING INDIAN COMTY. V. 17\\n NAT’L INDIAN GAMING COMM’N\\n\\n“Indian tribe.” IGRA does not authorize the non-federally\\nrecognized Community to engage in class II gaming.\\n\\n B. The Frank’s Landing Act Does Not Authorize the\\n Community to Engage in Indian Gaming.\\n\\n Despite IGRA’s requirement that an Indian tribe must be\\nrecognized by the Secretary, the Community argues that the\\nFrank’s Landing Act separately granted it permission to\\nengage in class II gaming. Specifically, the Community\\nargues that because Congress defined “Indian tribe” in IGRA\\nwith the “precise language” used in the Frank’s Landing Act,\\nCongress intended to authorize the Community to engage in\\nclass II gaming.\\n\\n Both the 1987 Frank’s Landing Act and the 1994\\namended Act provide that the Community is:\\n\\n recognized as eligible for the special\\n programs and services provided by the United\\n States to Indians because of their status as\\n Indians and is recognized as eligible to\\n contract, and to receive grants, under the\\n Indian Self-Determination and Education\\n Assistance Act for such services. . . .\\n\\nPub. L. No. 100-153, 101 Stat. 886, 889; § 10, Pub. L. No.\\n103-435, 108 Stat. 4566, 4569. Plainly, this language\\nindicates that Congress intended the Community to be\\neligible for special programs and services by virtue of their\\nstatus as Indians, and Congress made clear that it intended the\\nCommunity would contract for, and receive grants pursuant\\nto, the Indian Self-Determination and Education Assistance\\n\\f18 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\nAct. See Indian Self-Determination and Education Assistance\\nAct, 25 U.S.C. § 5321(a)(1)(E) (providing that upon request,\\nthe Secretary shall enter into self-determination contracts that\\ninclude programs “for the benefit of Indians because of their\\nstatus as Indians[.]”). But neither the 1987 Frank’s Landing\\nAct nor the 1994 amendments contain IGRA’s qualifying\\nphrase signaling federal recognition, “recognized as eligible\\nby the Secretary[,]” 25 U.S.C. § 2703(5) (emphasis added).\\nFurther, Congress underscored that the 1994 amendments\\nwere not intended to federally recognize the Community by\\nspecifying that “[n]othing in this section may be construed to\\nconstitute the recognition by the United States that the\\n[Community] is a federally recognized Indian tribe.” Pub. L.\\nNo. 103-435, § 8, 108 Stat. 4569.\\n\\n The Community argues that we should infer that Congress\\nintended to allow it to engage in class I or II gaming because\\nthe amended Act only expressly prohibits the Community\\nfrom engaging in class III gaming. To support this argument,\\nthe Community invokes the rule that a statute should be\\nconstrued so that no clause, sentence, or word is rendered\\nsuperfluous, and argues that there would have been no need\\nto expressly prohibit it from engaging in class III gaming if\\nlack of federal recognition rendered it ineligible to participate\\nin all IGRA gaming. Interior responds that Congress would\\nnot choose such an indirect route to convey an easily\\nexpressed message, and that there is no indication that this\\nclass III provision means anything other than what it\\nsays—that notwithstanding any other provision of law, i.e.,\\nnotwithstanding federal recognition at some future time, the\\nCommunity is prohibited from engaging in class III gaming.\\n\\f FRANKS LANDING INDIAN COMTY. V. 19\\n NAT’L INDIAN GAMING COMM’N\\n\\n Like the district court, we read the amended Act’s\\nprohibition of class III gaming as an express limitation on the\\nCommunity’s class III gaming rights if it is ever federally\\nrecognized. The words Congress used make this clear.\\nSubsections (b)(1) and (b)(2) provide that “[n]othing in this\\nsection may be construed to” alter or affect the criminal\\njurisdiction of Washington under Public Law 2807 or to\\nconstitute federal recognition of the Community. Congress\\nemployed language with a much broader sweep when it\\ndrafted subsection (b)(3). That subsection states\\n“[n]otwithstanding any other provision of law” the\\nCommunity shall not engage in class III gaming. The\\nSupreme Court recently observed that “[t]he ordinary\\nmeaning of ‘notwithstanding’ is ‘in spite of,’ or ‘without\\nprevention or obstruction from or by.’” N.L.R.B. v. SW Gen.,\\nInc., — U.S. —, 137 S. Ct. 929, 939 (2017) (quoting\\nWebster’s Third New International Dictionary 1545 (1986);\\nBlack’s Law Dictionary 1091 (7th ed. 1999)). The Court\\nexplained that the word “notwithstanding” demonstrates\\n“which provision prevails in the event of a clash,” id.\\n(quoting A. Scalia & B. Garner, Reading Law: The\\nInterpretation of Legal Texts 126–27 (2012)), and that a\\n“notwithstanding” clause “confirms rather than constrains\\nbreadth.” Id. at 940.\\n\\n Accordingly, we read “notwithstanding” in\\nsubsection (b)(3) as a signal that the amended Act’s broad\\nprohibition on class III gaming shall prevail in the event of a\\nconflict with “any other provision of law.” In contrast, the\\npreface to subsections (b)(1) and (b)(2)—“[n]othing in this\\nsection”—constrains the reach of the amended Act with\\n\\n 7\\n 18 U.S.C. § 1162.\\n\\f20 FRANKS LANDING INDIAN COMTY. V.\\n NAT’L INDIAN GAMING COMM’N\\n\\nrespect to Public Law 280 and federal recognition. This\\nvariation in language reflects a deliberate choice, see S.E.C.\\nv. McCarthy, 322 F.3d 650, 656 (9th Cir. 2003) (“the use of\\ndifferent words or terms within a statute demonstrates that\\nCongress intended to convey a different meaning for those\\nwords.”).\\n\\n The historical context of the Frank’s Landing Act also\\nconfirms that the Community is not authorized to engage in\\ngaming. On the same day Congress amended the Frank’s\\nLanding Act, November 2, 1994, see 108 Stat. 4566, 4569, it\\ncodified Interior’s longstanding Part 83 practice of publishing\\na list of recognized tribes, see 108 Stat. 4791. As explained,\\nthe List Act directed the Secretary to annually publish “a list\\nof all Indian tribes which the Secretary recognizes to be\\neligible for the special programs and services provided by the\\nUnited States to Indians because of their status as Indians.”\\n25 U.S.C. § 5131. Assuming, as we must, that Congress was\\nknowledgeable about Interior’s longstanding Part 83\\nregulations, IGRA’s requirement that tribes be recognized by\\nthe Secretary, and the List Act’s codification of the\\nSecretary’s annual list, it is conspicuous that Congress did not\\ntake the opportunity when it amended the Frank’s Landing\\nAct to specifically authorize the Community to engage in\\nclass II gaming. Indeed, Congress did just the opposite by\\nunequivocally stating that the Frank’s Landing Act\\namendments were not intended to confer federal recognition.\\nSee Pub. Law No. 103-435, 108 Stat. 4566, 4569.\\n\\n In light of the express statement in the Frank’s Landing\\nAct that the Community is not a federally recognized tribe,\\nthe amended Act’s broad limitation on class III gaming\\nrights, and the context in which the Act was amended, we\\n\\f FRANKS LANDING INDIAN COMTY. V. 21\\n NAT’L INDIAN GAMING COMM’N\\n\\nconclude it is not susceptible to more than one reasonable\\ninterpretation. The Frank’s Landing Act does not separately\\nauthorize the Community to engage in Indian gaming.\\n\\n CONCLUSION\\n\\n Congress clearly and unambiguously intended federal\\nrecognition by the Secretary to be a prerequisite for\\nparticipation in IGRA gaming. Congress did not separately\\nauthorize the non-federally recognized Community to engage\\nin class II gaming by amending the Frank’s Landing Act.\\nTherefore, Interior correctly concluded that the Community\\nwas not eligible to engage in IGRA gaming and we affirm the\\ndistrict court’s order granting summary judgment in favor of\\nthe appellees.\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1282,"cells":{"id":{"kind":"number","value":4599148,"string":"4,599,148"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"homeaway.com, Inc. v. City of Santa Monica"},"decision_date":{"kind":"string","value":"2019-03-13"},"docket_number":{"kind":"string","value":"18-55367"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Mary M. Schroeder and Jacqueline H. Nguyen, Circuit Judges, and Michael H. Simon, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"NGUYEN, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nHOMEAWAY.COM, INC.; AIRBNB No. 18-55367\\nINC.,\\n Plaintiffs-Appellants, D.C. Nos.\\n 2:16-cv-06641-\\n v. ODW-AFM\\n 2:16-cv-06645-\\nCITY OF SANTA MONICA, ODW-AFM\\n Defendant-Appellee.\\n\\n\\n\\nHOMEAWAY.COM, INC., No. 18-55805\\n Plaintiff-Appellant,\\n D.C. Nos.\\n and 2:16-cv-06641-\\n ODW-AFM\\nAIRBNB INC., 2:16-cv-06645-\\n Plaintiff, ODW-AFM\\n\\n v.\\n\\nCITY OF SANTA MONICA,\\n Defendant-Appellee.\\n\\f2 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\n\\n AIRBNB INC., No. 18-55806\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:16-cv-06645-\\n ODW-AFM\\n CITY OF SANTA MONICA,\\n Defendant-Appellee. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Otis D. Wright II, District Judge, Presiding\\n\\n Argued and Submitted October 12, 2018\\n Pasadena, California\\n\\n Filed March 13, 2019\\n\\n Before: Mary M. Schroeder and Jacqueline H. Nguyen,\\n Circuit Judges, and Michael H. Simon, * District Judge.\\n\\n Opinion by Judge Nguyen\\n\\n\\n\\n\\n *\\n The Honorable Michael H. Simon, United States District Judge for\\nthe District of Oregon, sitting by designation.\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 3\\n\\n SUMMARY **\\n\\n\\n Civil Rights\\n\\n The panel affirmed the district court’s dismissal, for\\nfailure to state a claim, of a complaint brought by\\nHomeAway.com and Airbnb Inc. challenging the City of\\nSanta Monica’s Ordinance 2535, which imposes various\\nobligations on companies that host online platforms for\\nshort-term vacation rentals.\\n\\n The Ordinance, as amended in 2017, imposes four\\nobligations on hosting platforms: (1) collecting and\\nremitting Transient Occupancy Taxes; (2) regularly\\ndisclosing listings and booking information to the City; (3)\\nrefraining from booking properties not licensed and listed on\\nthe City’s registry; (4) and refraining from collecting a fee\\nfor ancillary services.\\n\\n The panel rejected plaintiffs’ assertion that the\\nOrdinance violated the Communications Decency Act of\\n1996, 47 U.S.C. § 230 because it required them to monitor\\nand remove third-party content, and therefore interfered with\\nfederal policy protecting internet companies from liability\\nfor posting third-party content. The panel stated that the\\nOrdinance prohibits processing transactions for unregistered\\nproperties. It does not require the Platforms to review the\\ncontent provided by the hosts of listings on their websites.\\nRather, the panel noted that the only monitoring that\\nappeared necessary in order to comply with the Ordinance\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f4 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nrelated to incoming requests to complete a booking\\ntransaction—content that, while resulting from the third-\\nparty listings, was distinct, internal, and nonpublic. The\\npanel concluded that the Ordinance was not inconsistent\\nwith the Communications Decency Act, and therefore was\\nnot expressly preempted by its terms. The panel further\\nconcluded that the Ordinance would not pose an obstacle to\\nCongress’s aim to encourage self-monitoring of third-party\\ncontent, and therefore obstacle preemption did not preclude\\nSanta Monica from enforcing the Ordinance.\\n\\n The panel held that the Ordinance did not implicate\\nspeech protected by the First Amendment, concluding that\\nthe Ordinance’s prohibitions regulate nonexpressive\\nconduct, specifically booking transactions, and do not single\\nout those engaged in expressive activity.\\n\\n\\n COUNSEL\\n\\nDonald B. Verrilli, Jr. (argued) and Chad Golder, Munger,\\nTolles & Olson LLP, Washington, D.C.; Joseph W. Cotchett\\nand Alexandra P. Summer, Cotchett, Pitre & McCarthy,\\nLLP, Santa Monica, California; Jonathan H. Blavin and\\nJoshua Patashnik, Munger, Tolles & Olson LLP, San\\nFrancisco, California; John B. Major, Munger, Tolles &\\nOlson, Los Angeles, California; for Plaintiff-Appellant\\nAirbnb, Inc.\\n\\nStephen M. Rummage and Ambika K. Doran, Davis Wright\\nTremaine LLP, Seattle, Washington, for Plaintiff-Appellant\\nHomeAway.com, Inc.\\n\\nGeorge S. Cardona (argued), Deputy City Attorney; Lane\\nDilg, City Attorney; Yibin Shen, Chief Deputy City\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 5\\n\\nAttorney; Heidi Von Tongeln and Michael R. Cobden,\\nDeputy City Attorneys; Santa Monica City Attorney’s\\nOffice, Santa Monica, California, for Defendant-Appellee.\\n\\nDavid Salmons, Bryan Killian, and James Nelson, Morgan,\\nLewis & Bockius LLP, Washington, D.C., for Amici Curiae\\nChris Cox and NetChoice.\\n\\nCatherine R. Gellis, Esq., Sausalito, California, for Amicus\\nCuriae Floor64, Inc. d/b/a The Copia Institute.\\n\\nIan C. Ballon and Lori Chang, Greenberg Traurig, LLP, Los\\nAngeles, California, for Amici Curiae Ebay Inc., Glassdoor,\\nInc., Lyft, Inc., Offerup, Inc., Taskrabbit, Inc., Thumbtack,\\nInc., Uber Technologies, Inc., and Upwork, Inc.\\n\\nChristi Hogin, Best Best & Krieger LLP, Manhattan Beach,\\nCalifornia, for Amici Curiae League of California Cities,\\nInternational Municipal Lawyers Association, and\\nCalifornia State Association of Counties.\\n\\nJeremy B. Rosen, Eric S. Boorstin, and Ryan C. Chapman,\\nHorvitz & Levy LLP, Burbank, California; Michael T.\\nWilliams and Allison R. Mclaughlin, Wheeler Trigg\\nO’Donnell LLP, Denver, Colorado; David C. Frederick,\\nBrendan J. Crimmins, and Rachel P. May, Kellogg, Hansen,\\nTodd, Figel & Frederick, P.L.L.C., Washington, D.C., for\\nAmicus Curiae Apartment Investment and Management\\nCompany.\\n\\nHeidi Palutke, Sacramento, California, for Amicus Curiae\\nCalifornia Apartment Association.\\n\\nEdward M. Schulman, Ballston Tower, Arlington, Virginia,\\nfor Amicus Curiae AvalonBay Communities, Inc.\\n\\f6 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nGary S. Kessler, Kulik Gottesman Siegel & Ware LLP,\\nSherman Oaks, California, for Amicus Curiae Community\\nAssociations Institute.\\n\\nAbbey R. Stemler and Matthew C. Turk, Indiana University\\nKelley School of Business, Bloomington, Indiana; Jahan C.\\nSagafi and Relic Sun, Outten & Golden LLP, San Francisco,\\nCalifornia; Peter Romer-Friedman, Outten & Golden LLP,\\nWashington, D.C., for Amici Curiae Internet, Business, and\\nLocal Government Law Professors.\\n\\nRichard G. McCracken and Paul L. More, McCracken,\\nStemerman & Holsberry, LLP, San Francisco, California,\\nfor Amicus Curiae UNITE HERE International Union.\\n\\nDennis J. Herrera, San Francisco City Attorney; Christine\\nVan Aken, Chief of Appellate Litigation; Yvonne R. Meré,\\nChief of Complex and Affirmative Litigation; Sara J.\\nEisenberg, Deputy City Attorney; San Francisco, California\\nfor Amicus Curiae City and County of San Francisco.\\n\\nKarl A. Racine, Attorney General for the District of\\nColumbia, Washington, D.C., for Amicus Curiae District of\\nColumbia.\\n\\nAndre M. Davis, City Solicitor, City of Baltimore,\\nBaltimore, Maryland, for Amicus Curiae Mayor and City\\nCouncil of Baltimore.\\n\\nZach Klein, Columbus City Attorney, Columbus, Ohio, for\\nAmicus Curiae City of Columbus.\\n\\nBarbara J. Parker, Oakland City Attorney; Maria Bee,\\nSpecial Counsel; Erin Bernstein, Supervising Deputy City\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 7\\n\\nAttorney; Oakland, California; for Amicus Curiae City of\\nOakland.\\n\\nPeter S. Holmes, Seattle City Attorney, Seattle, Washington,\\nfor Amicus Curiae City of Seattle.\\n\\nJill E. Habig, Founder & President; Joanna Pearl, Legal\\nDirector; Oakland, California; for Amicus Curiae Public\\nRights Project, A Project of Tides Center.\\n\\n\\n OPINION\\n\\nNGUYEN, Circuit Judge:\\n\\n Located on the coast of Southern California, the city of\\nSanta Monica consists of only about eight square miles but\\nserves 90,000 residents and as many as 500,000 visitors on\\nweekends and holidays. Similar to other popular tourist\\ndestinations, Santa Monica is struggling to manage the\\ndisruptions brought about by the rise of short-term rentals\\nfacilitated by innovative startups such as Appellants\\nHomeAway.com, Inc. and Airbnb Inc. (the “Platforms”).\\nWebsites like those operated by the Platforms are essentially\\nonline marketplaces that allow “guests” seeking\\naccommodations and “hosts” offering accommodations to\\nconnect and enter into rental agreements with one another. 1\\nAs of February 2018, Airbnb had approximately 1,400\\nlistings in Santa Monica, of which about 30 percent are in\\n\\n 1\\n The Platforms do not own, lease, or manage any of the properties\\nlisted on their websites, nor are they parties to the rental agreements.\\nInstead, the content provided alongside the listings—such as description,\\nprice, and availability—are provided by the hosts. For their services, the\\nPlatforms collect a fee from each successful booking.\\n\\f8 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nthe “coastal zone” covered by the California Coastal Act,\\nwhile HomeAway.com had approximately 300 live listings\\nin Santa Monica, of which approximately 40 percent are in\\nthe coastal zone.\\n\\n Santa Monica’s council reported that the proliferation of\\nshort-term rentals had negatively impacted the quality and\\ncharacter of its neighborhoods by “bringing commercial\\nactivity and removing residential housing stock from the\\nmarket” at a time when California is already suffering from\\nsevere housing shortages. In response, the city passed an\\nordinance regulating the short-term vacation rental market\\nby authorizing licensed “home-sharing” (rentals where\\nresidents remain on-site with guests) but prohibiting all other\\nshort-term home rentals of 30 consecutive days or less.\\n\\n ]The Platforms filed suit, alleging that the city ordinance\\nis preempted by the Communications Decency Act and\\nimpermissibly infringes upon their First Amendment rights.\\nThe district court denied preliminary injunctive relief, and\\ndismissed the Platforms’ complaints for failure to state a\\nclaim under the Communications Decency Act and the First\\nAmendment. We affirm.\\n\\n BACKGROUND\\n\\n In May 2015, Santa Monica passed its initial ordinance\\nregulating the short-term vacation rental market by\\nauthorizing licensed “home-sharing” (rentals where\\nresidents remain on-site with guests) but prohibiting all other\\nforms of short-term rentals for 30 consecutive days or less.\\nSanta Monica Ordinance 2484 (May 12, 2015), codified as\\namended, Santa Monica Mun. Code §§ 6.20.010–6.20.100.\\nThe ordinance reflected the city’s housing goals of\\n“preserving its housing stock and preserving the quality and\\ncharacter of its existing single and multi-family residential\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 9\\n\\nneighborhoods.” Id. As originally enacted, the ordinance\\nprohibited hosting platforms from acting to “undertake,\\nmaintain, authorize, aid, facilitate or advertise any Home-\\nSharing activity” that was not authorized by the city.\\nHosting platforms also were required to collect and remit\\ntaxes, and to regularly disclose listings and booking\\ninformation to the city.\\n\\n The Platforms each filed a complaint in the Central\\nDistrict of California challenging the initial ordinance, and\\nthe district court consolidated the cases for discovery and\\npretrial matters. On September 21, 2016, the parties\\nstipulated to stay the case while the city considered\\namendments to the local ordinance. During the stay period,\\nthe district court for the Northern District of California\\ndenied a preliminary injunction requested by the plaintiffs in\\na separate case challenging a similar ordinance in San\\nFrancisco. See Airbnb Inc. v. City & County of San\\nFrancisco, 217 F. Supp. 3d 1066 (N.D. Cal. 2016). That\\ncase ended in a settlement in which the Platforms agreed to\\ncomply with an amended version of San Francisco’s\\nordinance that prohibited booking unlawful transactions but\\nprovided a safe harbor wherein any platform that complies\\nwith the responsibilities set out in the Ordinance will be\\npresumed to be in compliance with the law.\\n\\n In January 2017, Santa Monica likewise amended its\\nown ordinance. The version challenged here, Ordinance\\n2535 (the “Ordinance”), retains its prohibitions on most\\ntypes of short-term rentals, with the exception of licensed\\nhome-shares. In addition, the Ordinance imposes four\\nobligations on hosting platforms directly: (1) collecting and\\nremitting “Transient Occupancy Taxes,” (2) disclosing\\ncertain listing and booking information regularly,\\n(3) refraining from completing any booking transaction for\\n\\f10 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nproperties not licensed and listed on the City’s registry, and\\n(4) refraining from collecting or receiving a fee for\\n“facilitating or providing services ancillary to a vacation\\nrental or unregistered home-share.” If a housing platform\\noperates in compliance with these obligations, the Ordinance\\nprovides a safe harbor by presuming the platform to be in\\ncompliance with the law. Otherwise, violations are\\npunishable by a fine of up to $500 and/or imprisonment for\\nup to six months.\\n\\n After the district court lifted the stay, the Platforms\\namended their complaint to challenge the revised ordinance\\nand moved for a preliminary injunction. Santa Monica\\nmoved to dismiss the amended complaint. The court denied\\nthe Platforms’ motion for preliminary injunctive relief and\\nsubsequently granted Santa Monica’s motion to dismiss on\\nthe ground that the Platforms failed to state a claim under\\nfederal law, including the Communications Decency Act of\\n1996 and the First Amendment. The district court also\\ndeclined to exercise supplemental jurisdiction over their\\nremaining state-law claims. 2 The Platforms timely appealed\\nthese decisions, and we consolidated the appeals.\\n\\n\\n\\n\\n 2\\n The Platforms do not appeal the district court’s dismissal of other\\nfederal claims brought under the Fourth Amendment and the Stored\\nCommunications Act. Similarly, they do not challenge the court’s\\ndecision not to exercise supplemental jurisdiction over the state-law\\nclaims under the California Coastal Act if we affirm the dismissal of their\\nfederal claims. Because we affirm the district court’s dismissal, we need\\nnot consider the state-law claims. We deny Santa Monica’s motion for\\njudicial notice of its prior enforcement actions because the dispute as to\\nits prior actions relates only to the state-law claims.\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 11\\n\\n JURISDICTION AND STANDARD OF REVIEW\\n\\n We have jurisdiction under 28 U.S.C. § 1291. We\\nreview the district court’s order of dismissal de novo,\\n“accepting all factual allegations in the complaint as true and\\nconstruing them in the light most favorable to the\\nnonmoving party.” Yagman v. Garcetti, 852 F.3d 859, 863\\n(2017) (quoting Ebner v. Fresh, Inc., 838 F.3d 958, 962 (9th\\nCir. 2016)).\\n\\n DISCUSSION\\n\\n I. Communications Decency Act\\n\\n The Communications Decency Act of 1996 (“CDA” or\\nthe “Act”), 47 U.S.C. § 230, provides internet companies\\nwith immunity from certain claims in furtherance of its\\nstated policy “to promote the continued development of the\\nInternet and other interactive computer services.” Id.\\n§ 230(b)(1). Construing this immunity broadly, the\\nPlatforms argue that the Ordinance requires them to monitor\\nand remove third-party content, and therefore violates the\\nCDA by interfering with federal policy protecting internet\\ncompanies from liability for posting third-party content.\\nSanta Monica, on the other hand, argues that the Ordinance\\ndoes not implicate the CDA because it imposes no obligation\\non the Platforms to monitor or edit any listings provided by\\nhosts. Santa Monica contends that the Ordinance is simply\\nan exercise of its right to enact regulations to preserve\\nhousing by curtailing “incentives for landlords to evade rent\\ncontrol laws, evict tenants, and convert residential units into\\nde facto hotels.”\\n\\n We begin our analysis with the text of the CDA. See BP\\nAmerica Prod. Co. v. Burton, 549 U.S. 84, 91 (2006).\\nSection 230(c)(1) states that “[n]o provider or user of an\\n\\f12 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\ninteractive computer service shall be treated as the publisher\\nor speaker of any information provided by another\\ninformation content provider.” Id. § 230(c)(1). The CDA\\nexplicitly preempts inconsistent state laws: “Nothing in this\\nsection shall be construed to prevent any State from\\nenforcing any State law that is consistent with this section.\\nNo cause of action may be brought and no liability may be\\nimposed under any State or local law that is inconsistent with\\nthis section.” Id. § 230(e)(3).\\n\\n We have construed these provisions to extend immunity\\nto “(1) a provider or user of an interactive computer service\\n(2) whom a plaintiff seeks to treat, under a state law cause of\\naction, as a publisher or speaker (3) of information provided\\nby another information content provider.” Barnes v. Yahoo!,\\nInc., 570 F.3d 1096, 1100–01 (9th Cir. 2009). Only the\\nsecond element is at issue here: whether the Ordinance treats\\nthe Platforms as a “publisher or speaker” in a manner that is\\nbarred by the CDA. Although the CDA does not define\\n“publisher,” we have defined “publication” in this context to\\n“involve[] reviewing, editing, and deciding whether to\\npublish or to withdraw from publication third-party\\ncontent.” Id. at 1102 (citing Fair Hous. Council v.\\nRoommates.com, LLC, 521 F.3d 1157, 1170–71 (9th Cir.\\n2008) (en banc)).\\n\\n The Platforms offer two different theories as to how the\\nOrdinance in fact reaches “publication” activities. First, the\\nPlatforms claim that the Ordinance is expressly preempted\\nby the CDA because, as they argue, it implicitly requires\\nthem “to monitor the content of a third-party listing and\\ncompare it against the City’s short-term rental registry\\nbefore allowing any booking to proceed.” Relying on Doe\\nv. Internet Brands, 824 F.3d 846, 851 (9th Cir. 2016), the\\nPlatforms take the view that CDA immunity follows\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 13\\n\\nwhenever a legal duty “affects” how an internet company\\n“monitors” a website.\\n\\n However, the Platforms read Internet Brands too\\nbroadly. In that case, two individuals used the defendant’s\\nwebsite to message and lure the plaintiff to sham auditions\\nwhere she was drugged and raped. Id. at 848. We held that,\\nwhere the website provider was alleged to have known\\nindependently of the ongoing scheme beforehand, the CDA\\ndid not bar an action under state law for failure to warn. Id.\\nat 854. We observed that a duty to warn would not\\n“otherwise affect how [the defendant] publishes or\\nmonitors” user content. Id. at 851. Though the defendant\\ndid, in its business, act as a publisher of third-party content,\\nthe underlying legal duty at issue did not seek to hold the\\ndefendant liable as a “publisher or speaker” of third-party\\ncontent. Id. at 853; see 47 U.S.C. § 230(c)(1). We therefore\\ndeclined to extend CDA immunity to the defendant for the\\nplaintiff’s failure-to-warn claim. Internet Brands, 824 F.3d\\nat 854.\\n\\n We do not read Internet Brands to suggest that CDA\\nimmunity attaches any time a legal duty might lead a\\ncompany to respond with monitoring or other publication\\nactivities. It is not enough that third-party content is\\ninvolved; Internet Brands rejected use of a “but-for” test that\\nwould provide immunity under the CDA solely because a\\ncause of action would not otherwise have accrued but for the\\nthird-party content. Id. at 853. We look instead to what the\\nduty at issue actually requires: specifically, whether the duty\\nwould necessarily require an internet company to monitor\\nthird-party content. See id. at 851, 853.\\n\\n Here, the Ordinance does not require the Platforms to\\nmonitor third-party content and thus falls outside of the\\nCDA’s immunity. The Ordinance prohibits processing\\n\\f14 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\ntransactions for unregistered properties. It does not require\\nthe Platforms to review the content provided by the hosts of\\nlistings on their websites. Rather, the only monitoring that\\nappears necessary in order to comply with the Ordinance\\nrelates to incoming requests to complete a booking\\ntransaction—content that, while resulting from the third-\\nparty listings, is distinct, internal, and nonpublic. As in\\nInternet Brands, it is not enough that the third-party listings\\nare a “but-for” cause of such internal monitoring. See\\n824 F.3d at 853. The text of the CDA is “clear that neither\\nthis subsection nor any other declares a general immunity\\nfrom liability deriving from third-party content.” Barnes,\\n570 F.3d at 1100. To provide broad immunity “every time a\\nwebsite uses data initially obtained from third parties would\\neviscerate [the CDA].” Barnes, 570 F.3d at 1100 (quoting\\nRoommates.com, 521 F.3d at 1171 (9th Cir. 2008) (en\\nbanc)). That is not the result that Congress intended.\\n\\n Nor could a duty to cross-reference bookings against\\nSanta Monica’s property registry give rise to CDA\\nimmunity. While keeping track of the city’s registry is\\n“monitoring” third-party content in the most basic sense,\\nsuch conduct cannot be fairly classified as “publication” of\\nthird-party content. The Platforms have no editorial control\\nover the registry whatsoever. As with tax regulations or\\ncriminal statutes, the Ordinance can fairly charge parties\\nwith keeping abreast of the law without running afoul of the\\nCDA.\\n\\n Second, the Platforms argue that the Ordinance “in\\noperation and effect . . . forces [them] to remove third-party\\ncontent.” Although it is clear that the Ordinance does not\\nexpressly mandate that they do so, the Platforms claim that\\n“common sense explains” that they cannot “leave in place a\\nwebsite chock-full of un-bookable listings.” For purposes of\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 15\\n\\nour review, we accept at face value the Platforms’ assertion\\nthat they will choose to remove noncompliant third-party\\nlistings on their website as a consequence of the Ordinance. 3\\nNonetheless, their choice to remove listings is insufficient to\\nimplicate the CDA.\\n\\n On its face, the Ordinance does not proscribe, mandate,\\nor even discuss the content of the listings that the Platforms\\ndisplay on their websites. See Santa Monica Mun. Code\\n§§ 6.20.010–6.20.100. It requires only that transactions\\ninvolve licensed properties. We acknowledge that, as the\\nPlatforms explain in Airbnb’s complaint and in the briefing\\non appeal, removal of these listings would be the best option\\n“from a business standpoint.” But, as in Internet Brands, the\\nunderlying duty “could have been satisfied without changes\\nto content posted by the website’s users.” See 824 F.3d\\nat 851. Even assuming that removing certain listings may be\\nthe Platforms’ most practical compliance option, allowing\\ninternet companies to claim CDA immunity under these\\ncircumstances would risk exempting them from most local\\nregulations and would, as this court feared in\\nRoommates.com, 521 F.3d at 1164, “create a lawless no-\\nman’s-land on the Internet.” We hold that the Ordinance is\\nnot “inconsistent” with the CDA, and is therefore not\\nexpressly preempted by its terms. See 47 U.S.C. § 230(e)(3).\\n\\n Finally, the Platforms argue that, even if the Ordinance\\nis not expressly preempted by the CDA, the Ordinance\\nimposes “an obstacle to the accomplishment and execution\\n 3\\n The Platforms argued below that the district court must accept as\\ntrue their allegation that they would “have to” monitor and screen\\nlistings. As a matter of law, the Ordinance does not require them to do\\nso. Courts are “not bound to accept as true a legal conclusion couched\\nas a factual allegation.” Bell Atlantic Corp. v. Twombly, 550 U.S. 544,\\n555 (2007) (quoting Papasan v. Allain, 478 U.S. 265, 286 (1986)).\\n\\f16 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nof the full purposes and objectives of Congress.” See Crosby\\nv. Nat’l Foreign Trade Council, 530 U.S. 363, 372–73\\n(2000). Reading the CDA expansively, they argue that the\\nOrdinance conflicts with the CDA’s goal “to preserve the\\nvibrant and competitive free market that presently exists for\\nthe Internet . . . unfettered by Federal or State regulation.”\\nSee § 230(b)(2). We have consistently eschewed an\\nexpansive reading of the statute that would render unlawful\\nconduct “magically . . . lawful when [conducted] online,”\\nand therefore “giv[ing] online businesses an unfair\\nadvantage over their real-world counterparts.” See\\nRoommates.com, 521 F.3d at 1164, 1164–65 n.15. For the\\nsame reasons, while we acknowledge the Platforms’\\nconcerns about the difficulties of complying with numerous\\nstate and local regulations, the CDA does not provide\\ninternet companies with a one-size-fits-all body of law. Like\\ntheir brick-and-mortar counterparts, internet companies\\nmust also comply with any number of local regulations\\nconcerning, for example, employment, tax, or zoning.\\nBecause the Ordinance would not pose an obstacle to\\nCongress’s aim to encourage self-monitoring of third-party\\ncontent, we hold that obstacle preemption does not preclude\\nSanta Monica from enforcing the Ordinance.\\n\\n Fundamentally, the parties dispute how broadly to\\nconstrue the CDA so as to continue serving the purposes\\nCongress envisioned while allowing state and local\\ngovernments breathing room to address the pressing issues\\nfaced by their communities. We have previously\\nacknowledged that the CDA’s immunity reaches beyond the\\ninitial state court decision that sparked its enactment. See\\nFair Hous. Council v. Roommates.com, LLC, 521 F.3d 1157,\\n1163 (9th Cir. 2008) (en banc) (discussing Stratton\\nOakmont, Inc. v. Prodigy Servs. Co., which held an internet\\ncompany liable for defamation when it removed some, but\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 17\\n\\nnot all, harmful content from its public message boards,\\n1995 WL 323710 (N.Y. Sup. Ct. May 24, 1995)\\n(unpublished)). As the Platforms correctly note, the Act’s\\npolicy statements broadly promote “the vibrant and\\ncompetitive free market that presently exists for the Internet\\n. . . unfettered by Federal or State regulation.” See 47 U.S.C.\\n§ 230(b)(2). “[A] law’s scope often differs from its genesis,”\\nand we have repeatedly held the scope of immunity to reach\\nbeyond defamation cases. Barnes, 570 F.3d at 1101 (quoting\\nChicago Lawyers’ Comm. for Civil Rights Under Law, Inc.\\nv. Craigslist, Inc., 519 F.3d 666, 671 (7th Cir. 2008), as\\namended (May 2, 2008)) (citing cases applying immunity for\\ncauses of action including discrimination, fraud, and\\nnegligence).\\n\\n At the same time, our cases have hewn closely to the\\nstatutory language of the CDA and have limited the\\nexpansion of its immunity beyond the protection Congress\\nenvisioned. As we have observed, “the [relevant] section is\\ntitled ‘Protection for “good Samaritan” blocking and\\nscreening of offensive material.’” Roommates.com,\\n521 F.3d at 1163–64 (quoting 47 U.S.C. § 230(c)); see also\\nInternet Brands, 824 F.3d at 852. Congress intended to\\n“spare interactive computer services [the] grim choice”\\nbetween voluntarily filtering content and being subject to\\nliability on the one hand, and “ignoring all problematic posts\\naltogether [to] escape liability.” Roommates.com, 521 F.3d\\nat 1163–64. In contrast, the Platforms face no liability for\\nthe content of the bookings; rather, any liability arises only\\nfrom unlicensed bookings. We do not discount the\\nPlatforms’ concerns about the administrative burdens of\\nstate and local regulations, but we nonetheless disagree that\\n§ 230(c)(1) of the CDA may be read as broadly as they\\nadvocate, or that we may ourselves expand its provisions\\nbeyond what Congress initially intended.\\n\\f18 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\n In sum, neither express preemption nor obstacle\\npreemption apply to the Ordinance. We therefore affirm the\\ndistrict court’s dismissal for failure to state a claim under the\\nCDA.\\n\\n II. First Amendment\\n\\n The Platforms also contend that the district court erred in\\ndismissing their First Amendment claims. They argue that,\\neven if the plain language of the Ordinance only reaches\\n“conduct,” i.e., booking unlicensed properties, the law\\neffectively imposes a “content-based financial burden” on\\ncommercial speech and is thus subject to First Amendment\\nscrutiny. The district court concluded that the Ordinance\\n“regulates conduct, not speech, and that the conduct banned\\n. . . does not have such a ‘significant expressive element’ as\\nto draw First Amendment protection.” We agree.\\n\\n That the Ordinance regulates “conduct” is not alone\\ndispositive. The Supreme Court has previously applied First\\nAmendment scrutiny when “‘speech’ and ‘nonspeech’\\nelements are combined in the same course of conduct.” See\\nUnited States v. O’Brien, 391 U.S. 367, 376 (1968). But\\n“restrictions on protected expression are distinct from\\nrestrictions on economic activity or, more generally, on\\nnonexpressive conduct.” Sorrell v. IMS Health Inc.,\\n564 U.S. 552, 567 (2011). While the former is entitled to\\nprotection, “the First Amendment does not prevent\\nrestrictions directed at commerce or conduct from imposing\\nincidental burdens on speech.” Id.\\n\\n To determine whether the First Amendment applies, we\\nmust first ask the “threshold question [of] whether conduct\\nwith a ‘significant expressive element’ drew the legal\\nremedy or the ordinance has the inevitable effect of ‘singling\\nout those engaged in expressive activity.’” Int’l Franchise\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 19\\n\\nAss’n v. City of Seattle, 803 F.3d 389, 408 (9th Cir. 2015)\\n(quoting Arcara v. Cloud Books, Inc., 478 U.S. 697, 706–07\\n(1986)). A court may consider the “inevitable effect of a\\nstatute on its face,” as well as a statute’s “stated purpose.”\\nSorrell, 564 U.S. at 565. However, absent narrow\\ncircumstances, a court may not conduct an inquiry into\\nlegislative purpose or motive beyond what is stated within\\nthe statute itself. See O’Brien, 391 U.S. at 383 n.30.\\nBecause the conduct at issue—completing booking\\ntransactions for unlawful rentals—consists only of\\nnonspeech, nonexpressive conduct, we hold that the\\nOrdinance does not implicate the First Amendment.\\n\\n First, the prohibitions here did not target conduct with “a\\nsignificant expressive element.” See Arcara, 478 U.S.\\nat 706. Our decision in International Franchise Ass’n is\\nanalogous. There, the plaintiff challenged a minimum wage\\nordinance that would have accelerated the raising of the\\nminimum wage to $15 per hour for franchise owners and\\nother large employers. 803 F.3d at 389. In denying a\\npreliminary injunction, the district court held that the\\nplaintiffs were not likely to succeed on their First\\nAmendment argument that the ordinance treated them\\ndifferently based on their “speech and association” decisions\\nto operate within a franchise relationship framework. Id. at\\n408–09. We agreed, concluding that the “business\\nagreement or business dealings” were not conduct with a\\n“significant expressive element.” Id. at 408. Instead,\\n“Seattle’s minimum wage ordinance [was] plainly an\\neconomic regulation that [did] not target speech or\\nexpressive conduct.” Id.\\n\\n Similarly, here, the Ordinance is plainly a housing and\\nrental regulation. The “inevitable effect of the [Ordinance]\\non its face” is to regulate nonexpressive conduct—namely,\\n\\f20 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nbooking transactions—not speech. See Sorrell, 564 U.S.\\nat 565. As in International Franchise Ass’n, the “business\\nagreement or business dealings” associated with processing\\na booking is not conduct with a “significant expressive\\nelement.” See 803 F.3d at 408 (citation and quotation marks\\nomitted). Contrary to the Platforms’ claim, the Ordinance\\ndoes not “require” that they monitor or screen\\nadvertisements. It instead leaves them to decide how best to\\ncomply with the prohibition on booking unlawful\\ntransactions.\\n\\n Nor can the Platforms rely on the Ordinance’s “stated\\npurpose” to argue that it intends to regulate speech. The\\nOrdinance itself makes clear that the City’s “central and\\nsignificant goal . . . is preservation of its housing stock and\\npreserving the quality and nature of residential\\nneighborhoods.” As such, with respect to the Platforms, the\\nonly inevitable effect, and the stated purpose, of the\\nOrdinance is to prohibit them from completing booking\\ntransactions for unlawful rentals.\\n\\n As for the second prong of our inquiry, whether the\\nOrdinance has the effect of “singling out those engaged in\\nexpressive activity,” Arcara, 478 U.S. at 706–07, we\\nconclude that it does not. As the Platforms point out,\\nwebsites like Craigslist “advertise the very same properties,”\\nbut do not process transactions. Unlike the Platforms, those\\nwebsites would not be subject to the Ordinance,\\nunderscoring that the Ordinance does not target websites that\\npost listings, but rather companies that engage in unlawful\\nbooking transactions.\\n\\n Moreover, the incidental impacts on speech cited by the\\nPlatforms raise minimal concerns. The Platforms argue that\\nthe Ordinance chills commercial speech, namely,\\nadvertisements for third-party rentals. But even accepting\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 21\\n\\nthat the Platforms will need to engage in efforts to validate\\ntransactions before completing them, incidental burdens like\\nthese are not always sufficient to trigger First Amendment\\nscrutiny. See Int’l Franchise Ass’n, 803 F.3d at 408\\n(“[S]ubjecting every incidental impact on speech to First\\nAmendment scrutiny ‘would lead to the absurd result that\\nany government action that had some conceivable speech\\ninhibiting consequences . . . would require analysis under\\nthe First Amendment.’” (quoting Arcara, 478 U.S. at 708\\n(O’Connor, J., concurring))). Furthermore, to the extent that\\nthe speech chilled advertises unlawful rentals, “[a]ny First\\nAmendment interest . . . is altogether absent when the\\ncommercial activity itself is illegal and the restriction on\\nadvertising is incidental to a valid limitation on economic\\nactivity.” See Pittsburgh Press Co. v. Pittsburgh Comm’n\\non Human Relations, 413 U.S. 376, 389 (1973).\\n\\n Finally, because the Ordinance does not implicate speech\\nprotected by the First Amendment, we similarly reject the\\nPlatforms’ argument that the Ordinance is unconstitutional\\nwithout a scienter requirement. In most cases, there is no\\n“closed definition” on when a criminal statute must contain\\na scienter requirement. See Morissette v. United States,\\n342 U.S. 246, 260 (1952). However, the Supreme Court has\\ndrawn a bright line in certain contexts, such as holding that\\nthe First Amendment requires statutes imposing criminal\\nliability for obscenity or child pornography to contain a\\nscienter requirement. See New York v. Ferber, 458 U.S. 747,\\n765 (1982). Such a requirement prevents “a severe\\nlimitation on the public’s access to constitutionally protected\\nmatter” as would result from inflexible laws criminalizing\\n“bookshops and periodical stands.” Smith v. California,\\n361 U.S. 147, 153 (1959).\\n\\f22 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\n Here, even assuming that the Ordinance would lead the\\nPlatforms to voluntarily remove some advertisements for\\nlawful rentals, there would not be a “severe limitation on the\\npublic’s access” to lawful advertisements, especially\\nconsidering the existence of alternative channels like\\nCraigslist. Id. Such an incidental burden is far from “a\\nsubstantial restriction on the freedom of speech” that would\\nnecessitate a scienter requirement. Id. at 150. Otherwise,\\n“[t]here is no specific constitutional inhibition against\\nmaking the distributors of good[s] the strictest censors of\\ntheir merchandise.” Id. at 152.\\n\\n III. Remaining Claims\\n\\n On appeal, the Platforms do not challenge dismissal of\\ntheir other federal law claims “in light of the district court’s\\ninterpretation of the Ordinance as only requiring disclosure\\nof information pursuant to requests that comply with the\\nFourth Amendment and Stored Communications Act.”\\nSimilarly, the parties specified that they would “not\\nchallenge the district court’s decision to decline\\nsupplemental jurisdiction if all the Platforms’ federal claims\\nwere properly dismissed.” Accordingly, we need not\\nconsider the remaining claims.\\n\\n * * *\\n\\n Because the district court properly dismissed the\\nPlatforms’ complaints for failure to state a claim, we dismiss\\nas moot the appeals from the denial of preliminary injunctive\\nrelief.\\n\\n AFFIRMED in part, DISMISSED in part.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4376401/\", \"author_raw\": \"NGUYEN, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"MARY M SCHROEDER"},"judge2":{"kind":"string","value":"JACQUELINE H NGUYEN"},"judge3":{"kind":"string","value":"MICHAEL H SIMON"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4599148/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: homeaway.com, Inc. v. City of Santa Monica\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-13\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nHOMEAWAY.COM, INC.; AIRBNB No. 18-55367\nINC.,\n Plaintiffs-Appellants, D.C. Nos.\n 2:16-cv-06641-\n v. ODW-AFM\n 2:16-cv-06645-\nCITY OF SANTA MONICA, ODW-AFM\n Defendant-Appellee.\n\n\n\nHOMEAWAY.COM, INC., No. 18-55805\n Plaintiff-Appellant,\n D.C. Nos.\n and 2:16-cv-06641-\n ODW-AFM\nAIRBNB INC., 2:16-cv-06645-\n Plaintiff, ODW-AFM\n\n v.\n\nCITY OF SANTA MONICA,\n Defendant-Appellee.\n\f2 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\n\n AIRBNB INC., No. 18-55806\n Plaintiff-Appellant,\n D.C. No.\n v. 2:16-cv-06645-\n ODW-AFM\n CITY OF SANTA MONICA,\n Defendant-Appellee. OPINION\n\n\n\n Appeal from the United States District Court\n for the Central District of California\n Otis D. Wright II, District Judge, Presiding\n\n Argued and Submitted October 12, 2018\n Pasadena, California\n\n Filed March 13, 2019\n\n Before: Mary M. Schroeder and Jacqueline H. Nguyen,\n Circuit Judges, and Michael H. Simon, * District Judge.\n\n Opinion by Judge Nguyen\n\n\n\n\n *\n The Honorable Michael H. Simon, United States District Judge for\nthe District of Oregon, sitting by designation.\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 3\n\n SUMMARY **\n\n\n Civil Rights\n\n The panel affirmed the district court’s dismissal, for\nfailure to state a claim, of a complaint brought by\nHomeAway.com and Airbnb Inc. challenging the City of\nSanta Monica’s Ordinance 2535, which imposes various\nobligations on companies that host online platforms for\nshort-term vacation rentals.\n\n The Ordinance, as amended in 2017, imposes four\nobligations on hosting platforms: (1) collecting and\nremitting Transient Occupancy Taxes; (2) regularly\ndisclosing listings and booking information to the City; (3)\nrefraining from booking properties not licensed and listed on\nthe City’s registry; (4) and refraining from collecting a fee\nfor ancillary services.\n\n The panel rejected plaintiffs’ assertion that the\nOrdinance violated the Communications Decency Act of\n1996, 47 U.S.C. § 230 because it required them to monitor\nand remove third-party content, and therefore interfered with\nfederal policy protecting internet companies from liability\nfor posting third-party content. The panel stated that the\nOrdinance prohibits processing transactions for unregistered\nproperties. It does not require the Platforms to review the\ncontent provided by the hosts of listings on their websites.\nRather, the panel noted that the only monitoring that\nappeared necessary in order to comply with the Ordinance\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f4 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\nrelated to incoming requests to complete a booking\ntransaction—content that, while resulting from the third-\nparty listings, was distinct, internal, and nonpublic. The\npanel concluded that the Ordinance was not inconsistent\nwith the Communications Decency Act, and therefore was\nnot expressly preempted by its terms. The panel further\nconcluded that the Ordinance would not pose an obstacle to\nCongress’s aim to encourage self-monitoring of third-party\ncontent, and therefore obstacle preemption did not preclude\nSanta Monica from enforcing the Ordinance.\n\n The panel held that the Ordinance did not implicate\nspeech protected by the First Amendment, concluding that\nthe Ordinance’s prohibitions regulate nonexpressive\nconduct, specifically booking transactions, and do not single\nout those engaged in expressive activity.\n\n\n COUNSEL\n\nDonald B. Verrilli, Jr. (argued) and Chad Golder, Munger,\nTolles & Olson LLP, Washington, D.C.; Joseph W. Cotchett\nand Alexandra P. Summer, Cotchett, Pitre & McCarthy,\nLLP, Santa Monica, California; Jonathan H. Blavin and\nJoshua Patashnik, Munger, Tolles & Olson LLP, San\nFrancisco, California; John B. Major, Munger, Tolles &\nOlson, Los Angeles, California; for Plaintiff-Appellant\nAirbnb, Inc.\n\nStephen M. Rummage and Ambika K. Doran, Davis Wright\nTremaine LLP, Seattle, Washington, for Plaintiff-Appellant\nHomeAway.com, Inc.\n\nGeorge S. Cardona (argued), Deputy City Attorney; Lane\nDilg, City Attorney; Yibin Shen, Chief Deputy City\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 5\n\nAttorney; Heidi Von Tongeln and Michael R. Cobden,\nDeputy City Attorneys; Santa Monica City Attorney’s\nOffice, Santa Monica, California, for Defendant-Appellee.\n\nDavid Salmons, Bryan Killian, and James Nelson, Morgan,\nLewis & Bockius LLP, Washington, D.C., for Amici Curiae\nChris Cox and NetChoice.\n\nCatherine R. Gellis, Esq., Sausalito, California, for Amicus\nCuriae Floor64, Inc. d/b/a The Copia Institute.\n\nIan C. Ballon and Lori Chang, Greenberg Traurig, LLP, Los\nAngeles, California, for Amici Curiae Ebay Inc., Glassdoor,\nInc., Lyft, Inc., Offerup, Inc., Taskrabbit, Inc., Thumbtack,\nInc., Uber Technologies, Inc., and Upwork, Inc.\n\nChristi Hogin, Best Best & Krieger LLP, Manhattan Beach,\nCalifornia, for Amici Curiae League of California Cities,\nInternational Municipal Lawyers Association, and\nCalifornia State Association of Counties.\n\nJeremy B. Rosen, Eric S. Boorstin, and Ryan C. Chapman,\nHorvitz & Levy LLP, Burbank, California; Michael T.\nWilliams and Allison R. Mclaughlin, Wheeler Trigg\nO’Donnell LLP, Denver, Colorado; David C. Frederick,\nBrendan J. Crimmins, and Rachel P. May, Kellogg, Hansen,\nTodd, Figel & Frederick, P.L.L.C., Washington, D.C., for\nAmicus Curiae Apartment Investment and Management\nCompany.\n\nHeidi Palutke, Sacramento, California, for Amicus Curiae\nCalifornia Apartment Association.\n\nEdward M. Schulman, Ballston Tower, Arlington, Virginia,\nfor Amicus Curiae AvalonBay Communities, Inc.\n\f6 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\nGary S. Kessler, Kulik Gottesman Siegel & Ware LLP,\nSherman Oaks, California, for Amicus Curiae Community\nAssociations Institute.\n\nAbbey R. Stemler and Matthew C. Turk, Indiana University\nKelley School of Business, Bloomington, Indiana; Jahan C.\nSagafi and Relic Sun, Outten & Golden LLP, San Francisco,\nCalifornia; Peter Romer-Friedman, Outten & Golden LLP,\nWashington, D.C., for Amici Curiae Internet, Business, and\nLocal Government Law Professors.\n\nRichard G. McCracken and Paul L. More, McCracken,\nStemerman & Holsberry, LLP, San Francisco, California,\nfor Amicus Curiae UNITE HERE International Union.\n\nDennis J. Herrera, San Francisco City Attorney; Christine\nVan Aken, Chief of Appellate Litigation; Yvonne R. Meré,\nChief of Complex and Affirmative Litigation; Sara J.\nEisenberg, Deputy City Attorney; San Francisco, California\nfor Amicus Curiae City and County of San Francisco.\n\nKarl A. Racine, Attorney General for the District of\nColumbia, Washington, D.C., for Amicus Curiae District of\nColumbia.\n\nAndre M. Davis, City Solicitor, City of Baltimore,\nBaltimore, Maryland, for Amicus Curiae Mayor and City\nCouncil of Baltimore.\n\nZach Klein, Columbus City Attorney, Columbus, Ohio, for\nAmicus Curiae City of Columbus.\n\nBarbara J. Parker, Oakland City Attorney; Maria Bee,\nSpecial Counsel; Erin Bernstein, Supervising Deputy City\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 7\n\nAttorney; Oakland, California; for Amicus Curiae City of\nOakland.\n\nPeter S. Holmes, Seattle City Attorney, Seattle, Washington,\nfor Amicus Curiae City of Seattle.\n\nJill E. Habig, Founder & President; Joanna Pearl, Legal\nDirector; Oakland, California; for Amicus Curiae Public\nRights Project, A Project of Tides Center.\n\n\n OPINION\n\nNGUYEN, Circuit Judge:\n\n Located on the coast of Southern California, the city of\nSanta Monica consists of only about eight square miles but\nserves 90,000 residents and as many as 500,000 visitors on\nweekends and holidays. Similar to other popular tourist\ndestinations, Santa Monica is struggling to manage the\ndisruptions brought about by the rise of short-term rentals\nfacilitated by innovative startups such as Appellants\nHomeAway.com, Inc. and Airbnb Inc. (the “Platforms”).\nWebsites like those operated by the Platforms are essentially\nonline marketplaces that allow “guests” seeking\naccommodations and “hosts” offering accommodations to\nconnect and enter into rental agreements with one another. 1\nAs of February 2018, Airbnb had approximately 1,400\nlistings in Santa Monica, of which about 30 percent are in\n\n 1\n The Platforms do not own, lease, or manage any of the properties\nlisted on their websites, nor are they parties to the rental agreements.\nInstead, the content provided alongside the listings—such as description,\nprice, and availability—are provided by the hosts. For their services, the\nPlatforms collect a fee from each successful booking.\n\f8 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\nthe “coastal zone” covered by the California Coastal Act,\nwhile HomeAway.com had approximately 300 live listings\nin Santa Monica, of which approximately 40 percent are in\nthe coastal zone.\n\n Santa Monica’s council reported that the proliferation of\nshort-term rentals had negatively impacted the quality and\ncharacter of its neighborhoods by “bringing commercial\nactivity and removing residential housing stock from the\nmarket” at a time when California is already suffering from\nsevere housing shortages. In response, the city passed an\nordinance regulating the short-term vacation rental market\nby authorizing licensed “home-sharing” (rentals where\nresidents remain on-site with guests) but prohibiting all other\nshort-term home rentals of 30 consecutive days or less.\n\n ]The Platforms filed suit, alleging that the city ordinance\nis preempted by the Communications Decency Act and\nimpermissibly infringes upon their First Amendment rights.\nThe district court denied preliminary injunctive relief, and\ndismissed the Platforms’ complaints for failure to state a\nclaim under the Communications Decency Act and the First\nAmendment. We affirm.\n\n BACKGROUND\n\n In May 2015, Santa Monica passed its initial ordinance\nregulating the short-term vacation rental market by\nauthorizing licensed “home-sharing” (rentals where\nresidents remain on-site with guests) but prohibiting all other\nforms of short-term rentals for 30 consecutive days or less.\nSanta Monica Ordinance 2484 (May 12, 2015), codified as\namended, Santa Monica Mun. Code §§ 6.20.010–6.20.100.\nThe ordinance reflected the city’s housing goals of\n“preserving its housing stock and preserving the quality and\ncharacter of its existing single and multi-family residential\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 9\n\nneighborhoods.” Id. As originally enacted, the ordinance\nprohibited hosting platforms from acting to “undertake,\nmaintain, authorize, aid, facilitate or advertise any Home-\nSharing activity” that was not authorized by the city.\nHosting platforms also were required to collect and remit\ntaxes, and to regularly disclose listings and booking\ninformation to the city.\n\n The Platforms each filed a complaint in the Central\nDistrict of California challenging the initial ordinance, and\nthe district court consolidated the cases for discovery and\npretrial matters. On September 21, 2016, the parties\nstipulated to stay the case while the city considered\namendments to the local ordinance. During the stay period,\nthe district court for the Northern District of California\ndenied a preliminary injunction requested by the plaintiffs in\na separate case challenging a similar ordinance in San\nFrancisco. See Airbnb Inc. v. City & County of San\nFrancisco, 217 F. Supp. 3d 1066 (N.D. Cal. 2016). That\ncase ended in a settlement in which the Platforms agreed to\ncomply with an amended version of San Francisco’s\nordinance that prohibited booking unlawful transactions but\nprovided a safe harbor wherein any platform that complies\nwith the responsibilities set out in the Ordinance will be\npresumed to be in compliance with the law.\n\n In January 2017, Santa Monica likewise amended its\nown ordinance. The version challenged here, Ordinance\n2535 (the “Ordinance”), retains its prohibitions on most\ntypes of short-term rentals, with the exception of licensed\nhome-shares. In addition, the Ordinance imposes four\nobligations on hosting platforms directly: (1) collecting and\nremitting “Transient Occupancy Taxes,” (2) disclosing\ncertain listing and booking information regularly,\n(3) refraining from completing any booking transaction for\n\f10 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\nproperties not licensed and listed on the City’s registry, and\n(4) refraining from collecting or receiving a fee for\n“facilitating or providing services ancillary to a vacation\nrental or unregistered home-share.” If a housing platform\noperates in compliance with these obligations, the Ordinance\nprovides a safe harbor by presuming the platform to be in\ncompliance with the law. Otherwise, violations are\npunishable by a fine of up to $500 and/or imprisonment for\nup to six months.\n\n After the district court lifted the stay, the Platforms\namended their complaint to challenge the revised ordinance\nand moved for a preliminary injunction. Santa Monica\nmoved to dismiss the amended complaint. The court denied\nthe Platforms’ motion for preliminary injunctive relief and\nsubsequently granted Santa Monica’s motion to dismiss on\nthe ground that the Platforms failed to state a claim under\nfederal law, including the Communications Decency Act of\n1996 and the First Amendment. The district court also\ndeclined to exercise supplemental jurisdiction over their\nremaining state-law claims. 2 The Platforms timely appealed\nthese decisions, and we consolidated the appeals.\n\n\n\n\n 2\n The Platforms do not appeal the district court’s dismissal of other\nfederal claims brought under the Fourth Amendment and the Stored\nCommunications Act. Similarly, they do not challenge the court’s\ndecision not to exercise supplemental jurisdiction over the state-law\nclaims under the California Coastal Act if we affirm the dismissal of their\nfederal claims. Because we affirm the district court’s dismissal, we need\nnot consider the state-law claims. We deny Santa Monica’s motion for\njudicial notice of its prior enforcement actions because the dispute as to\nits prior actions relates only to the state-law claims.\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 11\n\n JURISDICTION AND STANDARD OF REVIEW\n\n We have jurisdiction under 28 U.S.C. § 1291. We\nreview the district court’s order of dismissal de novo,\n“accepting all factual allegations in the complaint as true and\nconstruing them in the light most favorable to the\nnonmoving party.” Yagman v. Garcetti, 852 F.3d 859, 863\n(2017) (quoting Ebner v. Fresh, Inc., 838 F.3d 958, 962 (9th\nCir. 2016)).\n\n DISCUSSION\n\n I. Communications Decency Act\n\n The Communications Decency Act of 1996 (“CDA” or\nthe “Act”), 47 U.S.C. § 230, provides internet companies\nwith immunity from certain claims in furtherance of its\nstated policy “to promote the continued development of the\nInternet and other interactive computer services.” Id.\n§ 230(b)(1). Construing this immunity broadly, the\nPlatforms argue that the Ordinance requires them to monitor\nand remove third-party content, and therefore violates the\nCDA by interfering with federal policy protecting internet\ncompanies from liability for posting third-party content.\nSanta Monica, on the other hand, argues that the Ordinance\ndoes not implicate the CDA because it imposes no obligation\non the Platforms to monitor or edit any listings provided by\nhosts. Santa Monica contends that the Ordinance is simply\nan exercise of its right to enact regulations to preserve\nhousing by curtailing “incentives for landlords to evade rent\ncontrol laws, evict tenants, and convert residential units into\nde facto hotels.”\n\n We begin our analysis with the text of the CDA. See BP\nAmerica Prod. Co. v. Burton, 549 U.S. 84, 91 (2006).\nSection 230(c)(1) states that “[n]o provider or user of an\n\f12 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\ninteractive computer service shall be treated as the publisher\nor speaker of any information provided by another\ninformation content provider.” Id. § 230(c)(1). The CDA\nexplicitly preempts inconsistent state laws: “Nothing in this\nsection shall be construed to prevent any State from\nenforcing any State law that is consistent with this section.\nNo cause of action may be brought and no liability may be\nimposed under any State or local law that is inconsistent with\nthis section.” Id. § 230(e)(3).\n\n We have construed these provisions to extend immunity\nto “(1) a provider or user of an interactive computer service\n(2) whom a plaintiff seeks to treat, under a state law cause of\naction, as a publisher or speaker (3) of information provided\nby another information content provider.” Barnes v. Yahoo!,\nInc., 570 F.3d 1096, 1100–01 (9th Cir. 2009). Only the\nsecond element is at issue here: whether the Ordinance treats\nthe Platforms as a “publisher or speaker” in a manner that is\nbarred by the CDA. Although the CDA does not define\n“publisher,” we have defined “publication” in this context to\n“involve[] reviewing, editing, and deciding whether to\npublish or to withdraw from publication third-party\ncontent.” Id. at 1102 (citing Fair Hous. Council v.\nRoommates.com, LLC, 521 F.3d 1157, 1170–71 (9th Cir.\n2008) (en banc)).\n\n The Platforms offer two different theories as to how the\nOrdinance in fact reaches “publication” activities. First, the\nPlatforms claim that the Ordinance is expressly preempted\nby the CDA because, as they argue, it implicitly requires\nthem “to monitor the content of a third-party listing and\ncompare it against the City’s short-term rental registry\nbefore allowing any booking to proceed.” Relying on Doe\nv. Internet Brands, 824 F.3d 846, 851 (9th Cir. 2016), the\nPlatforms take the view that CDA immunity follows\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 13\n\nwhenever a legal duty “affects” how an internet company\n“monitors” a website.\n\n However, the Platforms read Internet Brands too\nbroadly. In that case, two individuals used the defendant’s\nwebsite to message and lure the plaintiff to sham auditions\nwhere she was drugged and raped. Id. at 848. We held that,\nwhere the website provider was alleged to have known\nindependently of the ongoing scheme beforehand, the CDA\ndid not bar an action under state law for failure to warn. Id.\nat 854. We observed that a duty to warn would not\n“otherwise affect how [the defendant] publishes or\nmonitors” user content. Id. at 851. Though the defendant\ndid, in its business, act as a publisher of third-party content,\nthe underlying legal duty at issue did not seek to hold the\ndefendant liable as a “publisher or speaker” of third-party\ncontent. Id. at 853; see 47 U.S.C. § 230(c)(1). We therefore\ndeclined to extend CDA immunity to the defendant for the\nplaintiff’s failure-to-warn claim. Internet Brands, 824 F.3d\nat 854.\n\n We do not read Internet Brands to suggest that CDA\nimmunity attaches any time a legal duty might lead a\ncompany to respond with monitoring or other publication\nactivities. It is not enough that third-party content is\ninvolved; Internet Brands rejected use of a “but-for” test that\nwould provide immunity under the CDA solely because a\ncause of action would not otherwise have accrued but for the\nthird-party content. Id. at 853. We look instead to what the\nduty at issue actually requires: specifically, whether the duty\nwould necessarily require an internet company to monitor\nthird-party content. See id. at 851, 853.\n\n Here, the Ordinance does not require the Platforms to\nmonitor third-party content and thus falls outside of the\nCDA’s immunity. The Ordinance prohibits processing\n\f14 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\ntransactions for unregistered properties. It does not require\nthe Platforms to review the content provided by the hosts of\nlistings on their websites. Rather, the only monitoring that\nappears necessary in order to comply with the Ordinance\nrelates to incoming requests to complete a booking\ntransaction—content that, while resulting from the third-\nparty listings, is distinct, internal, and nonpublic. As in\nInternet Brands, it is not enough that the third-party listings\nare a “but-for” cause of such internal monitoring. See\n824 F.3d at 853. The text of the CDA is “clear that neither\nthis subsection nor any other declares a general immunity\nfrom liability deriving from third-party content.” Barnes,\n570 F.3d at 1100. To provide broad immunity “every time a\nwebsite uses data initially obtained from third parties would\neviscerate [the CDA].” Barnes, 570 F.3d at 1100 (quoting\nRoommates.com, 521 F.3d at 1171 (9th Cir. 2008) (en\nbanc)). That is not the result that Congress intended.\n\n Nor could a duty to cross-reference bookings against\nSanta Monica’s property registry give rise to CDA\nimmunity. While keeping track of the city’s registry is\n“monitoring” third-party content in the most basic sense,\nsuch conduct cannot be fairly classified as “publication” of\nthird-party content. The Platforms have no editorial control\nover the registry whatsoever. As with tax regulations or\ncriminal statutes, the Ordinance can fairly charge parties\nwith keeping abreast of the law without running afoul of the\nCDA.\n\n Second, the Platforms argue that the Ordinance “in\noperation and effect . . . forces [them] to remove third-party\ncontent.” Although it is clear that the Ordinance does not\nexpressly mandate that they do so, the Platforms claim that\n“common sense explains” that they cannot “leave in place a\nwebsite chock-full of un-bookable listings.” For purposes of\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 15\n\nour review, we accept at face value the Platforms’ assertion\nthat they will choose to remove noncompliant third-party\nlistings on their website as a consequence of the Ordinance. 3\nNonetheless, their choice to remove listings is insufficient to\nimplicate the CDA.\n\n On its face, the Ordinance does not proscribe, mandate,\nor even discuss the content of the listings that the Platforms\ndisplay on their websites. See Santa Monica Mun. Code\n§§ 6.20.010–6.20.100. It requires only that transactions\ninvolve licensed properties. We acknowledge that, as the\nPlatforms explain in Airbnb’s complaint and in the briefing\non appeal, removal of these listings would be the best option\n“from a business standpoint.” But, as in Internet Brands, the\nunderlying duty “could have been satisfied without changes\nto content posted by the website’s users.” See 824 F.3d\nat 851. Even assuming that removing certain listings may be\nthe Platforms’ most practical compliance option, allowing\ninternet companies to claim CDA immunity under these\ncircumstances would risk exempting them from most local\nregulations and would, as this court feared in\nRoommates.com, 521 F.3d at 1164, “create a lawless no-\nman’s-land on the Internet.” We hold that the Ordinance is\nnot “inconsistent” with the CDA, and is therefore not\nexpressly preempted by its terms. See 47 U.S.C. § 230(e)(3).\n\n Finally, the Platforms argue that, even if the Ordinance\nis not expressly preempted by the CDA, the Ordinance\nimposes “an obstacle to the accomplishment and execution\n 3\n The Platforms argued below that the district court must accept as\ntrue their allegation that they would “have to” monitor and screen\nlistings. As a matter of law, the Ordinance does not require them to do\nso. Courts are “not bound to accept as true a legal conclusion couched\nas a factual allegation.” Bell Atlantic Corp. v. Twombly, 550 U.S. 544,\n555 (2007) (quoting Papasan v. Allain, 478 U.S. 265, 286 (1986)).\n\f16 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\nof the full purposes and objectives of Congress.” See Crosby\nv. Nat’l Foreign Trade Council, 530 U.S. 363, 372–73\n(2000). Reading the CDA expansively, they argue that the\nOrdinance conflicts with the CDA’s goal “to preserve the\nvibrant and competitive free market that presently exists for\nthe Internet . . . unfettered by Federal or State regulation.”\nSee § 230(b)(2). We have consistently eschewed an\nexpansive reading of the statute that would render unlawful\nconduct “magically . . . lawful when [conducted] online,”\nand therefore “giv[ing] online businesses an unfair\nadvantage over their real-world counterparts.” See\nRoommates.com, 521 F.3d at 1164, 1164–65 n.15. For the\nsame reasons, while we acknowledge the Platforms’\nconcerns about the difficulties of complying with numerous\nstate and local regulations, the CDA does not provide\ninternet companies with a one-size-fits-all body of law. Like\ntheir brick-and-mortar counterparts, internet companies\nmust also comply with any number of local regulations\nconcerning, for example, employment, tax, or zoning.\nBecause the Ordinance would not pose an obstacle to\nCongress’s aim to encourage self-monitoring of third-party\ncontent, we hold that obstacle preemption does not preclude\nSanta Monica from enforcing the Ordinance.\n\n Fundamentally, the parties dispute how broadly to\nconstrue the CDA so as to continue serving the purposes\nCongress envisioned while allowing state and local\ngovernments breathing room to address the pressing issues\nfaced by their communities. We have previously\nacknowledged that the CDA’s immunity reaches beyond the\ninitial state court decision that sparked its enactment. See\nFair Hous. Council v. Roommates.com, LLC, 521 F.3d 1157,\n1163 (9th Cir. 2008) (en banc) (discussing Stratton\nOakmont, Inc. v. Prodigy Servs. Co., which held an internet\ncompany liable for defamation when it removed some, but\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 17\n\nnot all, harmful content from its public message boards,\n1995 WL 323710 (N.Y. Sup. Ct. May 24, 1995)\n(unpublished)). As the Platforms correctly note, the Act’s\npolicy statements broadly promote “the vibrant and\ncompetitive free market that presently exists for the Internet\n. . . unfettered by Federal or State regulation.” See 47 U.S.C.\n§ 230(b)(2). “[A] law’s scope often differs from its genesis,”\nand we have repeatedly held the scope of immunity to reach\nbeyond defamation cases. Barnes, 570 F.3d at 1101 (quoting\nChicago Lawyers’ Comm. for Civil Rights Under Law, Inc.\nv. Craigslist, Inc., 519 F.3d 666, 671 (7th Cir. 2008), as\namended (May 2, 2008)) (citing cases applying immunity for\ncauses of action including discrimination, fraud, and\nnegligence).\n\n At the same time, our cases have hewn closely to the\nstatutory language of the CDA and have limited the\nexpansion of its immunity beyond the protection Congress\nenvisioned. As we have observed, “the [relevant] section is\ntitled ‘Protection for “good Samaritan” blocking and\nscreening of offensive material.’” Roommates.com,\n521 F.3d at 1163–64 (quoting 47 U.S.C. § 230(c)); see also\nInternet Brands, 824 F.3d at 852. Congress intended to\n“spare interactive computer services [the] grim choice”\nbetween voluntarily filtering content and being subject to\nliability on the one hand, and “ignoring all problematic posts\naltogether [to] escape liability.” Roommates.com, 521 F.3d\nat 1163–64. In contrast, the Platforms face no liability for\nthe content of the bookings; rather, any liability arises only\nfrom unlicensed bookings. We do not discount the\nPlatforms’ concerns about the administrative burdens of\nstate and local regulations, but we nonetheless disagree that\n§ 230(c)(1) of the CDA may be read as broadly as they\nadvocate, or that we may ourselves expand its provisions\nbeyond what Congress initially intended.\n\f18 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\n In sum, neither express preemption nor obstacle\npreemption apply to the Ordinance. We therefore affirm the\ndistrict court’s dismissal for failure to state a claim under the\nCDA.\n\n II. First Amendment\n\n The Platforms also contend that the district court erred in\ndismissing their First Amendment claims. They argue that,\neven if the plain language of the Ordinance only reaches\n“conduct,” i.e., booking unlicensed properties, the law\neffectively imposes a “content-based financial burden” on\ncommercial speech and is thus subject to First Amendment\nscrutiny. The district court concluded that the Ordinance\n“regulates conduct, not speech, and that the conduct banned\n. . . does not have such a ‘significant expressive element’ as\nto draw First Amendment protection.” We agree.\n\n That the Ordinance regulates “conduct” is not alone\ndispositive. The Supreme Court has previously applied First\nAmendment scrutiny when “‘speech’ and ‘nonspeech’\nelements are combined in the same course of conduct.” See\nUnited States v. O’Brien, 391 U.S. 367, 376 (1968). But\n“restrictions on protected expression are distinct from\nrestrictions on economic activity or, more generally, on\nnonexpressive conduct.” Sorrell v. IMS Health Inc.,\n564 U.S. 552, 567 (2011). While the former is entitled to\nprotection, “the First Amendment does not prevent\nrestrictions directed at commerce or conduct from imposing\nincidental burdens on speech.” Id.\n\n To determine whether the First Amendment applies, we\nmust first ask the “threshold question [of] whether conduct\nwith a ‘significant expressive element’ drew the legal\nremedy or the ordinance has the inevitable effect of ‘singling\nout those engaged in expressive activity.’” Int’l Franchise\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 19\n\nAss’n v. City of Seattle, 803 F.3d 389, 408 (9th Cir. 2015)\n(quoting Arcara v. Cloud Books, Inc., 478 U.S. 697, 706–07\n(1986)). A court may consider the “inevitable effect of a\nstatute on its face,” as well as a statute’s “stated purpose.”\nSorrell, 564 U.S. at 565. However, absent narrow\ncircumstances, a court may not conduct an inquiry into\nlegislative purpose or motive beyond what is stated within\nthe statute itself. See O’Brien, 391 U.S. at 383 n.30.\nBecause the conduct at issue—completing booking\ntransactions for unlawful rentals—consists only of\nnonspeech, nonexpressive conduct, we hold that the\nOrdinance does not implicate the First Amendment.\n\n First, the prohibitions here did not target conduct with “a\nsignificant expressive element.” See Arcara, 478 U.S.\nat 706. Our decision in International Franchise Ass’n is\nanalogous. There, the plaintiff challenged a minimum wage\nordinance that would have accelerated the raising of the\nminimum wage to $15 per hour for franchise owners and\nother large employers. 803 F.3d at 389. In denying a\npreliminary injunction, the district court held that the\nplaintiffs were not likely to succeed on their First\nAmendment argument that the ordinance treated them\ndifferently based on their “speech and association” decisions\nto operate within a franchise relationship framework. Id. at\n408–09. We agreed, concluding that the “business\nagreement or business dealings” were not conduct with a\n“significant expressive element.” Id. at 408. Instead,\n“Seattle’s minimum wage ordinance [was] plainly an\neconomic regulation that [did] not target speech or\nexpressive conduct.” Id.\n\n Similarly, here, the Ordinance is plainly a housing and\nrental regulation. The “inevitable effect of the [Ordinance]\non its face” is to regulate nonexpressive conduct—namely,\n\f20 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\nbooking transactions—not speech. See Sorrell, 564 U.S.\nat 565. As in International Franchise Ass’n, the “business\nagreement or business dealings” associated with processing\na booking is not conduct with a “significant expressive\nelement.” See 803 F.3d at 408 (citation and quotation marks\nomitted). Contrary to the Platforms’ claim, the Ordinance\ndoes not “require” that they monitor or screen\nadvertisements. It instead leaves them to decide how best to\ncomply with the prohibition on booking unlawful\ntransactions.\n\n Nor can the Platforms rely on the Ordinance’s “stated\npurpose” to argue that it intends to regulate speech. The\nOrdinance itself makes clear that the City’s “central and\nsignificant goal . . . is preservation of its housing stock and\npreserving the quality and nature of residential\nneighborhoods.” As such, with respect to the Platforms, the\nonly inevitable effect, and the stated purpose, of the\nOrdinance is to prohibit them from completing booking\ntransactions for unlawful rentals.\n\n As for the second prong of our inquiry, whether the\nOrdinance has the effect of “singling out those engaged in\nexpressive activity,” Arcara, 478 U.S. at 706–07, we\nconclude that it does not. As the Platforms point out,\nwebsites like Craigslist “advertise the very same properties,”\nbut do not process transactions. Unlike the Platforms, those\nwebsites would not be subject to the Ordinance,\nunderscoring that the Ordinance does not target websites that\npost listings, but rather companies that engage in unlawful\nbooking transactions.\n\n Moreover, the incidental impacts on speech cited by the\nPlatforms raise minimal concerns. The Platforms argue that\nthe Ordinance chills commercial speech, namely,\nadvertisements for third-party rentals. But even accepting\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 21\n\nthat the Platforms will need to engage in efforts to validate\ntransactions before completing them, incidental burdens like\nthese are not always sufficient to trigger First Amendment\nscrutiny. See Int’l Franchise Ass’n, 803 F.3d at 408\n(“[S]ubjecting every incidental impact on speech to First\nAmendment scrutiny ‘would lead to the absurd result that\nany government action that had some conceivable speech\ninhibiting consequences . . . would require analysis under\nthe First Amendment.’” (quoting Arcara, 478 U.S. at 708\n(O’Connor, J., concurring))). Furthermore, to the extent that\nthe speech chilled advertises unlawful rentals, “[a]ny First\nAmendment interest . . . is altogether absent when the\ncommercial activity itself is illegal and the restriction on\nadvertising is incidental to a valid limitation on economic\nactivity.” See Pittsburgh Press Co. v. Pittsburgh Comm’n\non Human Relations, 413 U.S. 376, 389 (1973).\n\n Finally, because the Ordinance does not implicate speech\nprotected by the First Amendment, we similarly reject the\nPlatforms’ argument that the Ordinance is unconstitutional\nwithout a scienter requirement. In most cases, there is no\n“closed definition” on when a criminal statute must contain\na scienter requirement. See Morissette v. United States,\n342 U.S. 246, 260 (1952). However, the Supreme Court has\ndrawn a bright line in certain contexts, such as holding that\nthe First Amendment requires statutes imposing criminal\nliability for obscenity or child pornography to contain a\nscienter requirement. See New York v. Ferber, 458 U.S. 747,\n765 (1982). Such a requirement prevents “a severe\nlimitation on the public’s access to constitutionally protected\nmatter” as would result from inflexible laws criminalizing\n“bookshops and periodical stands.” Smith v. California,\n361 U.S. 147, 153 (1959).\n\f22 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\n Here, even assuming that the Ordinance would lead the\nPlatforms to voluntarily remove some advertisements for\nlawful rentals, there would not be a “severe limitation on the\npublic’s access” to lawful advertisements, especially\nconsidering the existence of alternative channels like\nCraigslist. Id. Such an incidental burden is far from “a\nsubstantial restriction on the freedom of speech” that would\nnecessitate a scienter requirement. Id. at 150. Otherwise,\n“[t]here is no specific constitutional inhibition against\nmaking the distributors of good[s] the strictest censors of\ntheir merchandise.” Id. at 152.\n\n III. Remaining Claims\n\n On appeal, the Platforms do not challenge dismissal of\ntheir other federal law claims “in light of the district court’s\ninterpretation of the Ordinance as only requiring disclosure\nof information pursuant to requests that comply with the\nFourth Amendment and Stored Communications Act.”\nSimilarly, the parties specified that they would “not\nchallenge the district court’s decision to decline\nsupplemental jurisdiction if all the Platforms’ federal claims\nwere properly dismissed.” Accordingly, we need not\nconsider the remaining claims.\n\n * * *\n\n Because the district court properly dismissed the\nPlatforms’ complaints for failure to state a claim, we dismiss\nas moot the appeals from the denial of preliminary injunctive\nrelief.\n\n AFFIRMED in part, DISMISSED in part.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: homeaway.com, Inc. v. City of Santa Monica\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-13\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nHOMEAWAY.COM, INC.; AIRBNB No. 18-55367\\nINC.,\\n Plaintiffs-Appellants, D.C. Nos.\\n 2:16-cv-06641-\\n v. ODW-AFM\\n 2:16-cv-06645-\\nCITY OF SANTA MONICA, ODW-AFM\\n Defendant-Appellee.\\n\\n\\n\\nHOMEAWAY.COM, INC., No. 18-55805\\n Plaintiff-Appellant,\\n D.C. Nos.\\n and 2:16-cv-06641-\\n ODW-AFM\\nAIRBNB INC., 2:16-cv-06645-\\n Plaintiff, ODW-AFM\\n\\n v.\\n\\nCITY OF SANTA MONICA,\\n Defendant-Appellee.\\n\\f2 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\n\\n AIRBNB INC., No. 18-55806\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:16-cv-06645-\\n ODW-AFM\\n CITY OF SANTA MONICA,\\n Defendant-Appellee. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Otis D. Wright II, District Judge, Presiding\\n\\n Argued and Submitted October 12, 2018\\n Pasadena, California\\n\\n Filed March 13, 2019\\n\\n Before: Mary M. Schroeder and Jacqueline H. Nguyen,\\n Circuit Judges, and Michael H. Simon, * District Judge.\\n\\n Opinion by Judge Nguyen\\n\\n\\n\\n\\n *\\n The Honorable Michael H. Simon, United States District Judge for\\nthe District of Oregon, sitting by designation.\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 3\\n\\n SUMMARY **\\n\\n\\n Civil Rights\\n\\n The panel affirmed the district court’s dismissal, for\\nfailure to state a claim, of a complaint brought by\\nHomeAway.com and Airbnb Inc. challenging the City of\\nSanta Monica’s Ordinance 2535, which imposes various\\nobligations on companies that host online platforms for\\nshort-term vacation rentals.\\n\\n The Ordinance, as amended in 2017, imposes four\\nobligations on hosting platforms: (1) collecting and\\nremitting Transient Occupancy Taxes; (2) regularly\\ndisclosing listings and booking information to the City; (3)\\nrefraining from booking properties not licensed and listed on\\nthe City’s registry; (4) and refraining from collecting a fee\\nfor ancillary services.\\n\\n The panel rejected plaintiffs’ assertion that the\\nOrdinance violated the Communications Decency Act of\\n1996, 47 U.S.C. § 230 because it required them to monitor\\nand remove third-party content, and therefore interfered with\\nfederal policy protecting internet companies from liability\\nfor posting third-party content. The panel stated that the\\nOrdinance prohibits processing transactions for unregistered\\nproperties. It does not require the Platforms to review the\\ncontent provided by the hosts of listings on their websites.\\nRather, the panel noted that the only monitoring that\\nappeared necessary in order to comply with the Ordinance\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f4 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nrelated to incoming requests to complete a booking\\ntransaction—content that, while resulting from the third-\\nparty listings, was distinct, internal, and nonpublic. The\\npanel concluded that the Ordinance was not inconsistent\\nwith the Communications Decency Act, and therefore was\\nnot expressly preempted by its terms. The panel further\\nconcluded that the Ordinance would not pose an obstacle to\\nCongress’s aim to encourage self-monitoring of third-party\\ncontent, and therefore obstacle preemption did not preclude\\nSanta Monica from enforcing the Ordinance.\\n\\n The panel held that the Ordinance did not implicate\\nspeech protected by the First Amendment, concluding that\\nthe Ordinance’s prohibitions regulate nonexpressive\\nconduct, specifically booking transactions, and do not single\\nout those engaged in expressive activity.\\n\\n\\n COUNSEL\\n\\nDonald B. Verrilli, Jr. (argued) and Chad Golder, Munger,\\nTolles & Olson LLP, Washington, D.C.; Joseph W. Cotchett\\nand Alexandra P. Summer, Cotchett, Pitre & McCarthy,\\nLLP, Santa Monica, California; Jonathan H. Blavin and\\nJoshua Patashnik, Munger, Tolles & Olson LLP, San\\nFrancisco, California; John B. Major, Munger, Tolles &\\nOlson, Los Angeles, California; for Plaintiff-Appellant\\nAirbnb, Inc.\\n\\nStephen M. Rummage and Ambika K. Doran, Davis Wright\\nTremaine LLP, Seattle, Washington, for Plaintiff-Appellant\\nHomeAway.com, Inc.\\n\\nGeorge S. Cardona (argued), Deputy City Attorney; Lane\\nDilg, City Attorney; Yibin Shen, Chief Deputy City\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 5\\n\\nAttorney; Heidi Von Tongeln and Michael R. Cobden,\\nDeputy City Attorneys; Santa Monica City Attorney’s\\nOffice, Santa Monica, California, for Defendant-Appellee.\\n\\nDavid Salmons, Bryan Killian, and James Nelson, Morgan,\\nLewis & Bockius LLP, Washington, D.C., for Amici Curiae\\nChris Cox and NetChoice.\\n\\nCatherine R. Gellis, Esq., Sausalito, California, for Amicus\\nCuriae Floor64, Inc. d/b/a The Copia Institute.\\n\\nIan C. Ballon and Lori Chang, Greenberg Traurig, LLP, Los\\nAngeles, California, for Amici Curiae Ebay Inc., Glassdoor,\\nInc., Lyft, Inc., Offerup, Inc., Taskrabbit, Inc., Thumbtack,\\nInc., Uber Technologies, Inc., and Upwork, Inc.\\n\\nChristi Hogin, Best Best & Krieger LLP, Manhattan Beach,\\nCalifornia, for Amici Curiae League of California Cities,\\nInternational Municipal Lawyers Association, and\\nCalifornia State Association of Counties.\\n\\nJeremy B. Rosen, Eric S. Boorstin, and Ryan C. Chapman,\\nHorvitz & Levy LLP, Burbank, California; Michael T.\\nWilliams and Allison R. Mclaughlin, Wheeler Trigg\\nO’Donnell LLP, Denver, Colorado; David C. Frederick,\\nBrendan J. Crimmins, and Rachel P. May, Kellogg, Hansen,\\nTodd, Figel & Frederick, P.L.L.C., Washington, D.C., for\\nAmicus Curiae Apartment Investment and Management\\nCompany.\\n\\nHeidi Palutke, Sacramento, California, for Amicus Curiae\\nCalifornia Apartment Association.\\n\\nEdward M. Schulman, Ballston Tower, Arlington, Virginia,\\nfor Amicus Curiae AvalonBay Communities, Inc.\\n\\f6 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nGary S. Kessler, Kulik Gottesman Siegel & Ware LLP,\\nSherman Oaks, California, for Amicus Curiae Community\\nAssociations Institute.\\n\\nAbbey R. Stemler and Matthew C. Turk, Indiana University\\nKelley School of Business, Bloomington, Indiana; Jahan C.\\nSagafi and Relic Sun, Outten & Golden LLP, San Francisco,\\nCalifornia; Peter Romer-Friedman, Outten & Golden LLP,\\nWashington, D.C., for Amici Curiae Internet, Business, and\\nLocal Government Law Professors.\\n\\nRichard G. McCracken and Paul L. More, McCracken,\\nStemerman & Holsberry, LLP, San Francisco, California,\\nfor Amicus Curiae UNITE HERE International Union.\\n\\nDennis J. Herrera, San Francisco City Attorney; Christine\\nVan Aken, Chief of Appellate Litigation; Yvonne R. Meré,\\nChief of Complex and Affirmative Litigation; Sara J.\\nEisenberg, Deputy City Attorney; San Francisco, California\\nfor Amicus Curiae City and County of San Francisco.\\n\\nKarl A. Racine, Attorney General for the District of\\nColumbia, Washington, D.C., for Amicus Curiae District of\\nColumbia.\\n\\nAndre M. Davis, City Solicitor, City of Baltimore,\\nBaltimore, Maryland, for Amicus Curiae Mayor and City\\nCouncil of Baltimore.\\n\\nZach Klein, Columbus City Attorney, Columbus, Ohio, for\\nAmicus Curiae City of Columbus.\\n\\nBarbara J. Parker, Oakland City Attorney; Maria Bee,\\nSpecial Counsel; Erin Bernstein, Supervising Deputy City\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 7\\n\\nAttorney; Oakland, California; for Amicus Curiae City of\\nOakland.\\n\\nPeter S. Holmes, Seattle City Attorney, Seattle, Washington,\\nfor Amicus Curiae City of Seattle.\\n\\nJill E. Habig, Founder & President; Joanna Pearl, Legal\\nDirector; Oakland, California; for Amicus Curiae Public\\nRights Project, A Project of Tides Center.\\n\\n\\n OPINION\\n\\nNGUYEN, Circuit Judge:\\n\\n Located on the coast of Southern California, the city of\\nSanta Monica consists of only about eight square miles but\\nserves 90,000 residents and as many as 500,000 visitors on\\nweekends and holidays. Similar to other popular tourist\\ndestinations, Santa Monica is struggling to manage the\\ndisruptions brought about by the rise of short-term rentals\\nfacilitated by innovative startups such as Appellants\\nHomeAway.com, Inc. and Airbnb Inc. (the “Platforms”).\\nWebsites like those operated by the Platforms are essentially\\nonline marketplaces that allow “guests” seeking\\naccommodations and “hosts” offering accommodations to\\nconnect and enter into rental agreements with one another. 1\\nAs of February 2018, Airbnb had approximately 1,400\\nlistings in Santa Monica, of which about 30 percent are in\\n\\n 1\\n The Platforms do not own, lease, or manage any of the properties\\nlisted on their websites, nor are they parties to the rental agreements.\\nInstead, the content provided alongside the listings—such as description,\\nprice, and availability—are provided by the hosts. For their services, the\\nPlatforms collect a fee from each successful booking.\\n\\f8 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nthe “coastal zone” covered by the California Coastal Act,\\nwhile HomeAway.com had approximately 300 live listings\\nin Santa Monica, of which approximately 40 percent are in\\nthe coastal zone.\\n\\n Santa Monica’s council reported that the proliferation of\\nshort-term rentals had negatively impacted the quality and\\ncharacter of its neighborhoods by “bringing commercial\\nactivity and removing residential housing stock from the\\nmarket” at a time when California is already suffering from\\nsevere housing shortages. In response, the city passed an\\nordinance regulating the short-term vacation rental market\\nby authorizing licensed “home-sharing” (rentals where\\nresidents remain on-site with guests) but prohibiting all other\\nshort-term home rentals of 30 consecutive days or less.\\n\\n ]The Platforms filed suit, alleging that the city ordinance\\nis preempted by the Communications Decency Act and\\nimpermissibly infringes upon their First Amendment rights.\\nThe district court denied preliminary injunctive relief, and\\ndismissed the Platforms’ complaints for failure to state a\\nclaim under the Communications Decency Act and the First\\nAmendment. We affirm.\\n\\n BACKGROUND\\n\\n In May 2015, Santa Monica passed its initial ordinance\\nregulating the short-term vacation rental market by\\nauthorizing licensed “home-sharing” (rentals where\\nresidents remain on-site with guests) but prohibiting all other\\nforms of short-term rentals for 30 consecutive days or less.\\nSanta Monica Ordinance 2484 (May 12, 2015), codified as\\namended, Santa Monica Mun. Code §§ 6.20.010–6.20.100.\\nThe ordinance reflected the city’s housing goals of\\n“preserving its housing stock and preserving the quality and\\ncharacter of its existing single and multi-family residential\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 9\\n\\nneighborhoods.” Id. As originally enacted, the ordinance\\nprohibited hosting platforms from acting to “undertake,\\nmaintain, authorize, aid, facilitate or advertise any Home-\\nSharing activity” that was not authorized by the city.\\nHosting platforms also were required to collect and remit\\ntaxes, and to regularly disclose listings and booking\\ninformation to the city.\\n\\n The Platforms each filed a complaint in the Central\\nDistrict of California challenging the initial ordinance, and\\nthe district court consolidated the cases for discovery and\\npretrial matters. On September 21, 2016, the parties\\nstipulated to stay the case while the city considered\\namendments to the local ordinance. During the stay period,\\nthe district court for the Northern District of California\\ndenied a preliminary injunction requested by the plaintiffs in\\na separate case challenging a similar ordinance in San\\nFrancisco. See Airbnb Inc. v. City & County of San\\nFrancisco, 217 F. Supp. 3d 1066 (N.D. Cal. 2016). That\\ncase ended in a settlement in which the Platforms agreed to\\ncomply with an amended version of San Francisco’s\\nordinance that prohibited booking unlawful transactions but\\nprovided a safe harbor wherein any platform that complies\\nwith the responsibilities set out in the Ordinance will be\\npresumed to be in compliance with the law.\\n\\n In January 2017, Santa Monica likewise amended its\\nown ordinance. The version challenged here, Ordinance\\n2535 (the “Ordinance”), retains its prohibitions on most\\ntypes of short-term rentals, with the exception of licensed\\nhome-shares. In addition, the Ordinance imposes four\\nobligations on hosting platforms directly: (1) collecting and\\nremitting “Transient Occupancy Taxes,” (2) disclosing\\ncertain listing and booking information regularly,\\n(3) refraining from completing any booking transaction for\\n\\f10 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nproperties not licensed and listed on the City’s registry, and\\n(4) refraining from collecting or receiving a fee for\\n“facilitating or providing services ancillary to a vacation\\nrental or unregistered home-share.” If a housing platform\\noperates in compliance with these obligations, the Ordinance\\nprovides a safe harbor by presuming the platform to be in\\ncompliance with the law. Otherwise, violations are\\npunishable by a fine of up to $500 and/or imprisonment for\\nup to six months.\\n\\n After the district court lifted the stay, the Platforms\\namended their complaint to challenge the revised ordinance\\nand moved for a preliminary injunction. Santa Monica\\nmoved to dismiss the amended complaint. The court denied\\nthe Platforms’ motion for preliminary injunctive relief and\\nsubsequently granted Santa Monica’s motion to dismiss on\\nthe ground that the Platforms failed to state a claim under\\nfederal law, including the Communications Decency Act of\\n1996 and the First Amendment. The district court also\\ndeclined to exercise supplemental jurisdiction over their\\nremaining state-law claims. 2 The Platforms timely appealed\\nthese decisions, and we consolidated the appeals.\\n\\n\\n\\n\\n 2\\n The Platforms do not appeal the district court’s dismissal of other\\nfederal claims brought under the Fourth Amendment and the Stored\\nCommunications Act. Similarly, they do not challenge the court’s\\ndecision not to exercise supplemental jurisdiction over the state-law\\nclaims under the California Coastal Act if we affirm the dismissal of their\\nfederal claims. Because we affirm the district court’s dismissal, we need\\nnot consider the state-law claims. We deny Santa Monica’s motion for\\njudicial notice of its prior enforcement actions because the dispute as to\\nits prior actions relates only to the state-law claims.\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 11\\n\\n JURISDICTION AND STANDARD OF REVIEW\\n\\n We have jurisdiction under 28 U.S.C. § 1291. We\\nreview the district court’s order of dismissal de novo,\\n“accepting all factual allegations in the complaint as true and\\nconstruing them in the light most favorable to the\\nnonmoving party.” Yagman v. Garcetti, 852 F.3d 859, 863\\n(2017) (quoting Ebner v. Fresh, Inc., 838 F.3d 958, 962 (9th\\nCir. 2016)).\\n\\n DISCUSSION\\n\\n I. Communications Decency Act\\n\\n The Communications Decency Act of 1996 (“CDA” or\\nthe “Act”), 47 U.S.C. § 230, provides internet companies\\nwith immunity from certain claims in furtherance of its\\nstated policy “to promote the continued development of the\\nInternet and other interactive computer services.” Id.\\n§ 230(b)(1). Construing this immunity broadly, the\\nPlatforms argue that the Ordinance requires them to monitor\\nand remove third-party content, and therefore violates the\\nCDA by interfering with federal policy protecting internet\\ncompanies from liability for posting third-party content.\\nSanta Monica, on the other hand, argues that the Ordinance\\ndoes not implicate the CDA because it imposes no obligation\\non the Platforms to monitor or edit any listings provided by\\nhosts. Santa Monica contends that the Ordinance is simply\\nan exercise of its right to enact regulations to preserve\\nhousing by curtailing “incentives for landlords to evade rent\\ncontrol laws, evict tenants, and convert residential units into\\nde facto hotels.”\\n\\n We begin our analysis with the text of the CDA. See BP\\nAmerica Prod. Co. v. Burton, 549 U.S. 84, 91 (2006).\\nSection 230(c)(1) states that “[n]o provider or user of an\\n\\f12 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\ninteractive computer service shall be treated as the publisher\\nor speaker of any information provided by another\\ninformation content provider.” Id. § 230(c)(1). The CDA\\nexplicitly preempts inconsistent state laws: “Nothing in this\\nsection shall be construed to prevent any State from\\nenforcing any State law that is consistent with this section.\\nNo cause of action may be brought and no liability may be\\nimposed under any State or local law that is inconsistent with\\nthis section.” Id. § 230(e)(3).\\n\\n We have construed these provisions to extend immunity\\nto “(1) a provider or user of an interactive computer service\\n(2) whom a plaintiff seeks to treat, under a state law cause of\\naction, as a publisher or speaker (3) of information provided\\nby another information content provider.” Barnes v. Yahoo!,\\nInc., 570 F.3d 1096, 1100–01 (9th Cir. 2009). Only the\\nsecond element is at issue here: whether the Ordinance treats\\nthe Platforms as a “publisher or speaker” in a manner that is\\nbarred by the CDA. Although the CDA does not define\\n“publisher,” we have defined “publication” in this context to\\n“involve[] reviewing, editing, and deciding whether to\\npublish or to withdraw from publication third-party\\ncontent.” Id. at 1102 (citing Fair Hous. Council v.\\nRoommates.com, LLC, 521 F.3d 1157, 1170–71 (9th Cir.\\n2008) (en banc)).\\n\\n The Platforms offer two different theories as to how the\\nOrdinance in fact reaches “publication” activities. First, the\\nPlatforms claim that the Ordinance is expressly preempted\\nby the CDA because, as they argue, it implicitly requires\\nthem “to monitor the content of a third-party listing and\\ncompare it against the City’s short-term rental registry\\nbefore allowing any booking to proceed.” Relying on Doe\\nv. Internet Brands, 824 F.3d 846, 851 (9th Cir. 2016), the\\nPlatforms take the view that CDA immunity follows\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 13\\n\\nwhenever a legal duty “affects” how an internet company\\n“monitors” a website.\\n\\n However, the Platforms read Internet Brands too\\nbroadly. In that case, two individuals used the defendant’s\\nwebsite to message and lure the plaintiff to sham auditions\\nwhere she was drugged and raped. Id. at 848. We held that,\\nwhere the website provider was alleged to have known\\nindependently of the ongoing scheme beforehand, the CDA\\ndid not bar an action under state law for failure to warn. Id.\\nat 854. We observed that a duty to warn would not\\n“otherwise affect how [the defendant] publishes or\\nmonitors” user content. Id. at 851. Though the defendant\\ndid, in its business, act as a publisher of third-party content,\\nthe underlying legal duty at issue did not seek to hold the\\ndefendant liable as a “publisher or speaker” of third-party\\ncontent. Id. at 853; see 47 U.S.C. § 230(c)(1). We therefore\\ndeclined to extend CDA immunity to the defendant for the\\nplaintiff’s failure-to-warn claim. Internet Brands, 824 F.3d\\nat 854.\\n\\n We do not read Internet Brands to suggest that CDA\\nimmunity attaches any time a legal duty might lead a\\ncompany to respond with monitoring or other publication\\nactivities. It is not enough that third-party content is\\ninvolved; Internet Brands rejected use of a “but-for” test that\\nwould provide immunity under the CDA solely because a\\ncause of action would not otherwise have accrued but for the\\nthird-party content. Id. at 853. We look instead to what the\\nduty at issue actually requires: specifically, whether the duty\\nwould necessarily require an internet company to monitor\\nthird-party content. See id. at 851, 853.\\n\\n Here, the Ordinance does not require the Platforms to\\nmonitor third-party content and thus falls outside of the\\nCDA’s immunity. The Ordinance prohibits processing\\n\\f14 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\ntransactions for unregistered properties. It does not require\\nthe Platforms to review the content provided by the hosts of\\nlistings on their websites. Rather, the only monitoring that\\nappears necessary in order to comply with the Ordinance\\nrelates to incoming requests to complete a booking\\ntransaction—content that, while resulting from the third-\\nparty listings, is distinct, internal, and nonpublic. As in\\nInternet Brands, it is not enough that the third-party listings\\nare a “but-for” cause of such internal monitoring. See\\n824 F.3d at 853. The text of the CDA is “clear that neither\\nthis subsection nor any other declares a general immunity\\nfrom liability deriving from third-party content.” Barnes,\\n570 F.3d at 1100. To provide broad immunity “every time a\\nwebsite uses data initially obtained from third parties would\\neviscerate [the CDA].” Barnes, 570 F.3d at 1100 (quoting\\nRoommates.com, 521 F.3d at 1171 (9th Cir. 2008) (en\\nbanc)). That is not the result that Congress intended.\\n\\n Nor could a duty to cross-reference bookings against\\nSanta Monica’s property registry give rise to CDA\\nimmunity. While keeping track of the city’s registry is\\n“monitoring” third-party content in the most basic sense,\\nsuch conduct cannot be fairly classified as “publication” of\\nthird-party content. The Platforms have no editorial control\\nover the registry whatsoever. As with tax regulations or\\ncriminal statutes, the Ordinance can fairly charge parties\\nwith keeping abreast of the law without running afoul of the\\nCDA.\\n\\n Second, the Platforms argue that the Ordinance “in\\noperation and effect . . . forces [them] to remove third-party\\ncontent.” Although it is clear that the Ordinance does not\\nexpressly mandate that they do so, the Platforms claim that\\n“common sense explains” that they cannot “leave in place a\\nwebsite chock-full of un-bookable listings.” For purposes of\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 15\\n\\nour review, we accept at face value the Platforms’ assertion\\nthat they will choose to remove noncompliant third-party\\nlistings on their website as a consequence of the Ordinance. 3\\nNonetheless, their choice to remove listings is insufficient to\\nimplicate the CDA.\\n\\n On its face, the Ordinance does not proscribe, mandate,\\nor even discuss the content of the listings that the Platforms\\ndisplay on their websites. See Santa Monica Mun. Code\\n§§ 6.20.010–6.20.100. It requires only that transactions\\ninvolve licensed properties. We acknowledge that, as the\\nPlatforms explain in Airbnb’s complaint and in the briefing\\non appeal, removal of these listings would be the best option\\n“from a business standpoint.” But, as in Internet Brands, the\\nunderlying duty “could have been satisfied without changes\\nto content posted by the website’s users.” See 824 F.3d\\nat 851. Even assuming that removing certain listings may be\\nthe Platforms’ most practical compliance option, allowing\\ninternet companies to claim CDA immunity under these\\ncircumstances would risk exempting them from most local\\nregulations and would, as this court feared in\\nRoommates.com, 521 F.3d at 1164, “create a lawless no-\\nman’s-land on the Internet.” We hold that the Ordinance is\\nnot “inconsistent” with the CDA, and is therefore not\\nexpressly preempted by its terms. See 47 U.S.C. § 230(e)(3).\\n\\n Finally, the Platforms argue that, even if the Ordinance\\nis not expressly preempted by the CDA, the Ordinance\\nimposes “an obstacle to the accomplishment and execution\\n 3\\n The Platforms argued below that the district court must accept as\\ntrue their allegation that they would “have to” monitor and screen\\nlistings. As a matter of law, the Ordinance does not require them to do\\nso. Courts are “not bound to accept as true a legal conclusion couched\\nas a factual allegation.” Bell Atlantic Corp. v. Twombly, 550 U.S. 544,\\n555 (2007) (quoting Papasan v. Allain, 478 U.S. 265, 286 (1986)).\\n\\f16 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nof the full purposes and objectives of Congress.” See Crosby\\nv. Nat’l Foreign Trade Council, 530 U.S. 363, 372–73\\n(2000). Reading the CDA expansively, they argue that the\\nOrdinance conflicts with the CDA’s goal “to preserve the\\nvibrant and competitive free market that presently exists for\\nthe Internet . . . unfettered by Federal or State regulation.”\\nSee § 230(b)(2). We have consistently eschewed an\\nexpansive reading of the statute that would render unlawful\\nconduct “magically . . . lawful when [conducted] online,”\\nand therefore “giv[ing] online businesses an unfair\\nadvantage over their real-world counterparts.” See\\nRoommates.com, 521 F.3d at 1164, 1164–65 n.15. For the\\nsame reasons, while we acknowledge the Platforms’\\nconcerns about the difficulties of complying with numerous\\nstate and local regulations, the CDA does not provide\\ninternet companies with a one-size-fits-all body of law. Like\\ntheir brick-and-mortar counterparts, internet companies\\nmust also comply with any number of local regulations\\nconcerning, for example, employment, tax, or zoning.\\nBecause the Ordinance would not pose an obstacle to\\nCongress’s aim to encourage self-monitoring of third-party\\ncontent, we hold that obstacle preemption does not preclude\\nSanta Monica from enforcing the Ordinance.\\n\\n Fundamentally, the parties dispute how broadly to\\nconstrue the CDA so as to continue serving the purposes\\nCongress envisioned while allowing state and local\\ngovernments breathing room to address the pressing issues\\nfaced by their communities. We have previously\\nacknowledged that the CDA’s immunity reaches beyond the\\ninitial state court decision that sparked its enactment. See\\nFair Hous. Council v. Roommates.com, LLC, 521 F.3d 1157,\\n1163 (9th Cir. 2008) (en banc) (discussing Stratton\\nOakmont, Inc. v. Prodigy Servs. Co., which held an internet\\ncompany liable for defamation when it removed some, but\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 17\\n\\nnot all, harmful content from its public message boards,\\n1995 WL 323710 (N.Y. Sup. Ct. May 24, 1995)\\n(unpublished)). As the Platforms correctly note, the Act’s\\npolicy statements broadly promote “the vibrant and\\ncompetitive free market that presently exists for the Internet\\n. . . unfettered by Federal or State regulation.” See 47 U.S.C.\\n§ 230(b)(2). “[A] law’s scope often differs from its genesis,”\\nand we have repeatedly held the scope of immunity to reach\\nbeyond defamation cases. Barnes, 570 F.3d at 1101 (quoting\\nChicago Lawyers’ Comm. for Civil Rights Under Law, Inc.\\nv. Craigslist, Inc., 519 F.3d 666, 671 (7th Cir. 2008), as\\namended (May 2, 2008)) (citing cases applying immunity for\\ncauses of action including discrimination, fraud, and\\nnegligence).\\n\\n At the same time, our cases have hewn closely to the\\nstatutory language of the CDA and have limited the\\nexpansion of its immunity beyond the protection Congress\\nenvisioned. As we have observed, “the [relevant] section is\\ntitled ‘Protection for “good Samaritan” blocking and\\nscreening of offensive material.’” Roommates.com,\\n521 F.3d at 1163–64 (quoting 47 U.S.C. § 230(c)); see also\\nInternet Brands, 824 F.3d at 852. Congress intended to\\n“spare interactive computer services [the] grim choice”\\nbetween voluntarily filtering content and being subject to\\nliability on the one hand, and “ignoring all problematic posts\\naltogether [to] escape liability.” Roommates.com, 521 F.3d\\nat 1163–64. In contrast, the Platforms face no liability for\\nthe content of the bookings; rather, any liability arises only\\nfrom unlicensed bookings. We do not discount the\\nPlatforms’ concerns about the administrative burdens of\\nstate and local regulations, but we nonetheless disagree that\\n§ 230(c)(1) of the CDA may be read as broadly as they\\nadvocate, or that we may ourselves expand its provisions\\nbeyond what Congress initially intended.\\n\\f18 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\n In sum, neither express preemption nor obstacle\\npreemption apply to the Ordinance. We therefore affirm the\\ndistrict court’s dismissal for failure to state a claim under the\\nCDA.\\n\\n II. First Amendment\\n\\n The Platforms also contend that the district court erred in\\ndismissing their First Amendment claims. They argue that,\\neven if the plain language of the Ordinance only reaches\\n“conduct,” i.e., booking unlicensed properties, the law\\neffectively imposes a “content-based financial burden” on\\ncommercial speech and is thus subject to First Amendment\\nscrutiny. The district court concluded that the Ordinance\\n“regulates conduct, not speech, and that the conduct banned\\n. . . does not have such a ‘significant expressive element’ as\\nto draw First Amendment protection.” We agree.\\n\\n That the Ordinance regulates “conduct” is not alone\\ndispositive. The Supreme Court has previously applied First\\nAmendment scrutiny when “‘speech’ and ‘nonspeech’\\nelements are combined in the same course of conduct.” See\\nUnited States v. O’Brien, 391 U.S. 367, 376 (1968). But\\n“restrictions on protected expression are distinct from\\nrestrictions on economic activity or, more generally, on\\nnonexpressive conduct.” Sorrell v. IMS Health Inc.,\\n564 U.S. 552, 567 (2011). While the former is entitled to\\nprotection, “the First Amendment does not prevent\\nrestrictions directed at commerce or conduct from imposing\\nincidental burdens on speech.” Id.\\n\\n To determine whether the First Amendment applies, we\\nmust first ask the “threshold question [of] whether conduct\\nwith a ‘significant expressive element’ drew the legal\\nremedy or the ordinance has the inevitable effect of ‘singling\\nout those engaged in expressive activity.’” Int’l Franchise\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 19\\n\\nAss’n v. City of Seattle, 803 F.3d 389, 408 (9th Cir. 2015)\\n(quoting Arcara v. Cloud Books, Inc., 478 U.S. 697, 706–07\\n(1986)). A court may consider the “inevitable effect of a\\nstatute on its face,” as well as a statute’s “stated purpose.”\\nSorrell, 564 U.S. at 565. However, absent narrow\\ncircumstances, a court may not conduct an inquiry into\\nlegislative purpose or motive beyond what is stated within\\nthe statute itself. See O’Brien, 391 U.S. at 383 n.30.\\nBecause the conduct at issue—completing booking\\ntransactions for unlawful rentals—consists only of\\nnonspeech, nonexpressive conduct, we hold that the\\nOrdinance does not implicate the First Amendment.\\n\\n First, the prohibitions here did not target conduct with “a\\nsignificant expressive element.” See Arcara, 478 U.S.\\nat 706. Our decision in International Franchise Ass’n is\\nanalogous. There, the plaintiff challenged a minimum wage\\nordinance that would have accelerated the raising of the\\nminimum wage to $15 per hour for franchise owners and\\nother large employers. 803 F.3d at 389. In denying a\\npreliminary injunction, the district court held that the\\nplaintiffs were not likely to succeed on their First\\nAmendment argument that the ordinance treated them\\ndifferently based on their “speech and association” decisions\\nto operate within a franchise relationship framework. Id. at\\n408–09. We agreed, concluding that the “business\\nagreement or business dealings” were not conduct with a\\n“significant expressive element.” Id. at 408. Instead,\\n“Seattle’s minimum wage ordinance [was] plainly an\\neconomic regulation that [did] not target speech or\\nexpressive conduct.” Id.\\n\\n Similarly, here, the Ordinance is plainly a housing and\\nrental regulation. The “inevitable effect of the [Ordinance]\\non its face” is to regulate nonexpressive conduct—namely,\\n\\f20 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nbooking transactions—not speech. See Sorrell, 564 U.S.\\nat 565. As in International Franchise Ass’n, the “business\\nagreement or business dealings” associated with processing\\na booking is not conduct with a “significant expressive\\nelement.” See 803 F.3d at 408 (citation and quotation marks\\nomitted). Contrary to the Platforms’ claim, the Ordinance\\ndoes not “require” that they monitor or screen\\nadvertisements. It instead leaves them to decide how best to\\ncomply with the prohibition on booking unlawful\\ntransactions.\\n\\n Nor can the Platforms rely on the Ordinance’s “stated\\npurpose” to argue that it intends to regulate speech. The\\nOrdinance itself makes clear that the City’s “central and\\nsignificant goal . . . is preservation of its housing stock and\\npreserving the quality and nature of residential\\nneighborhoods.” As such, with respect to the Platforms, the\\nonly inevitable effect, and the stated purpose, of the\\nOrdinance is to prohibit them from completing booking\\ntransactions for unlawful rentals.\\n\\n As for the second prong of our inquiry, whether the\\nOrdinance has the effect of “singling out those engaged in\\nexpressive activity,” Arcara, 478 U.S. at 706–07, we\\nconclude that it does not. As the Platforms point out,\\nwebsites like Craigslist “advertise the very same properties,”\\nbut do not process transactions. Unlike the Platforms, those\\nwebsites would not be subject to the Ordinance,\\nunderscoring that the Ordinance does not target websites that\\npost listings, but rather companies that engage in unlawful\\nbooking transactions.\\n\\n Moreover, the incidental impacts on speech cited by the\\nPlatforms raise minimal concerns. The Platforms argue that\\nthe Ordinance chills commercial speech, namely,\\nadvertisements for third-party rentals. But even accepting\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 21\\n\\nthat the Platforms will need to engage in efforts to validate\\ntransactions before completing them, incidental burdens like\\nthese are not always sufficient to trigger First Amendment\\nscrutiny. See Int’l Franchise Ass’n, 803 F.3d at 408\\n(“[S]ubjecting every incidental impact on speech to First\\nAmendment scrutiny ‘would lead to the absurd result that\\nany government action that had some conceivable speech\\ninhibiting consequences . . . would require analysis under\\nthe First Amendment.’” (quoting Arcara, 478 U.S. at 708\\n(O’Connor, J., concurring))). Furthermore, to the extent that\\nthe speech chilled advertises unlawful rentals, “[a]ny First\\nAmendment interest . . . is altogether absent when the\\ncommercial activity itself is illegal and the restriction on\\nadvertising is incidental to a valid limitation on economic\\nactivity.” See Pittsburgh Press Co. v. Pittsburgh Comm’n\\non Human Relations, 413 U.S. 376, 389 (1973).\\n\\n Finally, because the Ordinance does not implicate speech\\nprotected by the First Amendment, we similarly reject the\\nPlatforms’ argument that the Ordinance is unconstitutional\\nwithout a scienter requirement. In most cases, there is no\\n“closed definition” on when a criminal statute must contain\\na scienter requirement. See Morissette v. United States,\\n342 U.S. 246, 260 (1952). However, the Supreme Court has\\ndrawn a bright line in certain contexts, such as holding that\\nthe First Amendment requires statutes imposing criminal\\nliability for obscenity or child pornography to contain a\\nscienter requirement. See New York v. Ferber, 458 U.S. 747,\\n765 (1982). Such a requirement prevents “a severe\\nlimitation on the public’s access to constitutionally protected\\nmatter” as would result from inflexible laws criminalizing\\n“bookshops and periodical stands.” Smith v. California,\\n361 U.S. 147, 153 (1959).\\n\\f22 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\n Here, even assuming that the Ordinance would lead the\\nPlatforms to voluntarily remove some advertisements for\\nlawful rentals, there would not be a “severe limitation on the\\npublic’s access” to lawful advertisements, especially\\nconsidering the existence of alternative channels like\\nCraigslist. Id. Such an incidental burden is far from “a\\nsubstantial restriction on the freedom of speech” that would\\nnecessitate a scienter requirement. Id. at 150. Otherwise,\\n“[t]here is no specific constitutional inhibition against\\nmaking the distributors of good[s] the strictest censors of\\ntheir merchandise.” Id. at 152.\\n\\n III. Remaining Claims\\n\\n On appeal, the Platforms do not challenge dismissal of\\ntheir other federal law claims “in light of the district court’s\\ninterpretation of the Ordinance as only requiring disclosure\\nof information pursuant to requests that comply with the\\nFourth Amendment and Stored Communications Act.”\\nSimilarly, the parties specified that they would “not\\nchallenge the district court’s decision to decline\\nsupplemental jurisdiction if all the Platforms’ federal claims\\nwere properly dismissed.” Accordingly, we need not\\nconsider the remaining claims.\\n\\n * * *\\n\\n Because the district court properly dismissed the\\nPlatforms’ complaints for failure to state a claim, we dismiss\\nas moot the appeals from the denial of preliminary injunctive\\nrelief.\\n\\n AFFIRMED in part, DISMISSED in part.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1283,"cells":{"id":{"kind":"number","value":4599149,"string":"4,599,149"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Wildearth Guardians v. Heather Provencio"},"decision_date":{"kind":"string","value":"2019-03-13"},"docket_number":{"kind":"string","value":"17-17373"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: MICHAEL DALY HAWKINS, MILAN D. SMITH, JR., and ANDREW D. HURWITZ, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"M. SMITH, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nWILDEARTH GUARDIANS; GRAND No. 17-17373\\nCANYON WILDLANDS COUNCIL;\\nWILDLANDS NETWORK; SIERRA D.C. No.\\nCLUB, 3:16-cv-08010-\\n Plaintiffs-Appellants, SMM\\n\\n v.\\n OPINION\\nHEATHER PROVENCIO, in her official\\ncapacity as Kaibab National Forest\\nSupervisor; UNITED STATES FOREST\\nSERVICE,\\n Defendants-Appellees,\\n\\n and\\n\\nSTATE OF ARIZONA, on behalf of\\nArizona Department of Game and\\nFish; SAFARI CLUB INTERNATIONAL,\\n Intervenors-Defendants-Appellees.\\n\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Stephen M. McNamee, Senior District Judge, Presiding\\n\\n Argued and Submitted February 7, 2019\\n Phoenix, Arizona\\n\\f2 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n Filed March 13, 2019\\n\\n Before: MICHAEL DALY HAWKINS, MILAN D.\\n SMITH, JR., and ANDREW D. HURWITZ,\\n Circuit Judges.\\n\\n Opinion by Judge Milan D. Smith, Jr.\\n\\n\\n SUMMARY *\\n\\n\\n Environmental Law\\n\\n The panel affirmed the district court’s summary\\njudgment in favor of the United States Forest Service in an\\naction by plaintiff environmental groups challenging travel\\nmanagement plans implemented by the Forest Service to\\npermit limited motorized big game retrieval in three Ranger\\nDistricts of the Kaibab National Forest.\\n\\n The Travel Management Rule, promulgated by the U.S.\\nDepartment of Agriculture for Forest Service lands,\\ngenerally prohibits off-road, motorized travel, but permits\\nthe “limited” use of motor vehicles within a specified\\ndistance of “certain” forest roads for the purposes of\\ncamping or retrieval of downed big game animals. The\\npanel rejected plaintiffs’ contention that the Forest Service\\nviolated the Travel Management Rule by implementing\\nplans that did not sufficiently limit motorized big game\\nretrieval in the Ranger Districts. The panel concluded that\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 3\\n\\nthe Forest Service did not violate the plain terms of the\\nTravel Management Rule.\\n\\n Addressing plaintiffs’ claims under the National\\nEnvironmental Policy Act (“NEPA”), the panel held that the\\nplaintiffs had standing to bring their NEPA claims because\\nthey were trying to protect the environment, which was\\nwithin NEPA’s zone of interests. The panel concluded that\\nthe environmental impacts discussed in the environmental\\nassessments did not raise substantial concerns that\\nnecessitated the preparation of environmental impact\\nstatements. The panel held that there was no indication that\\nthe agency failed to satisfy NEPA’s procedural\\nrequirements. The panel concluded that the Forest Service\\ngave the requisite hard look and made determinations that\\nwere neither arbitrary nor capricious, and were consistent\\nwith the evidence before it; and accordingly, the Forest\\nService did not violate NEPA.\\n\\n The panel held that the Forest Service conducted the\\nrequired prefield work, consulted with the appropriate\\nentities, and reached a determination with the evidence\\nbefore it, and satisfied its procedural obligations under the\\nNational Historic Preservation Act.\\n\\f4 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n COUNSEL\\n\\nJohn R. Mellgren (argued), and Susan Jane Brown, Western\\nEnvironmental Law Center, Eugene, Oregon, for Plaintiffs-\\nAppellants.\\n\\nStuart Wilcox, WildEarth Guardians, Denver, Colorado, for\\nPlaintiff-Appellant WildEarth Guardians.\\n\\nMark R. Haag (argued), Allen M. Brabender, Attorneys,\\nEnvironment & Natural Resources Division; Eric Grant,\\nDeputy Assistant Attorney General, Jeffrey H. Wood,\\nActing Assistant Attorney General, United States\\nDepartment of Justice, Washington, D.C.; M’Leah\\nWoodard, Attorney Advisor, Albuquerque Field Office,\\nOffice of the General Counsel, United States Department of\\nAgriculture, Albuquerque, New Mexico; for Defendants-\\nAppellees.\\n\\nDominic Draye (argued), John LeSueur, Assistant Attorney\\nGeneral, and Mark Brnovich, Attorney General, Arizona\\nAttorney General’s Office, Phoenix, Arizona, for\\nIntervenor-Defendant-Appellee State of Arizona.\\n\\nAnna M. Seidman, Douglas S. Burdin, and Jeremy E. Clare,\\nSafari Club International, Washington, D.C., for Intervenor-\\nDefendant-Appellee Safari Club International.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 5\\n\\n OPINION\\n\\nM. SMITH, Circuit Judge:\\n\\n Plaintiffs-Appellants (Plaintiffs) are environmental\\nadvocacy groups that challenged travel management plans\\nimplemented by Defendant-Appellee United States Forest\\nService (the Forest Service) to permit limited motorized big\\ngame retrieval in three Ranger Districts of the Kaibab\\nNational Forest. The district court granted the Forest\\nService’s motion for summary judgment, concluding that the\\nForest Service complied with the Travel Management Rule,\\nthe National Environmental Policy Act (NEPA), and the\\nNational Historic Preservation Act (NHPA). We affirm.\\n\\n FACTUAL AND PROCEDURAL BACKGROUND\\n\\nI. Factual Background\\n\\n A. The Kaibab National Forest\\n\\n The Kaibab National Forest encompasses approximately\\n1.6 million acres of public land in northern Arizona,\\nincluding Grand Canyon National Park. It is comprised of\\nthree noncontiguous Ranger Districts: the Williams Ranger\\nDistrict, the Tusayan Ranger District, and the North Kaibab\\nRanger District.\\n\\n The Williams Ranger District is the southernmost,\\ncovering 560,305 acres approximately thirty-five miles west\\nof Flagstaff and sixty miles south of Grand Canyon National\\nPark. It includes the Kendrick Mountain Wilderness, which\\nextends into Coconino National Forest and features a diverse\\narray of vegetation including Douglas firs, white firs,\\nponderosa pines, and aspens. The Williams Ranger District\\nalso serves as a habitat for a number of endangered species,\\n\\f6 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nincluding the Mexican spotted owl, the California condor,\\nand the black-footed ferret. It contains six areas where\\nspotted owls are known to live and breed, and three spotted\\nowl critical habitats overlap the District.\\n\\n The Tusayan Ranger District, located just south of Grand\\nCanyon National Park’s south rim, encompasses 331,427\\nacres. It features varied terrain, from ponderosa pine forests\\nto grasslands, and is home to a number of sensitive species,\\nincluding bald eagles, goshawks, peregrine falcons,\\nburrowing owls, bats, and voles.\\n\\n The North Kaibab Ranger District covers 655,078 acres\\nimmediately north of Grand Canyon National Park. Like the\\nWilliams and Tusayan Ranger Districts, the North Kaibab\\nRanger District boasts diverse terrain and vegetation, as well\\nas sensitive animal species. Two federally listed endangered\\nspecies—the Mexican spotted owl and California condor—\\nlive in the District, which the U.S. Fish and Wildlife Service\\nhas designated as critical habitat for the spotted owl.\\n\\n B. The Travel Management Rule\\n\\n In 2005, the U.S. Department of Agriculture\\npromulgated a Travel Management Rule to “provide[] for a\\nsystem of National Forest System roads, National Forest\\nSystem trails, and areas on National Forest System lands that\\nare designated for motor vehicle use.” 36 C.F.R.\\n§ 212.50(a). 1 As part of this system, “[d]esignated roads,\\ntrails, and areas [are] identified on a motor vehicle map,”\\nwhich also “specif[ies] the classes of vehicles” and “the\\n 1\\n The Rule’s antecedents include executive orders issued by\\nPresidents Nixon and Carter that sought to limit the damage to federal\\npublic lands caused by off-road vehicles. See Utah Shared Access All.\\nv. Carpenter, 463 F.3d 1125, 1129–30 (10th Cir. 2006).\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 7\\n\\ntimes of year for which use is designated.” Id. § 212.56.\\nMotor vehicle use is prohibited on roads not so designated.\\nId. § 212.50(a). The regulations include a specific provision\\nconcerning the use of motor vehicles for dispersed camping\\nand big game retrieval, which allows that\\n\\n in designating routes, the responsible official\\n may include in the designation the limited use\\n of motor vehicles within a specified distance\\n of certain forest roads or trails where motor\\n vehicle use is allowed, and if appropriate\\n within specified time periods, solely for the\\n purposes of dispersed camping or retrieval of\\n a downed big game animal by an individual\\n who has legally taken that animal.\\n\\nId. § 212.51(b).\\n\\n The Forest Service’s Southwestern Regional Office\\nissued guidelines for implementation of the Travel\\nManagement Rule, including its motorized big game\\nretrieval provision. The guidelines noted that “National\\nForests in the Southwestern Region provide hunting\\nopportunities that are important to the public,” and directed\\nforests to identify designated routes for game retrieval “in\\nclose collaboration with the responsible State agency.” They\\nalso suggested, pursuant to discussions with the Arizona\\nGame and Fish Department, that motorized big game\\nretrieval be allowed “up to three miles from a designated\\nroute” for bison and “up to one mile from a designated route”\\nfor elk and mule deer.\\n\\f8 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n The Forest Service crafted travel management plans for\\neach of the three Ranger Districts in the Kaibab National\\nForest. It also prepared an Environmental Assessment (EA)\\nfor each plan to ascertain its environmental impact, but did\\nnot undertake a more rigorous Environmental Impact\\nStatement (EIS).\\n\\n i. The Williams Ranger District\\n\\n In July 2010, the Forest Service released the EA for the\\nWilliams Ranger District’s travel management plan, and\\nsubsequently issued a Decision Notice and Finding of No\\nSignificant Impact (DN/FONSI). The DN/FONSI generally\\n“prohibit[s] motorized travel off of designated routes on the\\nWilliams Ranger District,” but permits “the limited use of\\nmotor vehicles within one mile of all designated system\\nroads (except where prohibited) to retrieve a legally hunted\\nand tagged elk during all elk hunting seasons.” It allows\\nmotorized big game retrieval of elk (but not bison) up to one\\nmile off all designated open roads, so long as hunters make\\nonly “[o]ne trip that uses [the] most direct route and least\\nground disturbing.” The designated open road system\\nconsists of 1,114 miles of roadway, a reduction from\\nprevious motor vehicle activity, when 1,460 miles of roads\\nand 95 percent of the District were open to motor vehicle\\nuse. Several miles of the open roads pass through the spotted\\nowl critical habitat.\\n\\n ii. The Tusayan Ranger District\\n\\n Previously, the Tusayan Ranger District contained more\\nthan 700 miles of roads open to motor vehicles, and a vast\\nmajority of the District was open to cross-country motor\\nvehicle travel. The Forest Service’s final DN/FONSI for the\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 9\\n\\nDistrict 2 designated 566 miles of road open to motor\\nvehicles. The decision permits “[l]egally harvested elk [to]\\nbe retrieved during all legal elk hunting seasons” by motor\\nvehicles within one mile of designated roads. Motorized\\nretrieval of bison is not permitted, and the DN/FONSI limits\\nuse of motor vehicles when “conditions are such that travel\\nwould cause damage to natural and/or cultural resources,”\\nand mandated that “[m]otorized vehicles would not be\\npermitted to cross riparian areas, streams and rivers except\\nat hardened crossings or crossings with existing culverts.”\\n\\n iii. The North Kaibab Ranger District\\n\\n Prior to implementation of a new travel management\\nplan, 1,852 miles of road in the North Kaibab Ranger District\\nwere open to motor vehicle use, with 83 percent of the\\nDistrict open to cross-county travel. In September 2012, the\\nForest Service released an EA analyzing the District’s new\\nplan. Among other data, the EA noted that while “[c]ross-\\ncountry motorized travel, whether to retrieve game or for\\nother purposes, can adversely affect cultural resource sites if\\na vehicle is driven across a site,” only thirty-eight bison and\\nno elk were taken from the District in 2009.\\n\\n The Forest Service issued a DN/FONSI that designated\\n1,476 miles of open roads for motorized travel, including an\\nadditional 16 miles of unauthorized, user-created roads.\\nMotor vehicles can be used to retrieve elk or bison during\\nhunting seasons, under certain limiting conditions. Notably,\\n\\n 2\\n In April 2009, the Forest Service issued an initial EA that analyzed\\nthe impact of the new travel management plan, as well as a subsequent\\nDN/FONSI. In response to administrative appeals, that decision was\\nreversed, and a new environmental survey undertaken. The resulting\\nEA, issued in January 2011, reflected additional analysis and public\\ncomment.\\n\\f10 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nthe plan prohibits motorized retrieval of mule deer; the data\\nindicated that far more mule deer—1,020—were harvested\\nin the District in 2009 than bison or elk. The DN/FONSI\\nalso included guidance for monitoring and mitigation, as\\nwell as practices to limit the spread of invasive exotic weeds.\\n\\nII. Procedural Background\\n\\n The Districts’ travel management plans—their\\nmotorized big game retrieval provisions in particular—were\\nadministratively appealed, and the Regional Forester upheld\\nthem.\\n\\n Plaintiffs then filed a complaint for declaratory and\\ninjunctive relief in the district court. They challenged the\\ntravel management plans for each of the three Ranger\\nDistricts, alleging violations of the Travel Management\\nRule, the Administrative Procedure Act (APA), NEPA, and\\nthe NHPA. After the parties filed and briefed cross-motions\\nfor summary judgment, the district court granted the Forest\\nService’s motion and denied Plaintiffs’ motion. It further\\ndenied the motions for summary judgment filed by\\nIntervenors-Defendants-Appellees State of Arizona (the\\nState) and Safari Club International (Safari Club) as moot.\\nThis timely appeal followed.\\n\\n STANDARD OF REVIEW AND JURISDICTION\\n\\n We review de novo a district court’s order granting or\\ndenying a motion for summary judgment. Churchill County\\nv. Norton, 276 F.3d 1060, 1071 (9th Cir. 2001). Under the\\nAPA, agency action can be set aside if it is “arbitrary,\\ncapricious, an abuse of discretion, or otherwise not in\\naccordance with law.” 5 U.S.C. § 706(2)(A); see also\\nWildEarth Guardians v. Mont. Snowmobile Ass’n, 790 F.3d\\n920, 932 (9th Cir. 2015) (Travel Management Rule reviewed\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 11\\n\\nunder the APA); San Carlos Apache Tribe v. United States,\\n417 F.3d 1091, 1099 (9th Cir. 2005) (NHPA challenge\\nreviewed under the APA); Churchill County, 276 F.3d at\\n1071 (NEPA challenge reviewed under the APA).\\n\\n We have jurisdiction pursuant to 28 U.S.C. § 1291.\\n\\n ANALYSIS\\n\\nI. The Travel Management Rule\\n\\n Plaintiffs argue that the Forest Service violated the\\nTravel Management Rule by implementing plans that did not\\nsufficiently limit motorized big game retrieval in the\\nDistricts.\\n\\n Although the Travel Management Rule generally\\nprohibits off-road, motorized travel, it permits officials to\\ndesignate “the limited use of motor vehicles within a\\nspecified distance of certain forest roads . . . solely for the\\npurposes of dispersed camping or retrieval of a downed big\\ngame animal.” 36 C.F.R. § 212.51(b) (emphases added). As\\npart of this designation process, the Forest Service must\\nconsider various criteria, ranging from “public safety” and\\n“conflicts among uses of National Forest System lands” to\\n“[d]amage to soil, watershed, vegetation, and other forest\\nresources” and “[h]arassment of wildlife and significant\\ndisruption of wildlife habitats.” 36 C.F.R. § 212.55(a)–(b).\\nCompliance with these “minimization criteria” is\\nmandatory. Mont. Snowmobile, 790 F.3d at 929–32.\\n\\n Plaintiffs contend that the Forest Service violated the\\nTravel Management Rule by permitting off-road motorized\\nvehicle use to collect downed game within one mile of every\\nopen road in the Districts, in purported violation of the\\n\\f12 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nRule’s mandate that such activity be “limited” and only on\\n“certain” roads. We disagree.\\n\\n A. “Limited”\\n\\n First, Plaintiffs argue that “designating cross-country\\noff-road motor vehicle use for one mile off both sides of\\nevery single open road on each Ranger District is not a\\n‘limited’ designation as contemplated by the Travel\\nManagement Rule.” They note that the Forest Service’s own\\nregional office acknowledged as much when reviewing the\\nNorth Kaibab Ranger District’s plan proposal; the office\\ncommented, “Motorized [big game retrieval] is being\\nproposed on all system routes which is not consistent with\\nrule for ‘limited use.’” The assistant NEPA coordinator of\\nthat District also questioned the scope of the plan, noting,\\n“What I don’t see is how this got you to your proposed\\nactions, particularly the [motorized big game retrieval] on\\nALL roads with 1 mile corridor . . . . that’s huge.” In their\\nbrief, Plaintiffs include maps prepared by the Forest Service,\\nwhich illustrate the extent of areas (in light blue) where\\nmotorized big game retrieval is allowed in the Tusayan and\\nWilliams Ranger Districts. These maps indicate, as\\nPlaintiffs argue, that the plans permit motorized big game\\nretrieval across a vast swath of the Districts’ lands:\\n\\fWILDEARTH GUARDIANS V. PROVENCIO 13\\n\\f14 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n As the district court correctly noted, however, Plaintiffs’\\nfocus on only the spatial limitation of the Districts’ plans\\nignores the other restrictions on motorized big game\\nretrieval. The North Kaibab Ranger District plan illustrates\\nsome of these additional limitations: restricting retrieval to\\nlegally hunted elk and bison 3; permitting only one vehicle\\nper harvested animal; requiring hunters to “use the most\\ndirect and least ground disturbing route in and out of the area\\nto accomplish the retrieval”; and limiting the temporal\\nperiod to the “appropriate season as designated by the\\n[State], and for 24 hours following each season.” The\\nWilliams and Tusayan Ranger Districts further exclude the\\nretrieval of bison and require motor vehicles to cross streams\\nand rivers only at designated crossings. We agree with the\\ndistrict court that “these limitations are a significant\\ndeparture from the previous policy which did not limit the\\nnumber of trips . . . , did not limit the type of species which\\ncould be retrieved by motor vehicle, did not limit the\\ndistance traveled from system roads, and had no restrictions\\non seasons or weather conditions and no requirement for use\\nof a direct route.”\\n\\n Plaintiffs assert that “the nearly unlimited spatial\\nallowance . . . in and of itself[] violates the plain terms of the\\nTravel Management Rule,” but provide no authority for that\\nproposition. Because we find no support for it in either case\\nlaw or the applicable regulations, we conclude that the Forest\\nService did not abuse its discretion when it authorized plans\\nthat limit motorized big game retrieval based on factors other\\nthan geography. See 5 U.S.C. § 706(2)(A).\\n\\n\\n 3\\n The State notes that additional, non-qualifying big game species\\ncan also be found in the Kaibab National Forest, including mule deer,\\npronghorn, and black bears.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 15\\n\\n As for the regional office’s skepticism regarding the\\nscope of retrieval permitted under the plans, “the fact that a\\npreliminary determination by a local agency representative\\nis later overruled at a higher level within the agency does not\\nrender the decisionmaking process arbitrary and capricious,”\\nas agencies are “fully entitled” to “change[] their minds . . .\\nas long as the proper procedures were followed.” Nat’l\\nAss’n of Home Builders v. Defs. of Wildlife, 551 U.S. 644,\\n659 (2007). Such apparent inconsistencies might serve as\\nevidence of arbitrariness or capriciousness, see Barnes v.\\nU.S. Dep’t of Transp., 655 F.3d 1124, 1134 (9th Cir. 2011)\\n(noting that Home Builders “did not hold, however, that such\\npreliminary determinations are irrelevant in any context”),\\nbut absent other evidence that the plans violated the Travel\\nManagement Rule, we do not find these comments\\nparticularly enlightening, especially since the Regional\\nForester ultimately upheld the designation decisions as to\\neach Ranger District. 4\\n\\n Ultimately, as the district court concluded, “Plaintiffs\\nhave only identified dissatisfaction with the ultimate\\ndecisions made by the Forest Service in authorizing\\n[motorized big game retrieval] in the three ranger districts.”\\nAlthough it is not unreasonable to interpret “limited use”\\nspatially, Plaintiffs point to no statute, regulation, or ruling\\nthat requires a geographic limitation of this sort. The\\nDistricts’ plans limit motorized big game retrieval as to\\ntiming, qualified species, and number of vehicles, which is\\n\\n 4\\n The same can be said for any inconsistencies between the final\\nplans and the guidance provided by the Forest Service’s regional officer,\\nto which Plaintiffs point as evidence that the Rule was violated.\\nAlthough, for example, the guidance suggested that retrieval of elk not\\nbe allowed “between one hour before sunrise and 10:00 am,” that\\nlimitation was not a requirement, but merely a recommendation, and so\\na failure to implement it does not render the plans unlawful.\\n\\f16 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nboth a reasonable interpretation of the Rule and an indication\\nthat the Forest Service both considered and applied the\\nmandatory minimization criteria. See Mont. Snowmobile,\\n790 F.3d at 932. Given that much of the Districts’ land was\\nopen to cross-country motorized travel prior to\\nimplementation of the plans, we conclude that the new\\nrestrictions constitute a “limited” use of motorized vehicles.\\n\\n B. “Certain”\\n\\n Next, the Rule allows motorized retrieval on “certain\\nforest roads.” 36 C.F.R. § 212.51(b). Plaintiffs argue that\\nthe word “certain” requires that it only be allowed on “some,\\nbut not all” roads. Because the plans permit retrieval within\\none mile of all designated roads, Plaintiffs conclude that they\\nare unlawful. The Forest Service, however, correctly notes\\nthat while the word “certain” can mean “some, but not all,”\\nthe more common definition of the term is “definite” or\\n“fixed.” See, e.g., Webster’s Third New International\\nDictionary 367 (2002) (listing the primary definitions of\\n“certain” as “fixed, settled, stated” and “exact, precise”).\\n\\n Because the Forest Service limited motor vehicle use to\\na defined set of roads in each District, it complied with the\\nRule. Even if the proper interpretation of the word “certain”\\nwere ambiguous, the Forest Service’s definition of “fixed”\\nor “definite” is permissible, consistent with the text of the\\nRule, and entitled to deference. See Home Builders,\\n551 U.S. at 672 (“An agency’s interpretation of the meaning\\nof its own regulations is entitled to deference ‘unless plainly\\nerroneous or inconsistent with the regulation.’” (quoting\\nAuer v. Robbins, 519 U.S. 452, 461 (1997))).\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 17\\n\\n C. “Sparingly”\\n\\n Lastly, Plaintiffs contend that the Forest Service failed to\\nimplement motorized big game retrieval “sparingly.” Travel\\nManagement; Designated Routes and Areas for Motor\\nVehicle Use, 70 Fed. Reg. 68,264, 68,285 (Nov. 9, 2005)\\n(“The Department expects the Forest Service to apply this\\nprovision sparingly, on a local or State-wide basis, to avoid\\nundermining the purposes of the final rule and to promote\\nconsistency in implementation.” (emphasis added)). They\\nargue that “allowing motorized big game retrieval off of\\nevery single open road is not using [the Forest Service’s]\\nauthority sparingly,” and that the Forest Service erroneously\\nconcluded that the plans were sparing and limited because\\nthey permitted less motorized retrieval than under the prior\\npolicies that imposed no restrictions.\\n\\n This argument is unpersuasive. First, the word\\n“sparingly” does not appear in the Rule, but instead in its\\npreamble. We look to a preamble only when the regulation\\nitself is ambiguous. El Comité para el Bienstar de Earlimart\\nv. Warmerdam, 539 F.3d 1062, 1070 (9th Cir. 2008).\\nAccordingly, as the Forest Service correctly asserts, the\\npreamble does not “impose a duty above and beyond the\\nactual terms of the regulation.”\\n\\n Moreover, Plaintiffs’ argument fails for much the same\\nreason as their “limited” argument above: it relies on a\\nstrictly spatial conception of “sparingly,” while ignoring the\\nother restrictions that the plans impose on motorized big\\ngame retrieval. Absent authority requiring a strictly\\ngeographic interpretation of the words “limited” and\\n“sparingly,” we conclude that the Forest Service did not\\nviolate the plain terms of the Travel Management Rule.\\n\\f18 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nII. NEPA\\n\\n A. Standing\\n\\n As a preliminary matter, Safari Club challenges whether\\nPlaintiffs have standing to bring their NEPA claims. It\\nobserves that Plaintiffs’ “standing declarant, Kim Crumbo,\\nrevealed that his own activities in the forests cause the very\\nsame impact that [they] seek to attribute to motorized big\\ngame retrieval,” since Crumbo recounted “incidents during\\nwhich he, on his bicycle, interfered with young goshawk and\\na goshawk pursuing its prey.” Accordingly, Safari Club\\nsuggests that Plaintiffs cannot satisfy the redressability\\nrequirement for standing, see Salmon Spawning & Recovery\\nAll. v. Gutierrez, 545 F.3d 1220, 1226 (9th Cir. 2008),\\nbecause Plaintiffs “themselves are a source of” the negative\\neffects that they seek to analyze through an EIS, and so “[n]o\\nchange in process and no [EIS] analysis of motorized big\\ngame retrieval will prevent [Plaintiffs’] members from\\nengaging in conduct that is the source of effects” that they\\nseek to mitigate.\\n\\n Under Safari Club’s reasoning, a hypothetical plaintiff\\nchallenging an EPA decision on a CO2-emitting power plant\\nwould lack standing just because she also happens to exhale\\ncarbon dioxide. This result would not only be absurd, but\\nalso contrary to our prior precedent, for we have held that\\n\\n the mere existence of multiple causes of an\\n injury does not defeat redressability,\\n particularly for a procedural injury. So long\\n as a defendant is at least partially causing the\\n alleged injury, a plaintiff may sue that\\n defendant, even if the defendant is just one of\\n multiple causes of the plaintiff’s injury.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 19\\n\\nWildEarth Guardians v. U.S. Dep’t of Agric., 795 F.3d 1148,\\n1157 (9th Cir. 2015).\\n\\n Safari Club also challenges Plaintiffs’ NEPA standing\\nbased on their perceived motivation for bringing suit.\\nAsserting that Plaintiffs’ “goal in bringing this action was to\\nuse NEPA and the NHPA to force [the Forest Service] to\\nreverse [its] authorization of motorized big game retrieval,”\\nit contends that “NEPA does not provide a cause of action”\\nand that Plaintiffs lack prudential standing. Regardless of\\nPlaintiffs’ motivation in commencing this suit, however, if a\\ngroup is “trying to protect the environment,” then its “suit []\\nlies well within NEPA’s zone of interests.” Citizens for\\nBetter Forestry v. U.S. Dep’t of Agric., 341 F.3d 961, 976\\n(9th Cir. 2003); see also Havasupai Tribe v. Provencio,\\n906 F.3d 1155, 1166 (9th Cir. 2018). Accordingly, we\\nconclude that Plaintiffs have standing to bring their NEPA\\nclaims.\\n\\n B. Legal Framework\\n\\n NEPA requires federal agencies, including the Forest\\nService, to assess the environmental impact of proposed\\nactions that “significantly affect[] the quality of the human\\nenvironment.” 42 U.S.C. § 4332(C). It\\n\\n serves two fundamental objectives. First, it\\n “ensures that the agency, in reaching its\\n decision, will have available, and will\\n carefully consider, detailed information\\n concerning significant environmental\\n impacts.” And, second, it requires “that the\\n relevant information will be made available\\n to the larger audience that may also play a\\n role in both the decisionmaking process and\\n the implementation of that decision.”\\n\\f20 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nMont. Snowmobile, 790 F.3d at 924 (citation omitted)\\n(quoting Robertson v. Methow Valley Citizens Council,\\n490 U.S. 332, 349 (1989)). In short, “NEPA’s purpose is to\\nensure that ‘the agency will not act on incomplete\\ninformation, only to regret its decision after it is too late to\\ncorrect.’” Friends of Clearwater v. Dombeck, 222 F.3d 552,\\n557 (9th Cir. 2000) (quoting Marsh v. Or. Nat. Res. Council,\\n490 U.S. 360, 371 (1989)). We have admonished that “[w]e\\nmust . . . strictly interpret the procedural requirements in\\nNEPA . . . ‘to the fullest extent possible’ consistent with the\\npolicies embodied in NEPA. ‘[G]rudging, pro forma\\ncompliance will not do.’” Churchill County, 276 F.3d at\\n1072 (fourth alteration in original) (citation omitted)\\n(quoting Lathan v. Brinegar, 506 F.2d 677, 687, 693 (9th\\nCir. 1974) (en banc)). As part of this compliance, agencies\\nmust ensure “that environmental information is available to\\npublic officials and citizens before decisions are made and\\nbefore actions are taken. The information must be of high\\nquality. Accurate scientific analysis, expert agency\\ncomments, and public scrutiny are essential to implementing\\nNEPA.” 40 C.F.R. § 1500.1(b).\\n\\n The Forest Service must prepare an EIS—a more\\nthorough undertaking than an EA 5—if an action might\\nsignificantly affect environmental quality. As we have\\nexplained,\\n\\n\\n\\n 5\\n “Before deciding whether to complete an EIS, government\\nagencies may prepare a less formal EA which ‘briefly provides sufficient\\nevidence and analysis for determining whether to prepare an\\nenvironmental impact statement or a finding of no significant impact.’”\\nAnderson v. Evans, 371 F.3d 475, 488 (9th Cir. 2004) (quoting Tillamook\\nCounty v. U.S. Army Corps of Eng’rs, 288 F.3d 1140, 1144 (9th Cir.\\n2002)).\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 21\\n\\n An EIS must be prepared if “substantial\\n questions are raised as to whether a project\\n . . . may cause significant degradation of\\n some human environmental factor.” Thus, to\\n prevail on a claim that the Forest Service\\n violated its statutory duty to prepare an EIS,\\n a “plaintiff need not show that significant\\n effects will in fact occur.” It is enough for the\\n plaintiff to raise “substantial questions\\n whether a project may have a significant\\n effect” on the environment.\\n\\nBlue Mountains Biodiversity Project v. Blackwood, 161 F.3d\\n1208, 1212 (9th Cir. 1998) (alteration in original) (citations\\nomitted) (quoting Idaho Sporting Cong. v. Thomas, 137 F.3d\\n1146, 1149–50 (9th Cir. 1998)); see also Save the Yaak\\nComm. v. Block, 840 F.2d 714, 717 (9th Cir. 1988) (“[A]n\\nagency’s decision not to prepare an EIS will be considered\\nunreasonable if the agency fails to ‘supply a convincing\\nstatement of reasons why potential effects are\\ninsignificant.’” (quoting Steamboaters v. FERC, 759 F.2d\\n1382, 1393 (9th Cir. 1985))). The significance of an action\\ndepends on its context and intensity, the latter of which is\\nassessed using a list of criteria enumerated in the relevant\\nregulation. See 40 C.F.R. § 1508.27(a)–(b).\\n\\n C. Whether the Forest Service Needed to Prepare\\n EISs\\n\\n Plaintiffs contend that “the presence of several\\nsignificance factors indicating possible significant\\nenvironmental consequences of the proposed actions”\\nrequired the Forest Service to prepare EISs for each of the\\nDistricts’ travel management plans. We analyze in turn each\\nconsideration to which they point.\\n\\f22 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n i. Impacts\\n\\n The first enumerated consideration in evaluating an\\naction’s intensity concerns “[i]mpacts that may be both\\nbeneficial and adverse,” and notes that “[a] significant effect\\nmay exist even if the Federal agency believes that on balance\\nthe effect will be beneficial.” 40 C.F.R. § 1508.27(b)(1).\\n“Significance exists if it is reasonable to anticipate a\\ncumulatively significant impact on the environment.\\nSignificance cannot be avoided by terming an action\\ntemporary or by breaking it down into small component\\nparts.” Id. § 1508.27(b)(7).\\n\\n Plaintiffs suggest that “the travel management plans for\\neach Ranger District would have significant direct, indirect,\\nand cumulative impacts,” noting that both motorized vehicle\\nuse on open, designated roads and cross-country, off-road\\nmotorized vehicle use can have “significant detrimental\\neffects . . . on a variety of resources.”\\n\\n The assertion that motorized big game retrieval can have\\ndetrimental effects on the environment is consistent with the\\ninformation contained in the EAs prepared for the Williams\\nand Tusayan Ranger Districts. Those EAs noted that “[t]he\\nscientific literature documents a variety of negative effects\\nof roads and motorized travel on wildlife,” with\\n\\n [p]otential direct and indirect effects of roads\\n and motorized travel on wildlife includ[ing]\\n habitat loss, fragmentation, and degradation\\n caused by roads and cross country motorized\\n travel; roads can create barriers to\\n movements of certain species; animals can be\\n killed or injured as a result of being hit or run\\n over by motor vehicles; human disturbance\\n or harassment of animals caused by or\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 23\\n\\n facilitated by motorized travel; [and]\\n shooting or harvest of animals facilitated by\\n motor vehicle access to wildlife habitats.\\n\\nThe Williams Ranger District EA further indicated that off-\\nroad vehicle use “in areas with sensitive or moist soils can\\ncreate tracks, ruts and new user routes that may crush,\\ndisplace, and/or destroy cultural materials (i.e. artifacts,\\nfeatures, traditionally used plants), and damage significant\\ninformation that may contribute to our understanding of\\nhistory.” A particularly vexatious problem related to\\nmotorized vehicle use is the spread of invasive weeds. Each\\nof the three EAs noted that vehicles are a common cause of\\nweed introduction and spread, with the North Kaibab Ranger\\nDistrict EA reporting that “[t]he authorization of motorized\\nbig game retrieval will have an increased threat of invasive\\nspecies spread as every vehicle that travels cross-country has\\nthe ability to serve as a vector and create disturbance.” 6\\n\\n We do not disagree with Plaintiffs’ assertion that\\nmotorized big game retrieval can have a negative effect on\\nthe environment. But we nevertheless conclude that the\\nenvironmental impacts discussed in the EAs did not raise\\nsubstantial concerns that necessitated the preparation of\\nEISs. Plaintiffs might disagree with the Forest Service’s\\nsubstantive conclusions, but we see no indication that the\\nagency failed to satisfy NEPA’s procedural requirements—\\na crucial distinction that is lost in Plaintiffs’ formulation of\\nthe issue.\\n\\n\\n 6\\n This is because, as explained in the Williams Ranger District EA,\\n“[v]ehicles driving through populations of invasive plants often get\\nseed[s] entrapped in tire tread or undercarriages, move to another area\\nand then drop seeds into a previously uninfested area.”\\n\\f24 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n Plaintiffs’ treatment of the spread of invasive weeds is\\nillustrative. They tie the problem of weeds to another\\nenumerated NEPA consideration that implicates “[u]nique\\ncharacteristics of the geographic area such as proximity to\\nhistoric or cultural resources, park lands, prime farmlands,\\nwetlands, wild and scenic rivers, or ecologically critical\\nareas.” Id. § 1508.27(b)(3). They note that “both the North\\nKaibab and Tusayan Ranger Districts immediately abut\\nGrand Canyon National Park.” The EAs agreed: they\\ndescribed Grand Canyon National Park as “internationally\\nimportant,” and noted the Tusayan Ranger District’s “unique\\nlocation” as “a gateway to one of the most famous national\\nparks in the country.” In that District’s EA, the Forest\\nService acknowledged that recreation and hunting—\\nactivities facilitated by the travel management plan—“have\\nthe potential to introduce exotic plants” that “may then\\nspread to adjoining lands, including the Grand Canyon\\nNational Park.” The National Park Service, in a letter from\\nthe Acting Park Superintendent of Grand Canyon National\\nPark, advised the Forest Service to “institute a buffer zone\\nof 1-mile along the park boundary for any purpose including\\nbig-game retrieval, fuel-wood gathering, cross-county\\ntravel, etc.,” due to “increased pressure from motorized\\nvehicles at or near the southern park boundary over the past\\nseveral years”—a recommendation that was not adopted in\\nthe Tusayan Ranger District DN/FONSI.\\n\\n Clearly, the EAs demonstrated that motorized big game\\nretrieval risks the spread of invasive weeds, an undeniable\\nenvironmental impact. In response, the Forest Service relies\\nin part on questionable reasoning by focusing on the fact that\\nthe plans reduced negative impacts when compared with\\npre-plan activity. It notes that the plan eventually selected\\nfor the Tusayan Ranger District opted to limit the number of\\nroads open to the public, which, the EA noted, “reduces the\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 25\\n\\nnumber of opportunities for noxious and invasive exotic\\nweeds to be introduced and spread.” “Thus,” the Forest\\nService concludes, “contrary to [Plaintiffs’] argument, the\\nTusayan EA confirms that the Forest Service considered the\\nissue and reasonably concluded that the decision will reduce,\\nnot increase, the spread of exotic plants.”\\n\\n However, a conclusion, even a correct one, that a given\\naction might reduce a potential impact does not alone\\nindicate that the impact would not be significant. We have\\nnoted that the use of baselines is a helpful, and perhaps\\ninevitable, tool in conducting environmental surveys. See\\nOr. Nat. Desert Ass’n v. Jewell, 840 F.3d 562, 568 (9th Cir.\\n2016) (“The establishment of a ‘baseline is not an\\nindependent legal requirement, but rather, a practical\\nrequirement in environmental analysis often employed to\\nidentify the environmental consequences of a proposed\\nagency action.’” (quoting Am. Rivers v. FERC, 201 F.3d\\n1186, 1195 n.15 (9th Cir. 1999))). But we have also\\ndetermined that “simply because the Final Rule may be an\\nimprovement over the [previous] standard does not\\nnecessarily mean that it will not have a ‘significant effect’\\non the environment” where the agency “has not explained\\nwhy its rule will not have a significant effect.” Ctr. for\\nBiological Diversity v. NHTSA, 538 F.3d 1172, 1224 (9th\\nCir. 2008). The plans chosen by the Forest Service might\\nreduce (even substantially reduce) the spread of noxious\\nweeds in the Districts, but that alone does not address\\nwhether that reduced impact itself has a significant impact\\non the environment generally and Grand Canyon National\\nPark in particular. Therefore, the Forest Service cannot rely\\nsolely on the reduction of adverse impacts to demonstrate\\nthat those impacts are not significant.\\n\\f26 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n But that the Forest Service occasionally conflates\\nreduction with insignificance does not necessarily mean that\\nit violated NEPA. To demonstrate, we once again go into\\nthe weeds.\\n\\n The Forest Service acknowledged that the North Kaibab\\nRanger District contained “several species of invasive\\nweeds,” which “are spread [] via roads and forest visitors.”\\nThe plan that was eventually selected “reduce[d] the number\\nof roads that can be traveled on by 376 miles,” which, the\\nEA found, would “lower the amount of invasive species seed\\nintroduced or spread.” But notably, the EA continued:\\n\\n The authorization of motorized big game\\n retrieval will have an increased threat of\\n invasive species spread as every vehicle that\\n travels cross-country has the ability to serve\\n as a vector and create disturbance.\\n Alternative 2 [the selected plan] authorizes\\n motorized big game retrieval for only elk and\\n mule deer. This is expected to lead to only a\\n small increase in the potential for invasive\\n species spread and disturbance when\\n compared to Alternative 3 and should not\\n generate any realistic impacts.\\n\\nThis passage demonstrates that the Forest Service did not\\nmerely rely on the possibility of reduction, but also\\nconcluded that the plan would “not generate any realistic\\nimpacts”; in other words, that the effects would not be\\nsignificant. Furthermore, as the district court noted, the EA\\nmentioned that the North Kaibab Ranger District featured\\n“projects focus[ed] on treating known infestations across the\\nDistrict, prioritizing the species and locations that pose the\\ngreatest threats,” methods that had “proven successful in\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 27\\n\\neradicating or reducing potentially serious noxious species\\nthreats.” This language indicates that the Forest Service\\nacknowledged a potential environmental impact, and then\\ndetermined that, due to features of the travel management\\nplan and other remediation efforts, it was unlikely to be\\nsignificant. Contrary to Plaintiffs’ argument, the Forest\\nService did not merely determine that the problem would be\\nreduced; it also concluded that the impact would not be\\nsignificant. Plaintiffs disagree with the EA’s conclusion, but\\nthis is “a classic example of a factual dispute the resolution\\nof which implicates substantial agency expertise.” Marsh,\\n490 U.S. at 376. We agree with the district court: “Plaintiffs’\\nsingular and conclusory statement that exotic plants might\\nspread . . . does not raise substantial questions that would\\ntrigger the need for an EIS.”\\n\\n Similarly, although the Forest Service did not follow all\\nof the recommendations made by Grand Canyon National\\nPark’s Acting Park Superintendent, this fact does not mean\\nthat it ignored a significant environmental impact. Agencies\\ncan thoughtfully consider suggestions but ultimately decide\\nto reject them, and the presence of an articulated concern\\ndoes not alone trigger the need to conduct an EIS. See Native\\nEcosystems Council v. U.S. Forest Serv., 428 F.3d 1233,\\n1240 (9th Cir. 2005) (“NEPA permits a federal agency to\\ndisclose [] impacts without automatically triggering the\\n‘substantial questions’ threshold.”). Here, the record\\nindicates that the Forest Service explained why a buffer\\nwould not be employed (“We don’t expect to use a buffer\\nzone as many management actions will need to extend to the\\nForest Service-National Park boundary”), and further\\narticulated means of remedying the risk of illegal motor\\nvehicle use (such as “limit[ing] motorized big game retrieval\\nduring all elk seasons” and “work[ing] closely with Arizona\\nGame and Fish Department to monitor and enforce illegal\\n\\f28 WILDEARTH GUARDIANS V. PROVENCIO\\n\\ncross-country travel associated with hunting activities”). We\\nfind no indication in the record that the Forest Service did\\nnot adequately consider potential impacts, or that substantial\\nquestions remained that required the preparation of an EIS.\\n\\n The same conclusion ultimately applies to all of the\\nenvironmental impacts that Plaintiffs highlight in their\\nbriefs: although Plaintiffs disagree with the EAs’ factual\\nconclusions, the Forest Service nonetheless considered the\\nissues, gave them the requisite “hard look,” and thus fulfilled\\ntheir NEPA obligations. Save the Yaak, 840 F.2d at 717. In\\nreaching its conclusions that none of the impacts cited by\\nPlaintiffs were sufficiently significant to require the\\npreparation of EISs, the Forest Service did not “rel[y] on\\nfactors Congress did not intend it to consider, ‘entirely fail[]\\nto consider an important aspect of the problem,’ or offer[] an\\nexplanation ‘that runs counter to the evidence before the\\nagency or is so implausible that it could not be ascribed to a\\ndifference in view or the product of agency expertise.’”\\nLands Council v. McNair, 537 F.3d 981, 987 (9th Cir. 2008)\\n(en banc) (quoting Earth Island Inst. v. U.S. Forest Serv.,\\n442 F.3d 1147, 1157 (9th Cir. 2006)), overruled on other\\ngrounds by Winter v. Nat. Res. Def. Council, Inc., 555 U.S.\\n7 (2008). Instead, the evidence in the record indicates that,\\nalthough the EAs acknowledged that motorized big game\\nretrieval might have negative impacts on the environment,\\nthe Forest Service’s determination that these impacts would\\nnot be significant evinced “a rational connection between the\\nfacts found and the conclusions made.” Or. Nat. Res.\\nCouncil v. Lowe, 109 F.3d 521, 526 (9th Cir. 1997).\\nTherefore, its conclusions were not arbitrary and capricious,\\nand the Forest Service did not violate NEPA by declining to\\nprepare EISs based on the plans’ environmental impacts.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 29\\n\\n ii. Controversy and Uncertainty\\n\\n NEPA also requires the preparation of an EIS when an\\naction’s “effects on the quality of the human environment\\nare likely to be highly controversial,” and/or “are highly\\nuncertain or involve unique or unknown risks.” 40 C.F.R.\\n§ 1508.27(b)(4)–(5). “A project is ‘highly controversial’ if\\nthere is a ‘“substantial dispute [about] the size, nature, or\\neffect of the major Federal action rather than the existence\\nof opposition to a use.”’” Native Ecosystems Council,\\n428 F.3d at 1240 (alteration in original) (emphases added)\\n(quoting Blue Mountains, 161 F.3d at 1212); see also\\nWetlands Action Network v. U.S. Army Corps of Eng’rs,\\n222 F.3d 1105, 1122 (9th Cir. 2000) (“The existence of\\nopposition to a use, however, does not render an action\\ncontroversial.”), abrogated on other grounds by Wilderness\\nSoc’y v. U.S. Forest Serv., 630 F.3d 1173 (9th Cir. 2011).\\n“A substantial dispute exists when evidence, raised prior to\\nthe preparation of an EIS or FONSI casts serious doubt upon\\nthe reasonableness of an agency’s conclusions.” Nat’l Parks\\n& Conservation Ass’n v. Babbitt, 241 F.3d 722, 736 (9th Cir.\\n2001) (citation omitted), abrogated on other grounds by\\nMonsanto Co. v. Geertson Seed Farms, 561 U.S. 139 (2010).\\nAdditionally, because “[a]n agency must generally prepare\\nan EIS if the environmental effects of a proposed agency\\naction are highly uncertain,” such preparation “is mandated\\n‘where uncertainty may be resolved by further collection of\\ndata, or where the collection of such data may prevent\\nspeculation on potential . . . effects.’” Barnes, 655 F.3d at\\n1140 (second alteration in original) (quoting Native\\nEcosystems Council, 428 F.3d at 1240). However, NEPA\\nregulations “do not anticipate the need for an EIS anytime\\nthere is some uncertainty, but only if the effects of the project\\nare ‘highly’ uncertain.” Envtl. Prot. Info. Ctr. v. U.S. Forest\\nServ., 451 F.3d 1005, 1011 (9th Cir. 2006).\\n\\f30 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n Plaintiffs contend that “the travel management plans for\\nthe Tusayan, Williams, and North Kaibab Ranger Districts\\npresent highly controversial and highly uncertain effects that\\ninvolve unique or unknown risks,” because “significant\\ncontroversy exists as to the amount and type of motorized\\nrecreation that would be allowed across the three Ranger\\nDistricts.” But Plaintiffs point to nothing in the record\\nindicating the existence of a substantial dispute that casts\\ndoubt on the Forest Service’s conclusions about\\nenvironmental impacts. There may have been opposition to\\nthe plans, but mere opposition alone is insufficient to support\\na finding of controversy. The Forest Service “recognize[d]\\nthat elements of the Selected Alternative [] generated\\ncontroversy,” but concluded—apparently correctly—that\\nthere was “no substantiated scientific controversy over the\\neffects as described.”\\n\\n Plaintiffs assert that various questions raised during the\\nEA process revealed a high level of scientific uncertainty,\\nbut the record belies that assertion. For example, as to the\\npotential risks “based on the broad allowance of motorized\\nbig game retrieval,” the Forest Service attempted to estimate\\nhunting activity based on past data, and used this information\\nto conclude that the impacts of motorized big game retrieval\\nwould be limited. Plaintiffs identity nothing in the record to\\nsuggest that the Forest Service’s estimates were unduly\\nspeculative, or that it unreasonably relied upon these\\npredictions. Plaintiffs also argue that there was “uncertainty\\nregarding whether or not hunters will actually remove gut\\npiles” when retrieving carcasses, which they must do “to\\nprotect California condors from lead poisoning.” But\\nalthough the Forest Service acknowledged that this issue\\nmight present a problem, the record also indicates that it\\nconsidered the issue and reasonably concluded that it was\\nunlikely to significantly impact the North Kaibab Ranger\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 31\\n\\nDistrict’s condors because the Arizona Game and Fish\\nDepartment had provided to hunters, among other\\nincentives, lead-free ammunition. The Forest Service also\\nnoted that “there would be decreased risk of human\\ndisturbance of scavenging condors as a result of a reduced\\nopen road system and substantially restricted motorized\\ncross-country travel,” and concluded that the North Kaibab\\nRanger District plan “is not likely to jeopardize the\\ncontinued existence [of] California condors.” Plaintiffs\\nneither challenge nor address these conclusions, and instead\\nrely on the mere existence of potential problems as evidence\\nof significant uncertainty—a tactic that does not pass muster.\\nSee Native Ecosystems Council, 428 F.3d at 1240 (“Simply\\nbecause a challenger can cherry pick information and data\\nout of the administrative record to support its position does\\nnot mean that a project is highly controversial or highly\\nuncertain.”).\\n\\n iii. Precedent for Future Actions\\n\\n Another consideration for measuring an action’s\\nintensity for NEPA purposes is “[t]he degree to which the\\naction may establish a precedent for future actions with\\nsignificant effects or represents a decision in principle about\\na future consideration.” 40 C.F.R. § 1508.27(b)(6).\\nAlthough “EAs are usually highly specific to the project and\\nthe locale, thus creating no binding precedent,” In Def. of\\nAnimals v. U.S. Dep’t of Interior, 751 F.3d 1054, 1071 (9th\\nCir. 2014) (quoting Barnes, 655 F.3d at 1140), and the Forest\\nService here explicitly found that each of the three travel\\nmanagement plans was “not likely to establish a precedent\\nfor future actions with significant effects,” Plaintiffs assert\\nthat this consideration is nonetheless implicated because\\n“the Forest Service has [] made public statements indicating\\n\\f32 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nthat what the Kaibab National Forest does, so too will other\\nsouthwestern National Forests.”\\n\\n It is true that the record contains evidence to this effect—\\nincluding that the Coconino National Forest “will defer to\\nthe neighboring Kaibab National Forest’s policy for\\n[motorized big game retrieval] in units shared with the\\nWilliams Ranger District, regardless of how the Coconino\\nproposes to apply the Travel Management Rule,” and that\\nthe Prescott National Forest will “match them as best as we\\ncan”—but that does not mean that the Districts’ plans bind\\nor necessarily shape other forests’ plans in such a way that\\nthey should be considered precedential, especially since any\\nother forest’s plan would be subject to its own NEPA\\nanalysis. Cf. Sierra Club v. Marsh, 769 F.2d 868, 879 (1st\\nCir. 1985) (determining that an action was precedential\\nbecause “once Maine completes the causeway and port,\\npressure to develop the rest of the island could well prove\\nirreversible”). In Oregon Wild v. Bureau of Land\\nManagement, a district court dealt with a similar situation,\\nwhere a project was “part of a larger series of ‘pilot projects’\\naimed at ‘inform[ing] long-term planning’ for management\\nof [] lands in Oregon and California.” No. 6:14-CV-0110-\\nAA, 2015 WL 1190131, at *9 (D. Or. Mar. 14, 2015) (first\\nalteration in original). There, as here, the most that could be\\nconcluded from such a minor precedential effect is that this\\nconsideration “supports the conclusion that an EIS is\\nnecessary”—but “the precedential factor alone is not\\ndispositive.” Id.; see also Anderson v. Evans, 371 F.3d 475,\\n493 (9th Cir. 2004) (finding this factor “insufficient on its\\nown to demonstrate a significant environmental impact”\\nwhere an action is merely influential and not binding). Thus,\\nthis consideration alone did not require preparation of an\\nEIS.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 33\\n\\n iv. Threatened Species\\n\\n Finally, there is the issue of the Mexican spotted owl, a\\nthreatened species found in the Williams and North Kaibab\\nRanger Districts. The Forest Service must consider “[t]he\\ndegree to which the action may adversely affect an\\nendangered or threatened species or its habitat that has been\\ndetermined to be critical under the Endangered Species Act\\nof 1973.” 40 C.F.R. § 1508.27(b)(9). Plaintiffs point to\\nportions of the record that raise the specter of adverse\\neffects—including consultation letters between the Williams\\nand North Kaibab Ranger Districts and the U.S. Fish and\\nWildlife Service regarding the owls—but they again ignore\\nthe Forest Service’s overall conclusions. For example, the\\nWilliams Ranger District EA reported that some roads would\\npass through Mexican spotted owl critical habitats, which\\nmight result in “increased potential human disturbance.”\\nBut the EA ultimately concluded that the selected plan\\nwould be “primarily beneficial,” and would “not adversely\\naffect Mexican spotted owl or Mexican spotted owl\\ndesignated Critical Habitat.” In a separate biological\\nassessment, the Forest Service concluded that “[t]he effects\\ndetermination for Mexican spotted owl and Mexican spotted\\nowl critical habitat is may affect, not likely to adversely\\naffect,” based on the\\n\\n determination [] that potential effects of the\\n proposed action on the Mexican spotted owl\\n would be primarily beneficial. Closing roads\\n and restricting motorized cross-country\\n travel under the proposed action would result\\n in reduced motorized access to spotted owl\\n habitat compared to current management and\\n thus reduced risk of human disturbance to\\n spotted owls, reduced impacts to habitat of\\n\\f34 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n spotted owls and their small mammal prey\\n species, and reduced impacts to designated\\n critical habitat.\\n\\nNotably, the U.S. Fish and Wildlife Service concurred in this\\ndetermination.\\n\\n In short, although the Forest Service did not definitively\\nconclude that no Mexican spotted owls would be adversely\\naffected by the Districts’ travel management plans, the\\nrecord indicates that they sufficiently considered the issue\\nand arrived at a reasonable conclusion that the effects would\\nnot be significant, thus obviating the need for an EIS.\\n\\n v. Summation\\n\\n In the end, we conclude that the Forest Service’s\\ndetermination that no EISs were needed as to the Districts’\\ntravel management plans was reasonable. The plans might\\nhave some precedential effect, there is a possibility that\\nMexican spotted owls might be affected, and exotic weeds\\nmight be spread by motorized big game retrieval, but the\\nrecord ultimately supports the Forest Service’s conclusion\\nthat these concerns do not rise to the level of significance\\nthat would require EISs. The Forest Service gave the\\nrequisite hard look and made determinations that were\\nneither arbitrary nor capricious, and were consistent with the\\nevidence before it. Absent substantial questions that would\\nhave mandated EISs, the Forest Service did not violate\\nNEPA. 7\\n\\n\\n 7\\n Two other enumerated considerations are relevant in this case:\\n“[t]he degree to which the action may adversely affect . . . objects listed\\nin or eligible for listing in the National Register of Historic Places or may\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 35\\n\\nIII. NHPA\\n\\n Finally, Plaintiffs contend that the Forest Service\\nviolated the NHPA by failing to identify and evaluate the\\n“high density of cultural resources” that might be damaged\\nas a result of motorized travel in the Districts.\\n\\n The NHPA’s purpose is to “foster conditions under\\nwhich our modern society and our historic property can exist\\nin productive harmony,” 54 U.S.C. § 300101(1), and it\\nrequires federal agencies to “make a reasonable and good\\nfaith effort” to identify historic properties that might be\\naffected by an action, and to “take [those potential effects]\\ninto account.” 36 C.F.R. § 800.4(b)(1); see also 54 U.S.C.\\n§ 306108. 8 “Like NEPA, ‘[s]ection 106 of NHPA is a “stop,\\nlook, and listen” provision that requires each federal agency\\nto consider the effects of its programs.’” Te-Moak Tribe of\\nW. Shoshone of Nev. v. U.S. Dep’t of Interior, 608 F.3d 592,\\n\\n\\n\\n\\ncause loss or destruction of significant scientific, cultural, or historical\\nresources,” and “[w]hether the action threatens a violation of Federal,\\nState, or local law or requirements imposed for the protection of the\\nenvironment.” 40 C.F.R. § 1508.27(b)(8), (10). However, the former\\nconsideration essentially depends on whether the Forest Service\\ncomplied with the NHPA, while the latter hinges on the Forest Service’s\\ncompliance with both the NHPA and the Travel Management Rule.\\nBecause we conclude that the Forest Service complied with both, we also\\nconclude that no federal laws were violated, and no cultural resources\\nadversely affected, such that EISs were required based on these\\nconsiderations.\\n\\n 8\\n “Historic property means any prehistoric or historic district, site,\\nbuilding, structure, or object included in, or eligible for inclusion in, the\\nNational Register of Historic Places maintained by the Secretary of the\\nInterior.” 36 C.F.R. § 800.16(l)(1).\\n\\f36 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n607 (9th Cir. 2010) (quoting Muckleshoot Indian Tribe v.\\nU.S. Forest Serv., 177 F.3d 800, 805 (9th Cir. 1999)).\\n\\n Plaintiffs assert that the Forest Service violated the\\nNHPA in three ways: (1) by failing “to make a reasonable\\nand good faith effort to identify and evaluate cultural\\nproperties”; (2) by “erroneously determin[ing] that\\n‘Exemption Q’ excuses NHPA consultation for the Tusayan\\nand Williams Ranger Districts”; and (3) by arbitrarily\\nmaking “a ‘no adverse effect’ determination for the North\\nKaibab Ranger District after admitting that cross-country\\ntravel damages cultural resources.” We consider each\\nargument in turn.\\n\\n A. Identification of Cultural Properties\\n\\n Section 106 of the NHPA requires the Forest Service to\\n“make a reasonable and good faith effort to identify historic\\nproperties; determine whether identified properties are\\neligible for listing on the National Register . . .; [and] assess\\nthe effects of the undertaking on any eligible historic\\nproperties found.” Te-Moak Tribe, 608 F.3d at 607 (first\\nalteration in original) (quoting Muckleshoot Indian Tribe,\\n177 F.3d at 805). It must also engage in consultation with\\nthe State Historic Preservation Officer (SHPO) to\\n“[d]etermine and document the area of potential effects,”\\n“[g]ather information,” and “develop and evaluate\\nalternatives or modifications to the undertaking that could\\navoid, minimize, or mitigate adverse effects on historic\\nproperties.” 36 C.F.R. §§ 800.4(a), 800.6(a).\\n\\n Plaintiffs principally argue that the Forest Service failed\\nto “make a reasonable and good faith effort” because it did\\nnot complete 100 percent surveys of potentially affected\\nareas. The First Amended Programmatic Agreement\\nRegarding Historic Property Protection and Responsibilities\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 37\\n\\n(Programmatic Agreement), which the parties agree\\nprovided the relevant guidelines, mandated that “[t]he level\\nof need and extent of new field surveys or inspections will\\nbe proposed by a Professional Cultural Resource Specialist\\nand approved by the Forest Archaeologist based on the\\nguidelines provided in this section.” Those guidelines\\nincluded the use of “relevant information to assess the\\npotential to affect historic properties and the expected nature\\nand distribution of heritage properties that may be affected”;\\nnamely, “[t]he expected nature and severity of all associated\\nimpacts” and “[t]he expected nature and distribution of\\nheritage resources.” Based on the results of this “prefield\\nresearch,” the Resource Specialist and Forest Archeologist\\nwere directed to “determine the relative level of field survey\\nto be conducted.” The Programmatic Agreement called for\\n“100% surveys” where “site density is expected to be high”\\nand where “site densities are unknown and expected visitor\\nuse or impacts will be high.” By contrast, “areas may be\\nsurveyed at less than 100%” where “known site density is\\nlow.”\\n\\n Here, the record supports the Forest Service’s conclusion\\nthat the Programmatic Agreement did not require\\n100 percent surveys. At the time the travel management\\nplans were decided, the Forest Service had surveyed 42\\npercent of the Williams Ranger District, 23 percent of the\\nTusayan Ranger District, and 25 percent of the North Kaibab\\nRanger District. It concluded that no further surveying was\\nrequired, based on the expected density of cultural resources\\nand the nature and severity of impacts upon them.\\n\\n The record reinforces the determination that the expected\\nimpacts of the travel management plans would be low. After\\nconsidering historic hunting data, the EAs concluded that\\nless than 0.1 percent of each District’s acreage would be\\n\\f38 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nimpacted by motorized big game retrieval. However, the\\nfact that “[t]he expected nature and severity of all associated\\nimpacts” might have been low does not necessarily mean\\nthat the Forest Service did not need to conduct 100 percent\\nsurveys, for the Programmatic Agreement required\\n100 percent surveys for high-density sites regardless of\\nimpact level. Accordingly, the Forest Service’s obligation\\nrested on “[t]he expected nature and distribution of heritage\\nresources.”\\n\\n The record is somewhat unclear as to the density of\\nheritage resources in the areas open to motorized big game\\nretrieval. Plaintiffs note that many of the Districts’ cultural\\nresources are close to the roads, but that does not necessarily\\nspeak to the density of resources because the Programmatic\\nAgreement does not provide any benchmark or guidance as\\nto what constitutes a high-density site. Ultimately, the lack\\nof clarity weighs in the Forest Service’s favor: it was not\\nunreasonable for it to determine that the density was low or\\nunknown, either of which would have excused 100 percent\\nsurveys. 9 Thus, the Forest Service followed a reasonable\\ninterpretation of the Programmatic Agreement.\\n\\n B. Exemption Q\\n\\n The Programmatic Agreement’s Exemption Q provided\\nthat “[a]ctivities not involving ground or surface disturbance\\n(e.g., timber stand improvement and precommercial thinning\\nby hand)” are “exempt from further review and\\nconsultation.” Plaintiffs argue that the Forest Service\\n 9\\n The Programmatic Agreement required 100 percent surveys\\n“where site densities are unknown and expected visitor use or impacts\\nwill be high,” so it would have been reasonable for the Forest Service to\\nconclude that 100 percent surveys were not needed where densities were\\nunknown and use impacts were low.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 39\\n\\narbitrarily relied on Exemption Q to excuse review of\\nmotorized big game retrieval in the Williams and Tusayan\\nRanger Districts.\\n\\n Once again, however, the record does not support\\nPlaintiffs’ assertions. We agree with Plaintiffs that invoking\\nExemption Q would have been inappropriate here. It is\\nclear, and the Forest Service does not dispute, that motorized\\ntravel causes surface disturbance. But although the Forest\\nService’s correspondence with non-party Center for\\nBiological Diversity suggested that it applied Exemption\\nQ, 10 the record as a whole supports a contrary conclusion.\\nThe Forest Service consulted with both the Arizona SHPO\\nand potentially affected tribes as to each District’s travel\\nmanagement plan—consultations that would not have been\\nrequired if Exemption Q had been applied. Moreover, the\\nForest Service made no other references to Exemption Q as\\npart of the travel management plan decision process, which\\nfurther supports the conclusion that it was not in fact\\ninvoked. Accordingly, the references to Exemption Q at\\n\\n 10\\n The Forest Service claims that Plaintiffs’ “brief does not identify\\nwhere or how the Tusayan and Williams decisions purportedly relied on\\nExemption Q,” and that their “responses to the non-party Center for\\nBiological Diversity’s administrative appeals of those decisions” contain\\nonly “an isolated reference to Exemption Q,” but these assertions are\\nmisleading. In its response to the appeal regarding the Tusayan Ranger\\nDistrict, the Forest Service wrote that it “determined that motorized big\\ngame retrieval fell under Exemption Q of the PA, ‘Activities not\\ninvolving ground or surface disturbance (e.g., timber stand improvement\\nand precommercial thinning by hand)’ and would have limited impacts\\nsimilar to the examples cited in the exemption in the” Programmatic\\nAgreement. That is more than a mere “isolated reference,” as it implies\\nthat the Forest Service actually applied Exemption Q to the Tusayan\\nRanger District. This same language invoking Exemption Q appeared in\\nthe Forest Service’s response regarding the Williams Ranger District.\\n\\f40 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nmost amounted to harmless error, as they had no effect on\\nthe NHPA consultation process. See 5 U.S.C. § 706\\n(requiring the court to take “due account . . . of the rule of\\nprejudicial error” when conducting APA review); Tucson\\nHerpetological Soc’y v. Salazar, 566 F.3d 870, 880 (9th Cir.\\n2009) (“We have held that the harmless error doctrine ‘may\\nbe employed only “when a mistake of the administrative\\nbody is one that clearly had no bearing on the procedure\\nused or the substance of decision reached.”’” (quoting\\nGifford Pinchot Task Force v. U.S. Fish & Wildlife Serv.,\\n378 F.3d 1059, 1071 (9th Cir. 2004))). 11\\n\\n C. “No Adverse Effect” Determination\\n\\n Finally, Plaintiffs argue that the Forest Service’s\\nconclusion that motorized big game retrieval would have no\\nadverse effect on cultural resources was arbitrary. But to\\nbuttress this assertion, Plaintiffs again cherry pick isolated\\nsegments from the record without considering their broader\\ncontext. For example, the Cultural Resources Specialist\\nReport prepared for the North Kaibab Ranger District noted\\nthat “[c]ross country motorized travel, whether to retrieve\\ngame or for other purposes, can adversely affect cultural\\nresource sites if a vehicle is driven across a site,” since\\n“[v]ehicles can [] crush or displace artifacts and features\\n\\n 11\\n Curiously, although the Forest Service similarly argued in the\\ndistrict court that “the agency’s decisions did not rely on Exemption Q,”\\nwe note that the court concluded that Exemption Q did apply. On appeal,\\nthe Forest Service does not argue that the district court’s conclusion on\\nthis point was correct, and we can affirm the court’s ruling even though\\nwe agree with the parties that Exemption Q was not in fact applicable.\\nSee Applied Underwriters, Inc. v. Lichtenegger, 913 F.3d 884, 892 (9th\\nCir. 2019) (“We will affirm the district court’s correct legal results, even\\nif reached for the wrong reasons.” (quoting Alcaraz v. Block, 746 F.2d\\n593, 602 (9th Cir. 1984))).\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 41\\n\\nimpacting the physical integrity of the site and impairing or\\ndestroying scientific information that may contribute to the\\nunderstanding of the history and prehistory of an area.” But\\nthat same report also concluded as follows:\\n\\n Quantifying the potential for damage from\\n big game retrieval is difficult. The results\\n vary depending on the number of game\\n retrieval trips annually, the location of those\\n retrievals (high site probability areas versus\\n low probability), site types found in the area,\\n soil characteristics, routes used to access the\\n game and weather conditions at the time of\\n retrieval. However, the fewer number of\\n motorized trips that occur, the lower the\\n likelihood of encountering and impacting a\\n site. . . .\\n\\n Limiting cross-country travel will have a\\n beneficial effect on cultural resources by\\n reducing the potential for sites to be\\n damaged. This alternative would restrict\\n motorized big game retrieval to elk and\\n bison. In 2009, 38 buffalo and no elk were\\n taken . . . . While there is a possibility that\\n cross-country game retrieval of either of\\n these species could impact a cultural resource\\n site, given the low number of takes each year,\\n it is anticipated that the potential for adverse\\n effects to a site would be negligible:\\n 38 entries per year equates to less than\\n .0099% of the acreage on the NKRD. The\\n odds of adversely affecting a cultural\\n resource site under these conditions are\\n extremely low.\\n\\f42 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n(emphases added). In light of this ultimate conclusion, as\\nwell as the implementation of the suggested mitigation\\nmeasures, it was not arbitrary for the Forest Service to\\nconclude, based on the evidence before it, that adverse\\neffects were unlikely. See Motor Vehicle Mfrs. Ass’n of\\nU.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43\\n(1983) (“[A]n agency rule would be arbitrary and capricious\\nif the agency . . . offered an explanation for its decision that\\nruns counter to the evidence before the agency.”).\\n\\n D. Summation\\n\\n The NHPA—and NEPA—“create obligations that are\\nchiefly procedural in nature.” San Carlos Apache Tribe,\\n417 F.3d at 1097 (quoting Pres. Coal., Inc. v. Pierce,\\n667 F.2d 851, 859 (9th Cir. 1982)). Even if cultural\\nresources might be harmed as a result of motorized big game\\nretrieval, that fact alone does not indicate that the Forest\\nService violated the NHPA. The Forest Service conducted\\nthe required prefield work, consulted with the appropriate\\nentities, and reached a determination consistent with the\\nevidence before it—in short, satisfied its procedural\\nobligations.\\n\\n CONCLUSION\\n\\n The travel management plans that the Forest Service\\nimplemented in the three Ranger Districts of the Kaibab\\nNational Forest limited motorized big game retrieval to\\ncertain roads and imposed additional restrictions to reduce\\nthe level and effect of motorized activity. In crafting the\\nplans, the Forest Service investigated potential impacts on\\nboth the environment and historic properties and reasonably\\ndetermined that no further action was needed. Accordingly,\\nwe conclude that the Forest Service followed the Travel\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 43\\n\\nManagement Rule and fulfilled its procedural obligations\\nunder NEPA and the NHPA. 12\\n\\n AFFIRMED.\\n\\n\\n\\n\\n 12\\n Amicus Curiae Rocky Mountain Elk Foundation filed a motion\\nfor leave to file an amicus brief in support of the Forest Service. We\\nconclude that the proposed brief provides neither legal nor factual\\nsupport to help resolve the issues on appeal, and so deny the motion.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4376402/\", \"author_raw\": \"M. SMITH, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"MICHAEL DALY HAWKINS"},"judge2":{"kind":"string","value":"MILAN D SMITH JR"},"judge3":{"kind":"string","value":"ANDREW D HURWITZ"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{\"MICHAEL DALY HAWKINS\": \", Circuit\", \"MILAN D SMITH JR\": \", Circuit\", \"ANDREW D HURWITZ\": \", Circuit\"}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4599149/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Wildearth Guardians v. Heather Provencio\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-13\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nWILDEARTH GUARDIANS; GRAND No. 17-17373\nCANYON WILDLANDS COUNCIL;\nWILDLANDS NETWORK; SIERRA D.C. No.\nCLUB, 3:16-cv-08010-\n Plaintiffs-Appellants, SMM\n\n v.\n OPINION\nHEATHER PROVENCIO, in her official\ncapacity as Kaibab National Forest\nSupervisor; UNITED STATES FOREST\nSERVICE,\n Defendants-Appellees,\n\n and\n\nSTATE OF ARIZONA, on behalf of\nArizona Department of Game and\nFish; SAFARI CLUB INTERNATIONAL,\n Intervenors-Defendants-Appellees.\n\n\n\n Appeal from the United States District Court\n for the District of Arizona\n Stephen M. McNamee, Senior District Judge, Presiding\n\n Argued and Submitted February 7, 2019\n Phoenix, Arizona\n\f2 WILDEARTH GUARDIANS V. PROVENCIO\n\n Filed March 13, 2019\n\n Before: MICHAEL DALY HAWKINS, MILAN D.\n SMITH, JR., and ANDREW D. HURWITZ,\n Circuit Judges.\n\n Opinion by Judge Milan D. Smith, Jr.\n\n\n SUMMARY *\n\n\n Environmental Law\n\n The panel affirmed the district court’s summary\njudgment in favor of the United States Forest Service in an\naction by plaintiff environmental groups challenging travel\nmanagement plans implemented by the Forest Service to\npermit limited motorized big game retrieval in three Ranger\nDistricts of the Kaibab National Forest.\n\n The Travel Management Rule, promulgated by the U.S.\nDepartment of Agriculture for Forest Service lands,\ngenerally prohibits off-road, motorized travel, but permits\nthe “limited” use of motor vehicles within a specified\ndistance of “certain” forest roads for the purposes of\ncamping or retrieval of downed big game animals. The\npanel rejected plaintiffs’ contention that the Forest Service\nviolated the Travel Management Rule by implementing\nplans that did not sufficiently limit motorized big game\nretrieval in the Ranger Districts. The panel concluded that\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f WILDEARTH GUARDIANS V. PROVENCIO 3\n\nthe Forest Service did not violate the plain terms of the\nTravel Management Rule.\n\n Addressing plaintiffs’ claims under the National\nEnvironmental Policy Act (“NEPA”), the panel held that the\nplaintiffs had standing to bring their NEPA claims because\nthey were trying to protect the environment, which was\nwithin NEPA’s zone of interests. The panel concluded that\nthe environmental impacts discussed in the environmental\nassessments did not raise substantial concerns that\nnecessitated the preparation of environmental impact\nstatements. The panel held that there was no indication that\nthe agency failed to satisfy NEPA’s procedural\nrequirements. The panel concluded that the Forest Service\ngave the requisite hard look and made determinations that\nwere neither arbitrary nor capricious, and were consistent\nwith the evidence before it; and accordingly, the Forest\nService did not violate NEPA.\n\n The panel held that the Forest Service conducted the\nrequired prefield work, consulted with the appropriate\nentities, and reached a determination with the evidence\nbefore it, and satisfied its procedural obligations under the\nNational Historic Preservation Act.\n\f4 WILDEARTH GUARDIANS V. PROVENCIO\n\n COUNSEL\n\nJohn R. Mellgren (argued), and Susan Jane Brown, Western\nEnvironmental Law Center, Eugene, Oregon, for Plaintiffs-\nAppellants.\n\nStuart Wilcox, WildEarth Guardians, Denver, Colorado, for\nPlaintiff-Appellant WildEarth Guardians.\n\nMark R. Haag (argued), Allen M. Brabender, Attorneys,\nEnvironment & Natural Resources Division; Eric Grant,\nDeputy Assistant Attorney General, Jeffrey H. Wood,\nActing Assistant Attorney General, United States\nDepartment of Justice, Washington, D.C.; M’Leah\nWoodard, Attorney Advisor, Albuquerque Field Office,\nOffice of the General Counsel, United States Department of\nAgriculture, Albuquerque, New Mexico; for Defendants-\nAppellees.\n\nDominic Draye (argued), John LeSueur, Assistant Attorney\nGeneral, and Mark Brnovich, Attorney General, Arizona\nAttorney General’s Office, Phoenix, Arizona, for\nIntervenor-Defendant-Appellee State of Arizona.\n\nAnna M. Seidman, Douglas S. Burdin, and Jeremy E. Clare,\nSafari Club International, Washington, D.C., for Intervenor-\nDefendant-Appellee Safari Club International.\n\f WILDEARTH GUARDIANS V. PROVENCIO 5\n\n OPINION\n\nM. SMITH, Circuit Judge:\n\n Plaintiffs-Appellants (Plaintiffs) are environmental\nadvocacy groups that challenged travel management plans\nimplemented by Defendant-Appellee United States Forest\nService (the Forest Service) to permit limited motorized big\ngame retrieval in three Ranger Districts of the Kaibab\nNational Forest. The district court granted the Forest\nService’s motion for summary judgment, concluding that the\nForest Service complied with the Travel Management Rule,\nthe National Environmental Policy Act (NEPA), and the\nNational Historic Preservation Act (NHPA). We affirm.\n\n FACTUAL AND PROCEDURAL BACKGROUND\n\nI. Factual Background\n\n A. The Kaibab National Forest\n\n The Kaibab National Forest encompasses approximately\n1.6 million acres of public land in northern Arizona,\nincluding Grand Canyon National Park. It is comprised of\nthree noncontiguous Ranger Districts: the Williams Ranger\nDistrict, the Tusayan Ranger District, and the North Kaibab\nRanger District.\n\n The Williams Ranger District is the southernmost,\ncovering 560,305 acres approximately thirty-five miles west\nof Flagstaff and sixty miles south of Grand Canyon National\nPark. It includes the Kendrick Mountain Wilderness, which\nextends into Coconino National Forest and features a diverse\narray of vegetation including Douglas firs, white firs,\nponderosa pines, and aspens. The Williams Ranger District\nalso serves as a habitat for a number of endangered species,\n\f6 WILDEARTH GUARDIANS V. PROVENCIO\n\nincluding the Mexican spotted owl, the California condor,\nand the black-footed ferret. It contains six areas where\nspotted owls are known to live and breed, and three spotted\nowl critical habitats overlap the District.\n\n The Tusayan Ranger District, located just south of Grand\nCanyon National Park’s south rim, encompasses 331,427\nacres. It features varied terrain, from ponderosa pine forests\nto grasslands, and is home to a number of sensitive species,\nincluding bald eagles, goshawks, peregrine falcons,\nburrowing owls, bats, and voles.\n\n The North Kaibab Ranger District covers 655,078 acres\nimmediately north of Grand Canyon National Park. Like the\nWilliams and Tusayan Ranger Districts, the North Kaibab\nRanger District boasts diverse terrain and vegetation, as well\nas sensitive animal species. Two federally listed endangered\nspecies—the Mexican spotted owl and California condor—\nlive in the District, which the U.S. Fish and Wildlife Service\nhas designated as critical habitat for the spotted owl.\n\n B. The Travel Management Rule\n\n In 2005, the U.S. Department of Agriculture\npromulgated a Travel Management Rule to “provide[] for a\nsystem of National Forest System roads, National Forest\nSystem trails, and areas on National Forest System lands that\nare designated for motor vehicle use.” 36 C.F.R.\n§ 212.50(a). 1 As part of this system, “[d]esignated roads,\ntrails, and areas [are] identified on a motor vehicle map,”\nwhich also “specif[ies] the classes of vehicles” and “the\n 1\n The Rule’s antecedents include executive orders issued by\nPresidents Nixon and Carter that sought to limit the damage to federal\npublic lands caused by off-road vehicles. See Utah Shared Access All.\nv. Carpenter, 463 F.3d 1125, 1129–30 (10th Cir. 2006).\n\f WILDEARTH GUARDIANS V. PROVENCIO 7\n\ntimes of year for which use is designated.” Id. § 212.56.\nMotor vehicle use is prohibited on roads not so designated.\nId. § 212.50(a). The regulations include a specific provision\nconcerning the use of motor vehicles for dispersed camping\nand big game retrieval, which allows that\n\n in designating routes, the responsible official\n may include in the designation the limited use\n of motor vehicles within a specified distance\n of certain forest roads or trails where motor\n vehicle use is allowed, and if appropriate\n within specified time periods, solely for the\n purposes of dispersed camping or retrieval of\n a downed big game animal by an individual\n who has legally taken that animal.\n\nId. § 212.51(b).\n\n The Forest Service’s Southwestern Regional Office\nissued guidelines for implementation of the Travel\nManagement Rule, including its motorized big game\nretrieval provision. The guidelines noted that “National\nForests in the Southwestern Region provide hunting\nopportunities that are important to the public,” and directed\nforests to identify designated routes for game retrieval “in\nclose collaboration with the responsible State agency.” They\nalso suggested, pursuant to discussions with the Arizona\nGame and Fish Department, that motorized big game\nretrieval be allowed “up to three miles from a designated\nroute” for bison and “up to one mile from a designated route”\nfor elk and mule deer.\n\f8 WILDEARTH GUARDIANS V. PROVENCIO\n\n The Forest Service crafted travel management plans for\neach of the three Ranger Districts in the Kaibab National\nForest. It also prepared an Environmental Assessment (EA)\nfor each plan to ascertain its environmental impact, but did\nnot undertake a more rigorous Environmental Impact\nStatement (EIS).\n\n i. The Williams Ranger District\n\n In July 2010, the Forest Service released the EA for the\nWilliams Ranger District’s travel management plan, and\nsubsequently issued a Decision Notice and Finding of No\nSignificant Impact (DN/FONSI). The DN/FONSI generally\n“prohibit[s] motorized travel off of designated routes on the\nWilliams Ranger District,” but permits “the limited use of\nmotor vehicles within one mile of all designated system\nroads (except where prohibited) to retrieve a legally hunted\nand tagged elk during all elk hunting seasons.” It allows\nmotorized big game retrieval of elk (but not bison) up to one\nmile off all designated open roads, so long as hunters make\nonly “[o]ne trip that uses [the] most direct route and least\nground disturbing.” The designated open road system\nconsists of 1,114 miles of roadway, a reduction from\nprevious motor vehicle activity, when 1,460 miles of roads\nand 95 percent of the District were open to motor vehicle\nuse. Several miles of the open roads pass through the spotted\nowl critical habitat.\n\n ii. The Tusayan Ranger District\n\n Previously, the Tusayan Ranger District contained more\nthan 700 miles of roads open to motor vehicles, and a vast\nmajority of the District was open to cross-country motor\nvehicle travel. The Forest Service’s final DN/FONSI for the\n\f WILDEARTH GUARDIANS V. PROVENCIO 9\n\nDistrict 2 designated 566 miles of road open to motor\nvehicles. The decision permits “[l]egally harvested elk [to]\nbe retrieved during all legal elk hunting seasons” by motor\nvehicles within one mile of designated roads. Motorized\nretrieval of bison is not permitted, and the DN/FONSI limits\nuse of motor vehicles when “conditions are such that travel\nwould cause damage to natural and/or cultural resources,”\nand mandated that “[m]otorized vehicles would not be\npermitted to cross riparian areas, streams and rivers except\nat hardened crossings or crossings with existing culverts.”\n\n iii. The North Kaibab Ranger District\n\n Prior to implementation of a new travel management\nplan, 1,852 miles of road in the North Kaibab Ranger District\nwere open to motor vehicle use, with 83 percent of the\nDistrict open to cross-county travel. In September 2012, the\nForest Service released an EA analyzing the District’s new\nplan. Among other data, the EA noted that while “[c]ross-\ncountry motorized travel, whether to retrieve game or for\nother purposes, can adversely affect cultural resource sites if\na vehicle is driven across a site,” only thirty-eight bison and\nno elk were taken from the District in 2009.\n\n The Forest Service issued a DN/FONSI that designated\n1,476 miles of open roads for motorized travel, including an\nadditional 16 miles of unauthorized, user-created roads.\nMotor vehicles can be used to retrieve elk or bison during\nhunting seasons, under certain limiting conditions. Notably,\n\n 2\n In April 2009, the Forest Service issued an initial EA that analyzed\nthe impact of the new travel management plan, as well as a subsequent\nDN/FONSI. In response to administrative appeals, that decision was\nreversed, and a new environmental survey undertaken. The resulting\nEA, issued in January 2011, reflected additional analysis and public\ncomment.\n\f10 WILDEARTH GUARDIANS V. PROVENCIO\n\nthe plan prohibits motorized retrieval of mule deer; the data\nindicated that far more mule deer—1,020—were harvested\nin the District in 2009 than bison or elk. The DN/FONSI\nalso included guidance for monitoring and mitigation, as\nwell as practices to limit the spread of invasive exotic weeds.\n\nII. Procedural Background\n\n The Districts’ travel management plans—their\nmotorized big game retrieval provisions in particular—were\nadministratively appealed, and the Regional Forester upheld\nthem.\n\n Plaintiffs then filed a complaint for declaratory and\ninjunctive relief in the district court. They challenged the\ntravel management plans for each of the three Ranger\nDistricts, alleging violations of the Travel Management\nRule, the Administrative Procedure Act (APA), NEPA, and\nthe NHPA. After the parties filed and briefed cross-motions\nfor summary judgment, the district court granted the Forest\nService’s motion and denied Plaintiffs’ motion. It further\ndenied the motions for summary judgment filed by\nIntervenors-Defendants-Appellees State of Arizona (the\nState) and Safari Club International (Safari Club) as moot.\nThis timely appeal followed.\n\n STANDARD OF REVIEW AND JURISDICTION\n\n We review de novo a district court’s order granting or\ndenying a motion for summary judgment. Churchill County\nv. Norton, 276 F.3d 1060, 1071 (9th Cir. 2001). Under the\nAPA, agency action can be set aside if it is “arbitrary,\ncapricious, an abuse of discretion, or otherwise not in\naccordance with law.” 5 U.S.C. § 706(2)(A); see also\nWildEarth Guardians v. Mont. Snowmobile Ass’n, 790 F.3d\n920, 932 (9th Cir. 2015) (Travel Management Rule reviewed\n\f WILDEARTH GUARDIANS V. PROVENCIO 11\n\nunder the APA); San Carlos Apache Tribe v. United States,\n417 F.3d 1091, 1099 (9th Cir. 2005) (NHPA challenge\nreviewed under the APA); Churchill County, 276 F.3d at\n1071 (NEPA challenge reviewed under the APA).\n\n We have jurisdiction pursuant to 28 U.S.C. § 1291.\n\n ANALYSIS\n\nI. The Travel Management Rule\n\n Plaintiffs argue that the Forest Service violated the\nTravel Management Rule by implementing plans that did not\nsufficiently limit motorized big game retrieval in the\nDistricts.\n\n Although the Travel Management Rule generally\nprohibits off-road, motorized travel, it permits officials to\ndesignate “the limited use of motor vehicles within a\nspecified distance of certain forest roads . . . solely for the\npurposes of dispersed camping or retrieval of a downed big\ngame animal.” 36 C.F.R. § 212.51(b) (emphases added). As\npart of this designation process, the Forest Service must\nconsider various criteria, ranging from “public safety” and\n“conflicts among uses of National Forest System lands” to\n“[d]amage to soil, watershed, vegetation, and other forest\nresources” and “[h]arassment of wildlife and significant\ndisruption of wildlife habitats.” 36 C.F.R. § 212.55(a)–(b).\nCompliance with these “minimization criteria” is\nmandatory. Mont. Snowmobile, 790 F.3d at 929–32.\n\n Plaintiffs contend that the Forest Service violated the\nTravel Management Rule by permitting off-road motorized\nvehicle use to collect downed game within one mile of every\nopen road in the Districts, in purported violation of the\n\f12 WILDEARTH GUARDIANS V. PROVENCIO\n\nRule’s mandate that such activity be “limited” and only on\n“certain” roads. We disagree.\n\n A. “Limited”\n\n First, Plaintiffs argue that “designating cross-country\noff-road motor vehicle use for one mile off both sides of\nevery single open road on each Ranger District is not a\n‘limited’ designation as contemplated by the Travel\nManagement Rule.” They note that the Forest Service’s own\nregional office acknowledged as much when reviewing the\nNorth Kaibab Ranger District’s plan proposal; the office\ncommented, “Motorized [big game retrieval] is being\nproposed on all system routes which is not consistent with\nrule for ‘limited use.’” The assistant NEPA coordinator of\nthat District also questioned the scope of the plan, noting,\n“What I don’t see is how this got you to your proposed\nactions, particularly the [motorized big game retrieval] on\nALL roads with 1 mile corridor . . . . that’s huge.” In their\nbrief, Plaintiffs include maps prepared by the Forest Service,\nwhich illustrate the extent of areas (in light blue) where\nmotorized big game retrieval is allowed in the Tusayan and\nWilliams Ranger Districts. These maps indicate, as\nPlaintiffs argue, that the plans permit motorized big game\nretrieval across a vast swath of the Districts’ lands:\n\fWILDEARTH GUARDIANS V. PROVENCIO 13\n\f14 WILDEARTH GUARDIANS V. PROVENCIO\n\n As the district court correctly noted, however, Plaintiffs’\nfocus on only the spatial limitation of the Districts’ plans\nignores the other restrictions on motorized big game\nretrieval. The North Kaibab Ranger District plan illustrates\nsome of these additional limitations: restricting retrieval to\nlegally hunted elk and bison 3; permitting only one vehicle\nper harvested animal; requiring hunters to “use the most\ndirect and least ground disturbing route in and out of the area\nto accomplish the retrieval”; and limiting the temporal\nperiod to the “appropriate season as designated by the\n[State], and for 24 hours following each season.” The\nWilliams and Tusayan Ranger Districts further exclude the\nretrieval of bison and require motor vehicles to cross streams\nand rivers only at designated crossings. We agree with the\ndistrict court that “these limitations are a significant\ndeparture from the previous policy which did not limit the\nnumber of trips . . . , did not limit the type of species which\ncould be retrieved by motor vehicle, did not limit the\ndistance traveled from system roads, and had no restrictions\non seasons or weather conditions and no requirement for use\nof a direct route.”\n\n Plaintiffs assert that “the nearly unlimited spatial\nallowance . . . in and of itself[] violates the plain terms of the\nTravel Management Rule,” but provide no authority for that\nproposition. Because we find no support for it in either case\nlaw or the applicable regulations, we conclude that the Forest\nService did not abuse its discretion when it authorized plans\nthat limit motorized big game retrieval based on factors other\nthan geography. See 5 U.S.C. § 706(2)(A).\n\n\n 3\n The State notes that additional, non-qualifying big game species\ncan also be found in the Kaibab National Forest, including mule deer,\npronghorn, and black bears.\n\f WILDEARTH GUARDIANS V. PROVENCIO 15\n\n As for the regional office’s skepticism regarding the\nscope of retrieval permitted under the plans, “the fact that a\npreliminary determination by a local agency representative\nis later overruled at a higher level within the agency does not\nrender the decisionmaking process arbitrary and capricious,”\nas agencies are “fully entitled” to “change[] their minds . . .\nas long as the proper procedures were followed.” Nat’l\nAss’n of Home Builders v. Defs. of Wildlife, 551 U.S. 644,\n659 (2007). Such apparent inconsistencies might serve as\nevidence of arbitrariness or capriciousness, see Barnes v.\nU.S. Dep’t of Transp., 655 F.3d 1124, 1134 (9th Cir. 2011)\n(noting that Home Builders “did not hold, however, that such\npreliminary determinations are irrelevant in any context”),\nbut absent other evidence that the plans violated the Travel\nManagement Rule, we do not find these comments\nparticularly enlightening, especially since the Regional\nForester ultimately upheld the designation decisions as to\neach Ranger District. 4\n\n Ultimately, as the district court concluded, “Plaintiffs\nhave only identified dissatisfaction with the ultimate\ndecisions made by the Forest Service in authorizing\n[motorized big game retrieval] in the three ranger districts.”\nAlthough it is not unreasonable to interpret “limited use”\nspatially, Plaintiffs point to no statute, regulation, or ruling\nthat requires a geographic limitation of this sort. The\nDistricts’ plans limit motorized big game retrieval as to\ntiming, qualified species, and number of vehicles, which is\n\n 4\n The same can be said for any inconsistencies between the final\nplans and the guidance provided by the Forest Service’s regional officer,\nto which Plaintiffs point as evidence that the Rule was violated.\nAlthough, for example, the guidance suggested that retrieval of elk not\nbe allowed “between one hour before sunrise and 10:00 am,” that\nlimitation was not a requirement, but merely a recommendation, and so\na failure to implement it does not render the plans unlawful.\n\f16 WILDEARTH GUARDIANS V. PROVENCIO\n\nboth a reasonable interpretation of the Rule and an indication\nthat the Forest Service both considered and applied the\nmandatory minimization criteria. See Mont. Snowmobile,\n790 F.3d at 932. Given that much of the Districts’ land was\nopen to cross-country motorized travel prior to\nimplementation of the plans, we conclude that the new\nrestrictions constitute a “limited” use of motorized vehicles.\n\n B. “Certain”\n\n Next, the Rule allows motorized retrieval on “certain\nforest roads.” 36 C.F.R. § 212.51(b). Plaintiffs argue that\nthe word “certain” requires that it only be allowed on “some,\nbut not all” roads. Because the plans permit retrieval within\none mile of all designated roads, Plaintiffs conclude that they\nare unlawful. The Forest Service, however, correctly notes\nthat while the word “certain” can mean “some, but not all,”\nthe more common definition of the term is “definite” or\n“fixed.” See, e.g., Webster’s Third New International\nDictionary 367 (2002) (listing the primary definitions of\n“certain” as “fixed, settled, stated” and “exact, precise”).\n\n Because the Forest Service limited motor vehicle use to\na defined set of roads in each District, it complied with the\nRule. Even if the proper interpretation of the word “certain”\nwere ambiguous, the Forest Service’s definition of “fixed”\nor “definite” is permissible, consistent with the text of the\nRule, and entitled to deference. See Home Builders,\n551 U.S. at 672 (“An agency’s interpretation of the meaning\nof its own regulations is entitled to deference ‘unless plainly\nerroneous or inconsistent with the regulation.’” (quoting\nAuer v. Robbins, 519 U.S. 452, 461 (1997))).\n\f WILDEARTH GUARDIANS V. PROVENCIO 17\n\n C. “Sparingly”\n\n Lastly, Plaintiffs contend that the Forest Service failed to\nimplement motorized big game retrieval “sparingly.” Travel\nManagement; Designated Routes and Areas for Motor\nVehicle Use, 70 Fed. Reg. 68,264, 68,285 (Nov. 9, 2005)\n(“The Department expects the Forest Service to apply this\nprovision sparingly, on a local or State-wide basis, to avoid\nundermining the purposes of the final rule and to promote\nconsistency in implementation.” (emphasis added)). They\nargue that “allowing motorized big game retrieval off of\nevery single open road is not using [the Forest Service’s]\nauthority sparingly,” and that the Forest Service erroneously\nconcluded that the plans were sparing and limited because\nthey permitted less motorized retrieval than under the prior\npolicies that imposed no restrictions.\n\n This argument is unpersuasive. First, the word\n“sparingly” does not appear in the Rule, but instead in its\npreamble. We look to a preamble only when the regulation\nitself is ambiguous. El Comité para el Bienstar de Earlimart\nv. Warmerdam, 539 F.3d 1062, 1070 (9th Cir. 2008).\nAccordingly, as the Forest Service correctly asserts, the\npreamble does not “impose a duty above and beyond the\nactual terms of the regulation.”\n\n Moreover, Plaintiffs’ argument fails for much the same\nreason as their “limited” argument above: it relies on a\nstrictly spatial conception of “sparingly,” while ignoring the\nother restrictions that the plans impose on motorized big\ngame retrieval. Absent authority requiring a strictly\ngeographic interpretation of the words “limited” and\n“sparingly,” we conclude that the Forest Service did not\nviolate the plain terms of the Travel Management Rule.\n\f18 WILDEARTH GUARDIANS V. PROVENCIO\n\nII. NEPA\n\n A. Standing\n\n As a preliminary matter, Safari Club challenges whether\nPlaintiffs have standing to bring their NEPA claims. It\nobserves that Plaintiffs’ “standing declarant, Kim Crumbo,\nrevealed that his own activities in the forests cause the very\nsame impact that [they] seek to attribute to motorized big\ngame retrieval,” since Crumbo recounted “incidents during\nwhich he, on his bicycle, interfered with young goshawk and\na goshawk pursuing its prey.” Accordingly, Safari Club\nsuggests that Plaintiffs cannot satisfy the redressability\nrequirement for standing, see Salmon Spawning & Recovery\nAll. v. Gutierrez, 545 F.3d 1220, 1226 (9th Cir. 2008),\nbecause Plaintiffs “themselves are a source of” the negative\neffects that they seek to analyze through an EIS, and so “[n]o\nchange in process and no [EIS] analysis of motorized big\ngame retrieval will prevent [Plaintiffs’] members from\nengaging in conduct that is the source of effects” that they\nseek to mitigate.\n\n Under Safari Club’s reasoning, a hypothetical plaintiff\nchallenging an EPA decision on a CO2-emitting power plant\nwould lack standing just because she also happens to exhale\ncarbon dioxide. This result would not only be absurd, but\nalso contrary to our prior precedent, for we have held that\n\n the mere existence of multiple causes of an\n injury does not defeat redressability,\n particularly for a procedural injury. So long\n as a defendant is at least partially causing the\n alleged injury, a plaintiff may sue that\n defendant, even if the defendant is just one of\n multiple causes of the plaintiff’s injury.\n\f WILDEARTH GUARDIANS V. PROVENCIO 19\n\nWildEarth Guardians v. U.S. Dep’t of Agric., 795 F.3d 1148,\n1157 (9th Cir. 2015).\n\n Safari Club also challenges Plaintiffs’ NEPA standing\nbased on their perceived motivation for bringing suit.\nAsserting that Plaintiffs’ “goal in bringing this action was to\nuse NEPA and the NHPA to force [the Forest Service] to\nreverse [its] authorization of motorized big game retrieval,”\nit contends that “NEPA does not provide a cause of action”\nand that Plaintiffs lack prudential standing. Regardless of\nPlaintiffs’ motivation in commencing this suit, however, if a\ngroup is “trying to protect the environment,” then its “suit []\nlies well within NEPA’s zone of interests.” Citizens for\nBetter Forestry v. U.S. Dep’t of Agric., 341 F.3d 961, 976\n(9th Cir. 2003); see also Havasupai Tribe v. Provencio,\n906 F.3d 1155, 1166 (9th Cir. 2018). Accordingly, we\nconclude that Plaintiffs have standing to bring their NEPA\nclaims.\n\n B. Legal Framework\n\n NEPA requires federal agencies, including the Forest\nService, to assess the environmental impact of proposed\nactions that “significantly affect[] the quality of the human\nenvironment.” 42 U.S.C. § 4332(C). It\n\n serves two fundamental objectives. First, it\n “ensures that the agency, in reaching its\n decision, will have available, and will\n carefully consider, detailed information\n concerning significant environmental\n impacts.” And, second, it requires “that the\n relevant information will be made available\n to the larger audience that may also play a\n role in both the decisionmaking process and\n the implementation of that decision.”\n\f20 WILDEARTH GUARDIANS V. PROVENCIO\n\nMont. Snowmobile, 790 F.3d at 924 (citation omitted)\n(quoting Robertson v. Methow Valley Citizens Council,\n490 U.S. 332, 349 (1989)). In short, “NEPA’s purpose is to\nensure that ‘the agency will not act on incomplete\ninformation, only to regret its decision after it is too late to\ncorrect.’” Friends of Clearwater v. Dombeck, 222 F.3d 552,\n557 (9th Cir. 2000) (quoting Marsh v. Or. Nat. Res. Council,\n490 U.S. 360, 371 (1989)). We have admonished that “[w]e\nmust . . . strictly interpret the procedural requirements in\nNEPA . . . ‘to the fullest extent possible’ consistent with the\npolicies embodied in NEPA. ‘[G]rudging, pro forma\ncompliance will not do.’” Churchill County, 276 F.3d at\n1072 (fourth alteration in original) (citation omitted)\n(quoting Lathan v. Brinegar, 506 F.2d 677, 687, 693 (9th\nCir. 1974) (en banc)). As part of this compliance, agencies\nmust ensure “that environmental information is available to\npublic officials and citizens before decisions are made and\nbefore actions are taken. The information must be of high\nquality. Accurate scientific analysis, expert agency\ncomments, and public scrutiny are essential to implementing\nNEPA.” 40 C.F.R. § 1500.1(b).\n\n The Forest Service must prepare an EIS—a more\nthorough undertaking than an EA 5—if an action might\nsignificantly affect environmental quality. As we have\nexplained,\n\n\n\n 5\n “Before deciding whether to complete an EIS, government\nagencies may prepare a less formal EA which ‘briefly provides sufficient\nevidence and analysis for determining whether to prepare an\nenvironmental impact statement or a finding of no significant impact.’”\nAnderson v. Evans, 371 F.3d 475, 488 (9th Cir. 2004) (quoting Tillamook\nCounty v. U.S. Army Corps of Eng’rs, 288 F.3d 1140, 1144 (9th Cir.\n2002)).\n\f WILDEARTH GUARDIANS V. PROVENCIO 21\n\n An EIS must be prepared if “substantial\n questions are raised as to whether a project\n . . . may cause significant degradation of\n some human environmental factor.” Thus, to\n prevail on a claim that the Forest Service\n violated its statutory duty to prepare an EIS,\n a “plaintiff need not show that significant\n effects will in fact occur.” It is enough for the\n plaintiff to raise “substantial questions\n whether a project may have a significant\n effect” on the environment.\n\nBlue Mountains Biodiversity Project v. Blackwood, 161 F.3d\n1208, 1212 (9th Cir. 1998) (alteration in original) (citations\nomitted) (quoting Idaho Sporting Cong. v. Thomas, 137 F.3d\n1146, 1149–50 (9th Cir. 1998)); see also Save the Yaak\nComm. v. Block, 840 F.2d 714, 717 (9th Cir. 1988) (“[A]n\nagency’s decision not to prepare an EIS will be considered\nunreasonable if the agency fails to ‘supply a convincing\nstatement of reasons why potential effects are\ninsignificant.’” (quoting Steamboaters v. FERC, 759 F.2d\n1382, 1393 (9th Cir. 1985))). The significance of an action\ndepends on its context and intensity, the latter of which is\nassessed using a list of criteria enumerated in the relevant\nregulation. See 40 C.F.R. § 1508.27(a)–(b).\n\n C. Whether the Forest Service Needed to Prepare\n EISs\n\n Plaintiffs contend that “the presence of several\nsignificance factors indicating possible significant\nenvironmental consequences of the proposed actions”\nrequired the Forest Service to prepare EISs for each of the\nDistricts’ travel management plans. We analyze in turn each\nconsideration to which they point.\n\f22 WILDEARTH GUARDIANS V. PROVENCIO\n\n i. Impacts\n\n The first enumerated consideration in evaluating an\naction’s intensity concerns “[i]mpacts that may be both\nbeneficial and adverse,” and notes that “[a] significant effect\nmay exist even if the Federal agency believes that on balance\nthe effect will be beneficial.” 40 C.F.R. § 1508.27(b)(1).\n“Significance exists if it is reasonable to anticipate a\ncumulatively significant impact on the environment.\nSignificance cannot be avoided by terming an action\ntemporary or by breaking it down into small component\nparts.” Id. § 1508.27(b)(7).\n\n Plaintiffs suggest that “the travel management plans for\neach Ranger District would have significant direct, indirect,\nand cumulative impacts,” noting that both motorized vehicle\nuse on open, designated roads and cross-country, off-road\nmotorized vehicle use can have “significant detrimental\neffects . . . on a variety of resources.”\n\n The assertion that motorized big game retrieval can have\ndetrimental effects on the environment is consistent with the\ninformation contained in the EAs prepared for the Williams\nand Tusayan Ranger Districts. Those EAs noted that “[t]he\nscientific literature documents a variety of negative effects\nof roads and motorized travel on wildlife,” with\n\n [p]otential direct and indirect effects of roads\n and motorized travel on wildlife includ[ing]\n habitat loss, fragmentation, and degradation\n caused by roads and cross country motorized\n travel; roads can create barriers to\n movements of certain species; animals can be\n killed or injured as a result of being hit or run\n over by motor vehicles; human disturbance\n or harassment of animals caused by or\n\f WILDEARTH GUARDIANS V. PROVENCIO 23\n\n facilitated by motorized travel; [and]\n shooting or harvest of animals facilitated by\n motor vehicle access to wildlife habitats.\n\nThe Williams Ranger District EA further indicated that off-\nroad vehicle use “in areas with sensitive or moist soils can\ncreate tracks, ruts and new user routes that may crush,\ndisplace, and/or destroy cultural materials (i.e. artifacts,\nfeatures, traditionally used plants), and damage significant\ninformation that may contribute to our understanding of\nhistory.” A particularly vexatious problem related to\nmotorized vehicle use is the spread of invasive weeds. Each\nof the three EAs noted that vehicles are a common cause of\nweed introduction and spread, with the North Kaibab Ranger\nDistrict EA reporting that “[t]he authorization of motorized\nbig game retrieval will have an increased threat of invasive\nspecies spread as every vehicle that travels cross-country has\nthe ability to serve as a vector and create disturbance.” 6\n\n We do not disagree with Plaintiffs’ assertion that\nmotorized big game retrieval can have a negative effect on\nthe environment. But we nevertheless conclude that the\nenvironmental impacts discussed in the EAs did not raise\nsubstantial concerns that necessitated the preparation of\nEISs. Plaintiffs might disagree with the Forest Service’s\nsubstantive conclusions, but we see no indication that the\nagency failed to satisfy NEPA’s procedural requirements—\na crucial distinction that is lost in Plaintiffs’ formulation of\nthe issue.\n\n\n 6\n This is because, as explained in the Williams Ranger District EA,\n“[v]ehicles driving through populations of invasive plants often get\nseed[s] entrapped in tire tread or undercarriages, move to another area\nand then drop seeds into a previously uninfested area.”\n\f24 WILDEARTH GUARDIANS V. PROVENCIO\n\n Plaintiffs’ treatment of the spread of invasive weeds is\nillustrative. They tie the problem of weeds to another\nenumerated NEPA consideration that implicates “[u]nique\ncharacteristics of the geographic area such as proximity to\nhistoric or cultural resources, park lands, prime farmlands,\nwetlands, wild and scenic rivers, or ecologically critical\nareas.” Id. § 1508.27(b)(3). They note that “both the North\nKaibab and Tusayan Ranger Districts immediately abut\nGrand Canyon National Park.” The EAs agreed: they\ndescribed Grand Canyon National Park as “internationally\nimportant,” and noted the Tusayan Ranger District’s “unique\nlocation” as “a gateway to one of the most famous national\nparks in the country.” In that District’s EA, the Forest\nService acknowledged that recreation and hunting—\nactivities facilitated by the travel management plan—“have\nthe potential to introduce exotic plants” that “may then\nspread to adjoining lands, including the Grand Canyon\nNational Park.” The National Park Service, in a letter from\nthe Acting Park Superintendent of Grand Canyon National\nPark, advised the Forest Service to “institute a buffer zone\nof 1-mile along the park boundary for any purpose including\nbig-game retrieval, fuel-wood gathering, cross-county\ntravel, etc.,” due to “increased pressure from motorized\nvehicles at or near the southern park boundary over the past\nseveral years”—a recommendation that was not adopted in\nthe Tusayan Ranger District DN/FONSI.\n\n Clearly, the EAs demonstrated that motorized big game\nretrieval risks the spread of invasive weeds, an undeniable\nenvironmental impact. In response, the Forest Service relies\nin part on questionable reasoning by focusing on the fact that\nthe plans reduced negative impacts when compared with\npre-plan activity. It notes that the plan eventually selected\nfor the Tusayan Ranger District opted to limit the number of\nroads open to the public, which, the EA noted, “reduces the\n\f WILDEARTH GUARDIANS V. PROVENCIO 25\n\nnumber of opportunities for noxious and invasive exotic\nweeds to be introduced and spread.” “Thus,” the Forest\nService concludes, “contrary to [Plaintiffs’] argument, the\nTusayan EA confirms that the Forest Service considered the\nissue and reasonably concluded that the decision will reduce,\nnot increase, the spread of exotic plants.”\n\n However, a conclusion, even a correct one, that a given\naction might reduce a potential impact does not alone\nindicate that the impact would not be significant. We have\nnoted that the use of baselines is a helpful, and perhaps\ninevitable, tool in conducting environmental surveys. See\nOr. Nat. Desert Ass’n v. Jewell, 840 F.3d 562, 568 (9th Cir.\n2016) (“The establishment of a ‘baseline is not an\nindependent legal requirement, but rather, a practical\nrequirement in environmental analysis often employed to\nidentify the environmental consequences of a proposed\nagency action.’” (quoting Am. Rivers v. FERC, 201 F.3d\n1186, 1195 n.15 (9th Cir. 1999))). But we have also\ndetermined that “simply because the Final Rule may be an\nimprovement over the [previous] standard does not\nnecessarily mean that it will not have a ‘significant effect’\non the environment” where the agency “has not explained\nwhy its rule will not have a significant effect.” Ctr. for\nBiological Diversity v. NHTSA, 538 F.3d 1172, 1224 (9th\nCir. 2008). The plans chosen by the Forest Service might\nreduce (even substantially reduce) the spread of noxious\nweeds in the Districts, but that alone does not address\nwhether that reduced impact itself has a significant impact\non the environment generally and Grand Canyon National\nPark in particular. Therefore, the Forest Service cannot rely\nsolely on the reduction of adverse impacts to demonstrate\nthat those impacts are not significant.\n\f26 WILDEARTH GUARDIANS V. PROVENCIO\n\n But that the Forest Service occasionally conflates\nreduction with insignificance does not necessarily mean that\nit violated NEPA. To demonstrate, we once again go into\nthe weeds.\n\n The Forest Service acknowledged that the North Kaibab\nRanger District contained “several species of invasive\nweeds,” which “are spread [] via roads and forest visitors.”\nThe plan that was eventually selected “reduce[d] the number\nof roads that can be traveled on by 376 miles,” which, the\nEA found, would “lower the amount of invasive species seed\nintroduced or spread.” But notably, the EA continued:\n\n The authorization of motorized big game\n retrieval will have an increased threat of\n invasive species spread as every vehicle that\n travels cross-country has the ability to serve\n as a vector and create disturbance.\n Alternative 2 [the selected plan] authorizes\n motorized big game retrieval for only elk and\n mule deer. This is expected to lead to only a\n small increase in the potential for invasive\n species spread and disturbance when\n compared to Alternative 3 and should not\n generate any realistic impacts.\n\nThis passage demonstrates that the Forest Service did not\nmerely rely on the possibility of reduction, but also\nconcluded that the plan would “not generate any realistic\nimpacts”; in other words, that the effects would not be\nsignificant. Furthermore, as the district court noted, the EA\nmentioned that the North Kaibab Ranger District featured\n“projects focus[ed] on treating known infestations across the\nDistrict, prioritizing the species and locations that pose the\ngreatest threats,” methods that had “proven successful in\n\f WILDEARTH GUARDIANS V. PROVENCIO 27\n\neradicating or reducing potentially serious noxious species\nthreats.” This language indicates that the Forest Service\nacknowledged a potential environmental impact, and then\ndetermined that, due to features of the travel management\nplan and other remediation efforts, it was unlikely to be\nsignificant. Contrary to Plaintiffs’ argument, the Forest\nService did not merely determine that the problem would be\nreduced; it also concluded that the impact would not be\nsignificant. Plaintiffs disagree with the EA’s conclusion, but\nthis is “a classic example of a factual dispute the resolution\nof which implicates substantial agency expertise.” Marsh,\n490 U.S. at 376. We agree with the district court: “Plaintiffs’\nsingular and conclusory statement that exotic plants might\nspread . . . does not raise substantial questions that would\ntrigger the need for an EIS.”\n\n Similarly, although the Forest Service did not follow all\nof the recommendations made by Grand Canyon National\nPark’s Acting Park Superintendent, this fact does not mean\nthat it ignored a significant environmental impact. Agencies\ncan thoughtfully consider suggestions but ultimately decide\nto reject them, and the presence of an articulated concern\ndoes not alone trigger the need to conduct an EIS. See Native\nEcosystems Council v. U.S. Forest Serv., 428 F.3d 1233,\n1240 (9th Cir. 2005) (“NEPA permits a federal agency to\ndisclose [] impacts without automatically triggering the\n‘substantial questions’ threshold.”). Here, the record\nindicates that the Forest Service explained why a buffer\nwould not be employed (“We don’t expect to use a buffer\nzone as many management actions will need to extend to the\nForest Service-National Park boundary”), and further\narticulated means of remedying the risk of illegal motor\nvehicle use (such as “limit[ing] motorized big game retrieval\nduring all elk seasons” and “work[ing] closely with Arizona\nGame and Fish Department to monitor and enforce illegal\n\f28 WILDEARTH GUARDIANS V. PROVENCIO\n\ncross-country travel associated with hunting activities”). We\nfind no indication in the record that the Forest Service did\nnot adequately consider potential impacts, or that substantial\nquestions remained that required the preparation of an EIS.\n\n The same conclusion ultimately applies to all of the\nenvironmental impacts that Plaintiffs highlight in their\nbriefs: although Plaintiffs disagree with the EAs’ factual\nconclusions, the Forest Service nonetheless considered the\nissues, gave them the requisite “hard look,” and thus fulfilled\ntheir NEPA obligations. Save the Yaak, 840 F.2d at 717. In\nreaching its conclusions that none of the impacts cited by\nPlaintiffs were sufficiently significant to require the\npreparation of EISs, the Forest Service did not “rel[y] on\nfactors Congress did not intend it to consider, ‘entirely fail[]\nto consider an important aspect of the problem,’ or offer[] an\nexplanation ‘that runs counter to the evidence before the\nagency or is so implausible that it could not be ascribed to a\ndifference in view or the product of agency expertise.’”\nLands Council v. McNair, 537 F.3d 981, 987 (9th Cir. 2008)\n(en banc) (quoting Earth Island Inst. v. U.S. Forest Serv.,\n442 F.3d 1147, 1157 (9th Cir. 2006)), overruled on other\ngrounds by Winter v. Nat. Res. Def. Council, Inc., 555 U.S.\n7 (2008). Instead, the evidence in the record indicates that,\nalthough the EAs acknowledged that motorized big game\nretrieval might have negative impacts on the environment,\nthe Forest Service’s determination that these impacts would\nnot be significant evinced “a rational connection between the\nfacts found and the conclusions made.” Or. Nat. Res.\nCouncil v. Lowe, 109 F.3d 521, 526 (9th Cir. 1997).\nTherefore, its conclusions were not arbitrary and capricious,\nand the Forest Service did not violate NEPA by declining to\nprepare EISs based on the plans’ environmental impacts.\n\f WILDEARTH GUARDIANS V. PROVENCIO 29\n\n ii. Controversy and Uncertainty\n\n NEPA also requires the preparation of an EIS when an\naction’s “effects on the quality of the human environment\nare likely to be highly controversial,” and/or “are highly\nuncertain or involve unique or unknown risks.” 40 C.F.R.\n§ 1508.27(b)(4)–(5). “A project is ‘highly controversial’ if\nthere is a ‘“substantial dispute [about] the size, nature, or\neffect of the major Federal action rather than the existence\nof opposition to a use.”’” Native Ecosystems Council,\n428 F.3d at 1240 (alteration in original) (emphases added)\n(quoting Blue Mountains, 161 F.3d at 1212); see also\nWetlands Action Network v. U.S. Army Corps of Eng’rs,\n222 F.3d 1105, 1122 (9th Cir. 2000) (“The existence of\nopposition to a use, however, does not render an action\ncontroversial.”), abrogated on other grounds by Wilderness\nSoc’y v. U.S. Forest Serv., 630 F.3d 1173 (9th Cir. 2011).\n“A substantial dispute exists when evidence, raised prior to\nthe preparation of an EIS or FONSI casts serious doubt upon\nthe reasonableness of an agency’s conclusions.” Nat’l Parks\n& Conservation Ass’n v. Babbitt, 241 F.3d 722, 736 (9th Cir.\n2001) (citation omitted), abrogated on other grounds by\nMonsanto Co. v. Geertson Seed Farms, 561 U.S. 139 (2010).\nAdditionally, because “[a]n agency must generally prepare\nan EIS if the environmental effects of a proposed agency\naction are highly uncertain,” such preparation “is mandated\n‘where uncertainty may be resolved by further collection of\ndata, or where the collection of such data may prevent\nspeculation on potential . . . effects.’” Barnes, 655 F.3d at\n1140 (second alteration in original) (quoting Native\nEcosystems Council, 428 F.3d at 1240). However, NEPA\nregulations “do not anticipate the need for an EIS anytime\nthere is some uncertainty, but only if the effects of the project\nare ‘highly’ uncertain.” Envtl. Prot. Info. Ctr. v. U.S. Forest\nServ., 451 F.3d 1005, 1011 (9th Cir. 2006).\n\f30 WILDEARTH GUARDIANS V. PROVENCIO\n\n Plaintiffs contend that “the travel management plans for\nthe Tusayan, Williams, and North Kaibab Ranger Districts\npresent highly controversial and highly uncertain effects that\ninvolve unique or unknown risks,” because “significant\ncontroversy exists as to the amount and type of motorized\nrecreation that would be allowed across the three Ranger\nDistricts.” But Plaintiffs point to nothing in the record\nindicating the existence of a substantial dispute that casts\ndoubt on the Forest Service’s conclusions about\nenvironmental impacts. There may have been opposition to\nthe plans, but mere opposition alone is insufficient to support\na finding of controversy. The Forest Service “recognize[d]\nthat elements of the Selected Alternative [] generated\ncontroversy,” but concluded—apparently correctly—that\nthere was “no substantiated scientific controversy over the\neffects as described.”\n\n Plaintiffs assert that various questions raised during the\nEA process revealed a high level of scientific uncertainty,\nbut the record belies that assertion. For example, as to the\npotential risks “based on the broad allowance of motorized\nbig game retrieval,” the Forest Service attempted to estimate\nhunting activity based on past data, and used this information\nto conclude that the impacts of motorized big game retrieval\nwould be limited. Plaintiffs identity nothing in the record to\nsuggest that the Forest Service’s estimates were unduly\nspeculative, or that it unreasonably relied upon these\npredictions. Plaintiffs also argue that there was “uncertainty\nregarding whether or not hunters will actually remove gut\npiles” when retrieving carcasses, which they must do “to\nprotect California condors from lead poisoning.” But\nalthough the Forest Service acknowledged that this issue\nmight present a problem, the record also indicates that it\nconsidered the issue and reasonably concluded that it was\nunlikely to significantly impact the North Kaibab Ranger\n\f WILDEARTH GUARDIANS V. PROVENCIO 31\n\nDistrict’s condors because the Arizona Game and Fish\nDepartment had provided to hunters, among other\nincentives, lead-free ammunition. The Forest Service also\nnoted that “there would be decreased risk of human\ndisturbance of scavenging condors as a result of a reduced\nopen road system and substantially restricted motorized\ncross-country travel,” and concluded that the North Kaibab\nRanger District plan “is not likely to jeopardize the\ncontinued existence [of] California condors.” Plaintiffs\nneither challenge nor address these conclusions, and instead\nrely on the mere existence of potential problems as evidence\nof significant uncertainty—a tactic that does not pass muster.\nSee Native Ecosystems Council, 428 F.3d at 1240 (“Simply\nbecause a challenger can cherry pick information and data\nout of the administrative record to support its position does\nnot mean that a project is highly controversial or highly\nuncertain.”).\n\n iii. Precedent for Future Actions\n\n Another consideration for measuring an action’s\nintensity for NEPA purposes is “[t]he degree to which the\naction may establish a precedent for future actions with\nsignificant effects or represents a decision in principle about\na future consideration.” 40 C.F.R. § 1508.27(b)(6).\nAlthough “EAs are usually highly specific to the project and\nthe locale, thus creating no binding precedent,” In Def. of\nAnimals v. U.S. Dep’t of Interior, 751 F.3d 1054, 1071 (9th\nCir. 2014) (quoting Barnes, 655 F.3d at 1140), and the Forest\nService here explicitly found that each of the three travel\nmanagement plans was “not likely to establish a precedent\nfor future actions with significant effects,” Plaintiffs assert\nthat this consideration is nonetheless implicated because\n“the Forest Service has [] made public statements indicating\n\f32 WILDEARTH GUARDIANS V. PROVENCIO\n\nthat what the Kaibab National Forest does, so too will other\nsouthwestern National Forests.”\n\n It is true that the record contains evidence to this effect—\nincluding that the Coconino National Forest “will defer to\nthe neighboring Kaibab National Forest’s policy for\n[motorized big game retrieval] in units shared with the\nWilliams Ranger District, regardless of how the Coconino\nproposes to apply the Travel Management Rule,” and that\nthe Prescott National Forest will “match them as best as we\ncan”—but that does not mean that the Districts’ plans bind\nor necessarily shape other forests’ plans in such a way that\nthey should be considered precedential, especially since any\nother forest’s plan would be subject to its own NEPA\nanalysis. Cf. Sierra Club v. Marsh, 769 F.2d 868, 879 (1st\nCir. 1985) (determining that an action was precedential\nbecause “once Maine completes the causeway and port,\npressure to develop the rest of the island could well prove\nirreversible”). In Oregon Wild v. Bureau of Land\nManagement, a district court dealt with a similar situation,\nwhere a project was “part of a larger series of ‘pilot projects’\naimed at ‘inform[ing] long-term planning’ for management\nof [] lands in Oregon and California.” No. 6:14-CV-0110-\nAA, 2015 WL 1190131, at *9 (D. Or. Mar. 14, 2015) (first\nalteration in original). There, as here, the most that could be\nconcluded from such a minor precedential effect is that this\nconsideration “supports the conclusion that an EIS is\nnecessary”—but “the precedential factor alone is not\ndispositive.” Id.; see also Anderson v. Evans, 371 F.3d 475,\n493 (9th Cir. 2004) (finding this factor “insufficient on its\nown to demonstrate a significant environmental impact”\nwhere an action is merely influential and not binding). Thus,\nthis consideration alone did not require preparation of an\nEIS.\n\f WILDEARTH GUARDIANS V. PROVENCIO 33\n\n iv. Threatened Species\n\n Finally, there is the issue of the Mexican spotted owl, a\nthreatened species found in the Williams and North Kaibab\nRanger Districts. The Forest Service must consider “[t]he\ndegree to which the action may adversely affect an\nendangered or threatened species or its habitat that has been\ndetermined to be critical under the Endangered Species Act\nof 1973.” 40 C.F.R. § 1508.27(b)(9). Plaintiffs point to\nportions of the record that raise the specter of adverse\neffects—including consultation letters between the Williams\nand North Kaibab Ranger Districts and the U.S. Fish and\nWildlife Service regarding the owls—but they again ignore\nthe Forest Service’s overall conclusions. For example, the\nWilliams Ranger District EA reported that some roads would\npass through Mexican spotted owl critical habitats, which\nmight result in “increased potential human disturbance.”\nBut the EA ultimately concluded that the selected plan\nwould be “primarily beneficial,” and would “not adversely\naffect Mexican spotted owl or Mexican spotted owl\ndesignated Critical Habitat.” In a separate biological\nassessment, the Forest Service concluded that “[t]he effects\ndetermination for Mexican spotted owl and Mexican spotted\nowl critical habitat is may affect, not likely to adversely\naffect,” based on the\n\n determination [] that potential effects of the\n proposed action on the Mexican spotted owl\n would be primarily beneficial. Closing roads\n and restricting motorized cross-country\n travel under the proposed action would result\n in reduced motorized access to spotted owl\n habitat compared to current management and\n thus reduced risk of human disturbance to\n spotted owls, reduced impacts to habitat of\n\f34 WILDEARTH GUARDIANS V. PROVENCIO\n\n spotted owls and their small mammal prey\n species, and reduced impacts to designated\n critical habitat.\n\nNotably, the U.S. Fish and Wildlife Service concurred in this\ndetermination.\n\n In short, although the Forest Service did not definitively\nconclude that no Mexican spotted owls would be adversely\naffected by the Districts’ travel management plans, the\nrecord indicates that they sufficiently considered the issue\nand arrived at a reasonable conclusion that the effects would\nnot be significant, thus obviating the need for an EIS.\n\n v. Summation\n\n In the end, we conclude that the Forest Service’s\ndetermination that no EISs were needed as to the Districts’\ntravel management plans was reasonable. The plans might\nhave some precedential effect, there is a possibility that\nMexican spotted owls might be affected, and exotic weeds\nmight be spread by motorized big game retrieval, but the\nrecord ultimately supports the Forest Service’s conclusion\nthat these concerns do not rise to the level of significance\nthat would require EISs. The Forest Service gave the\nrequisite hard look and made determinations that were\nneither arbitrary nor capricious, and were consistent with the\nevidence before it. Absent substantial questions that would\nhave mandated EISs, the Forest Service did not violate\nNEPA. 7\n\n\n 7\n Two other enumerated considerations are relevant in this case:\n“[t]he degree to which the action may adversely affect . . . objects listed\nin or eligible for listing in the National Register of Historic Places or may\n\f WILDEARTH GUARDIANS V. PROVENCIO 35\n\nIII. NHPA\n\n Finally, Plaintiffs contend that the Forest Service\nviolated the NHPA by failing to identify and evaluate the\n“high density of cultural resources” that might be damaged\nas a result of motorized travel in the Districts.\n\n The NHPA’s purpose is to “foster conditions under\nwhich our modern society and our historic property can exist\nin productive harmony,” 54 U.S.C. § 300101(1), and it\nrequires federal agencies to “make a reasonable and good\nfaith effort” to identify historic properties that might be\naffected by an action, and to “take [those potential effects]\ninto account.” 36 C.F.R. § 800.4(b)(1); see also 54 U.S.C.\n§ 306108. 8 “Like NEPA, ‘[s]ection 106 of NHPA is a “stop,\nlook, and listen” provision that requires each federal agency\nto consider the effects of its programs.’” Te-Moak Tribe of\nW. Shoshone of Nev. v. U.S. Dep’t of Interior, 608 F.3d 592,\n\n\n\n\ncause loss or destruction of significant scientific, cultural, or historical\nresources,” and “[w]hether the action threatens a violation of Federal,\nState, or local law or requirements imposed for the protection of the\nenvironment.” 40 C.F.R. § 1508.27(b)(8), (10). However, the former\nconsideration essentially depends on whether the Forest Service\ncomplied with the NHPA, while the latter hinges on the Forest Service’s\ncompliance with both the NHPA and the Travel Management Rule.\nBecause we conclude that the Forest Service complied with both, we also\nconclude that no federal laws were violated, and no cultural resources\nadversely affected, such that EISs were required based on these\nconsiderations.\n\n 8\n “Historic property means any prehistoric or historic district, site,\nbuilding, structure, or object included in, or eligible for inclusion in, the\nNational Register of Historic Places maintained by the Secretary of the\nInterior.” 36 C.F.R. § 800.16(l)(1).\n\f36 WILDEARTH GUARDIANS V. PROVENCIO\n\n607 (9th Cir. 2010) (quoting Muckleshoot Indian Tribe v.\nU.S. Forest Serv., 177 F.3d 800, 805 (9th Cir. 1999)).\n\n Plaintiffs assert that the Forest Service violated the\nNHPA in three ways: (1) by failing “to make a reasonable\nand good faith effort to identify and evaluate cultural\nproperties”; (2) by “erroneously determin[ing] that\n‘Exemption Q’ excuses NHPA consultation for the Tusayan\nand Williams Ranger Districts”; and (3) by arbitrarily\nmaking “a ‘no adverse effect’ determination for the North\nKaibab Ranger District after admitting that cross-country\ntravel damages cultural resources.” We consider each\nargument in turn.\n\n A. Identification of Cultural Properties\n\n Section 106 of the NHPA requires the Forest Service to\n“make a reasonable and good faith effort to identify historic\nproperties; determine whether identified properties are\neligible for listing on the National Register . . .; [and] assess\nthe effects of the undertaking on any eligible historic\nproperties found.” Te-Moak Tribe, 608 F.3d at 607 (first\nalteration in original) (quoting Muckleshoot Indian Tribe,\n177 F.3d at 805). It must also engage in consultation with\nthe State Historic Preservation Officer (SHPO) to\n“[d]etermine and document the area of potential effects,”\n“[g]ather information,” and “develop and evaluate\nalternatives or modifications to the undertaking that could\navoid, minimize, or mitigate adverse effects on historic\nproperties.” 36 C.F.R. §§ 800.4(a), 800.6(a).\n\n Plaintiffs principally argue that the Forest Service failed\nto “make a reasonable and good faith effort” because it did\nnot complete 100 percent surveys of potentially affected\nareas. The First Amended Programmatic Agreement\nRegarding Historic Property Protection and Responsibilities\n\f WILDEARTH GUARDIANS V. PROVENCIO 37\n\n(Programmatic Agreement), which the parties agree\nprovided the relevant guidelines, mandated that “[t]he level\nof need and extent of new field surveys or inspections will\nbe proposed by a Professional Cultural Resource Specialist\nand approved by the Forest Archaeologist based on the\nguidelines provided in this section.” Those guidelines\nincluded the use of “relevant information to assess the\npotential to affect historic properties and the expected nature\nand distribution of heritage properties that may be affected”;\nnamely, “[t]he expected nature and severity of all associated\nimpacts” and “[t]he expected nature and distribution of\nheritage resources.” Based on the results of this “prefield\nresearch,” the Resource Specialist and Forest Archeologist\nwere directed to “determine the relative level of field survey\nto be conducted.” The Programmatic Agreement called for\n“100% surveys” where “site density is expected to be high”\nand where “site densities are unknown and expected visitor\nuse or impacts will be high.” By contrast, “areas may be\nsurveyed at less than 100%” where “known site density is\nlow.”\n\n Here, the record supports the Forest Service’s conclusion\nthat the Programmatic Agreement did not require\n100 percent surveys. At the time the travel management\nplans were decided, the Forest Service had surveyed 42\npercent of the Williams Ranger District, 23 percent of the\nTusayan Ranger District, and 25 percent of the North Kaibab\nRanger District. It concluded that no further surveying was\nrequired, based on the expected density of cultural resources\nand the nature and severity of impacts upon them.\n\n The record reinforces the determination that the expected\nimpacts of the travel management plans would be low. After\nconsidering historic hunting data, the EAs concluded that\nless than 0.1 percent of each District’s acreage would be\n\f38 WILDEARTH GUARDIANS V. PROVENCIO\n\nimpacted by motorized big game retrieval. However, the\nfact that “[t]he expected nature and severity of all associated\nimpacts” might have been low does not necessarily mean\nthat the Forest Service did not need to conduct 100 percent\nsurveys, for the Programmatic Agreement required\n100 percent surveys for high-density sites regardless of\nimpact level. Accordingly, the Forest Service’s obligation\nrested on “[t]he expected nature and distribution of heritage\nresources.”\n\n The record is somewhat unclear as to the density of\nheritage resources in the areas open to motorized big game\nretrieval. Plaintiffs note that many of the Districts’ cultural\nresources are close to the roads, but that does not necessarily\nspeak to the density of resources because the Programmatic\nAgreement does not provide any benchmark or guidance as\nto what constitutes a high-density site. Ultimately, the lack\nof clarity weighs in the Forest Service’s favor: it was not\nunreasonable for it to determine that the density was low or\nunknown, either of which would have excused 100 percent\nsurveys. 9 Thus, the Forest Service followed a reasonable\ninterpretation of the Programmatic Agreement.\n\n B. Exemption Q\n\n The Programmatic Agreement’s Exemption Q provided\nthat “[a]ctivities not involving ground or surface disturbance\n(e.g., timber stand improvement and precommercial thinning\nby hand)” are “exempt from further review and\nconsultation.” Plaintiffs argue that the Forest Service\n 9\n The Programmatic Agreement required 100 percent surveys\n“where site densities are unknown and expected visitor use or impacts\nwill be high,” so it would have been reasonable for the Forest Service to\nconclude that 100 percent surveys were not needed where densities were\nunknown and use impacts were low.\n\f WILDEARTH GUARDIANS V. PROVENCIO 39\n\narbitrarily relied on Exemption Q to excuse review of\nmotorized big game retrieval in the Williams and Tusayan\nRanger Districts.\n\n Once again, however, the record does not support\nPlaintiffs’ assertions. We agree with Plaintiffs that invoking\nExemption Q would have been inappropriate here. It is\nclear, and the Forest Service does not dispute, that motorized\ntravel causes surface disturbance. But although the Forest\nService’s correspondence with non-party Center for\nBiological Diversity suggested that it applied Exemption\nQ, 10 the record as a whole supports a contrary conclusion.\nThe Forest Service consulted with both the Arizona SHPO\nand potentially affected tribes as to each District’s travel\nmanagement plan—consultations that would not have been\nrequired if Exemption Q had been applied. Moreover, the\nForest Service made no other references to Exemption Q as\npart of the travel management plan decision process, which\nfurther supports the conclusion that it was not in fact\ninvoked. Accordingly, the references to Exemption Q at\n\n 10\n The Forest Service claims that Plaintiffs’ “brief does not identify\nwhere or how the Tusayan and Williams decisions purportedly relied on\nExemption Q,” and that their “responses to the non-party Center for\nBiological Diversity’s administrative appeals of those decisions” contain\nonly “an isolated reference to Exemption Q,” but these assertions are\nmisleading. In its response to the appeal regarding the Tusayan Ranger\nDistrict, the Forest Service wrote that it “determined that motorized big\ngame retrieval fell under Exemption Q of the PA, ‘Activities not\ninvolving ground or surface disturbance (e.g., timber stand improvement\nand precommercial thinning by hand)’ and would have limited impacts\nsimilar to the examples cited in the exemption in the” Programmatic\nAgreement. That is more than a mere “isolated reference,” as it implies\nthat the Forest Service actually applied Exemption Q to the Tusayan\nRanger District. This same language invoking Exemption Q appeared in\nthe Forest Service’s response regarding the Williams Ranger District.\n\f40 WILDEARTH GUARDIANS V. PROVENCIO\n\nmost amounted to harmless error, as they had no effect on\nthe NHPA consultation process. See 5 U.S.C. § 706\n(requiring the court to take “due account . . . of the rule of\nprejudicial error” when conducting APA review); Tucson\nHerpetological Soc’y v. Salazar, 566 F.3d 870, 880 (9th Cir.\n2009) (“We have held that the harmless error doctrine ‘may\nbe employed only “when a mistake of the administrative\nbody is one that clearly had no bearing on the procedure\nused or the substance of decision reached.”’” (quoting\nGifford Pinchot Task Force v. U.S. Fish & Wildlife Serv.,\n378 F.3d 1059, 1071 (9th Cir. 2004))). 11\n\n C. “No Adverse Effect” Determination\n\n Finally, Plaintiffs argue that the Forest Service’s\nconclusion that motorized big game retrieval would have no\nadverse effect on cultural resources was arbitrary. But to\nbuttress this assertion, Plaintiffs again cherry pick isolated\nsegments from the record without considering their broader\ncontext. For example, the Cultural Resources Specialist\nReport prepared for the North Kaibab Ranger District noted\nthat “[c]ross country motorized travel, whether to retrieve\ngame or for other purposes, can adversely affect cultural\nresource sites if a vehicle is driven across a site,” since\n“[v]ehicles can [] crush or displace artifacts and features\n\n 11\n Curiously, although the Forest Service similarly argued in the\ndistrict court that “the agency’s decisions did not rely on Exemption Q,”\nwe note that the court concluded that Exemption Q did apply. On appeal,\nthe Forest Service does not argue that the district court’s conclusion on\nthis point was correct, and we can affirm the court’s ruling even though\nwe agree with the parties that Exemption Q was not in fact applicable.\nSee Applied Underwriters, Inc. v. Lichtenegger, 913 F.3d 884, 892 (9th\nCir. 2019) (“We will affirm the district court’s correct legal results, even\nif reached for the wrong reasons.” (quoting Alcaraz v. Block, 746 F.2d\n593, 602 (9th Cir. 1984))).\n\f WILDEARTH GUARDIANS V. PROVENCIO 41\n\nimpacting the physical integrity of the site and impairing or\ndestroying scientific information that may contribute to the\nunderstanding of the history and prehistory of an area.” But\nthat same report also concluded as follows:\n\n Quantifying the potential for damage from\n big game retrieval is difficult. The results\n vary depending on the number of game\n retrieval trips annually, the location of those\n retrievals (high site probability areas versus\n low probability), site types found in the area,\n soil characteristics, routes used to access the\n game and weather conditions at the time of\n retrieval. However, the fewer number of\n motorized trips that occur, the lower the\n likelihood of encountering and impacting a\n site. . . .\n\n Limiting cross-country travel will have a\n beneficial effect on cultural resources by\n reducing the potential for sites to be\n damaged. This alternative would restrict\n motorized big game retrieval to elk and\n bison. In 2009, 38 buffalo and no elk were\n taken . . . . While there is a possibility that\n cross-country game retrieval of either of\n these species could impact a cultural resource\n site, given the low number of takes each year,\n it is anticipated that the potential for adverse\n effects to a site would be negligible:\n 38 entries per year equates to less than\n .0099% of the acreage on the NKRD. The\n odds of adversely affecting a cultural\n resource site under these conditions are\n extremely low.\n\f42 WILDEARTH GUARDIANS V. PROVENCIO\n\n(emphases added). In light of this ultimate conclusion, as\nwell as the implementation of the suggested mitigation\nmeasures, it was not arbitrary for the Forest Service to\nconclude, based on the evidence before it, that adverse\neffects were unlikely. See Motor Vehicle Mfrs. Ass’n of\nU.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43\n(1983) (“[A]n agency rule would be arbitrary and capricious\nif the agency . . . offered an explanation for its decision that\nruns counter to the evidence before the agency.”).\n\n D. Summation\n\n The NHPA—and NEPA—“create obligations that are\nchiefly procedural in nature.” San Carlos Apache Tribe,\n417 F.3d at 1097 (quoting Pres. Coal., Inc. v. Pierce,\n667 F.2d 851, 859 (9th Cir. 1982)). Even if cultural\nresources might be harmed as a result of motorized big game\nretrieval, that fact alone does not indicate that the Forest\nService violated the NHPA. The Forest Service conducted\nthe required prefield work, consulted with the appropriate\nentities, and reached a determination consistent with the\nevidence before it—in short, satisfied its procedural\nobligations.\n\n CONCLUSION\n\n The travel management plans that the Forest Service\nimplemented in the three Ranger Districts of the Kaibab\nNational Forest limited motorized big game retrieval to\ncertain roads and imposed additional restrictions to reduce\nthe level and effect of motorized activity. In crafting the\nplans, the Forest Service investigated potential impacts on\nboth the environment and historic properties and reasonably\ndetermined that no further action was needed. Accordingly,\nwe conclude that the Forest Service followed the Travel\n\f WILDEARTH GUARDIANS V. PROVENCIO 43\n\nManagement Rule and fulfilled its procedural obligations\nunder NEPA and the NHPA. 12\n\n AFFIRMED.\n\n\n\n\n 12\n Amicus Curiae Rocky Mountain Elk Foundation filed a motion\nfor leave to file an amicus brief in support of the Forest Service. We\nconclude that the proposed brief provides neither legal nor factual\nsupport to help resolve the issues on appeal, and so deny the motion.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Wildearth Guardians v. Heather Provencio\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-13\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nWILDEARTH GUARDIANS; GRAND No. 17-17373\\nCANYON WILDLANDS COUNCIL;\\nWILDLANDS NETWORK; SIERRA D.C. No.\\nCLUB, 3:16-cv-08010-\\n Plaintiffs-Appellants, SMM\\n\\n v.\\n OPINION\\nHEATHER PROVENCIO, in her official\\ncapacity as Kaibab National Forest\\nSupervisor; UNITED STATES FOREST\\nSERVICE,\\n Defendants-Appellees,\\n\\n and\\n\\nSTATE OF ARIZONA, on behalf of\\nArizona Department of Game and\\nFish; SAFARI CLUB INTERNATIONAL,\\n Intervenors-Defendants-Appellees.\\n\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Stephen M. McNamee, Senior District Judge, Presiding\\n\\n Argued and Submitted February 7, 2019\\n Phoenix, Arizona\\n\\f2 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n Filed March 13, 2019\\n\\n Before: MICHAEL DALY HAWKINS, MILAN D.\\n SMITH, JR., and ANDREW D. HURWITZ,\\n Circuit Judges.\\n\\n Opinion by Judge Milan D. Smith, Jr.\\n\\n\\n SUMMARY *\\n\\n\\n Environmental Law\\n\\n The panel affirmed the district court’s summary\\njudgment in favor of the United States Forest Service in an\\naction by plaintiff environmental groups challenging travel\\nmanagement plans implemented by the Forest Service to\\npermit limited motorized big game retrieval in three Ranger\\nDistricts of the Kaibab National Forest.\\n\\n The Travel Management Rule, promulgated by the U.S.\\nDepartment of Agriculture for Forest Service lands,\\ngenerally prohibits off-road, motorized travel, but permits\\nthe “limited” use of motor vehicles within a specified\\ndistance of “certain” forest roads for the purposes of\\ncamping or retrieval of downed big game animals. The\\npanel rejected plaintiffs’ contention that the Forest Service\\nviolated the Travel Management Rule by implementing\\nplans that did not sufficiently limit motorized big game\\nretrieval in the Ranger Districts. The panel concluded that\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 3\\n\\nthe Forest Service did not violate the plain terms of the\\nTravel Management Rule.\\n\\n Addressing plaintiffs’ claims under the National\\nEnvironmental Policy Act (“NEPA”), the panel held that the\\nplaintiffs had standing to bring their NEPA claims because\\nthey were trying to protect the environment, which was\\nwithin NEPA’s zone of interests. The panel concluded that\\nthe environmental impacts discussed in the environmental\\nassessments did not raise substantial concerns that\\nnecessitated the preparation of environmental impact\\nstatements. The panel held that there was no indication that\\nthe agency failed to satisfy NEPA’s procedural\\nrequirements. The panel concluded that the Forest Service\\ngave the requisite hard look and made determinations that\\nwere neither arbitrary nor capricious, and were consistent\\nwith the evidence before it; and accordingly, the Forest\\nService did not violate NEPA.\\n\\n The panel held that the Forest Service conducted the\\nrequired prefield work, consulted with the appropriate\\nentities, and reached a determination with the evidence\\nbefore it, and satisfied its procedural obligations under the\\nNational Historic Preservation Act.\\n\\f4 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n COUNSEL\\n\\nJohn R. Mellgren (argued), and Susan Jane Brown, Western\\nEnvironmental Law Center, Eugene, Oregon, for Plaintiffs-\\nAppellants.\\n\\nStuart Wilcox, WildEarth Guardians, Denver, Colorado, for\\nPlaintiff-Appellant WildEarth Guardians.\\n\\nMark R. Haag (argued), Allen M. Brabender, Attorneys,\\nEnvironment & Natural Resources Division; Eric Grant,\\nDeputy Assistant Attorney General, Jeffrey H. Wood,\\nActing Assistant Attorney General, United States\\nDepartment of Justice, Washington, D.C.; M’Leah\\nWoodard, Attorney Advisor, Albuquerque Field Office,\\nOffice of the General Counsel, United States Department of\\nAgriculture, Albuquerque, New Mexico; for Defendants-\\nAppellees.\\n\\nDominic Draye (argued), John LeSueur, Assistant Attorney\\nGeneral, and Mark Brnovich, Attorney General, Arizona\\nAttorney General’s Office, Phoenix, Arizona, for\\nIntervenor-Defendant-Appellee State of Arizona.\\n\\nAnna M. Seidman, Douglas S. Burdin, and Jeremy E. Clare,\\nSafari Club International, Washington, D.C., for Intervenor-\\nDefendant-Appellee Safari Club International.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 5\\n\\n OPINION\\n\\nM. SMITH, Circuit Judge:\\n\\n Plaintiffs-Appellants (Plaintiffs) are environmental\\nadvocacy groups that challenged travel management plans\\nimplemented by Defendant-Appellee United States Forest\\nService (the Forest Service) to permit limited motorized big\\ngame retrieval in three Ranger Districts of the Kaibab\\nNational Forest. The district court granted the Forest\\nService’s motion for summary judgment, concluding that the\\nForest Service complied with the Travel Management Rule,\\nthe National Environmental Policy Act (NEPA), and the\\nNational Historic Preservation Act (NHPA). We affirm.\\n\\n FACTUAL AND PROCEDURAL BACKGROUND\\n\\nI. Factual Background\\n\\n A. The Kaibab National Forest\\n\\n The Kaibab National Forest encompasses approximately\\n1.6 million acres of public land in northern Arizona,\\nincluding Grand Canyon National Park. It is comprised of\\nthree noncontiguous Ranger Districts: the Williams Ranger\\nDistrict, the Tusayan Ranger District, and the North Kaibab\\nRanger District.\\n\\n The Williams Ranger District is the southernmost,\\ncovering 560,305 acres approximately thirty-five miles west\\nof Flagstaff and sixty miles south of Grand Canyon National\\nPark. It includes the Kendrick Mountain Wilderness, which\\nextends into Coconino National Forest and features a diverse\\narray of vegetation including Douglas firs, white firs,\\nponderosa pines, and aspens. The Williams Ranger District\\nalso serves as a habitat for a number of endangered species,\\n\\f6 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nincluding the Mexican spotted owl, the California condor,\\nand the black-footed ferret. It contains six areas where\\nspotted owls are known to live and breed, and three spotted\\nowl critical habitats overlap the District.\\n\\n The Tusayan Ranger District, located just south of Grand\\nCanyon National Park’s south rim, encompasses 331,427\\nacres. It features varied terrain, from ponderosa pine forests\\nto grasslands, and is home to a number of sensitive species,\\nincluding bald eagles, goshawks, peregrine falcons,\\nburrowing owls, bats, and voles.\\n\\n The North Kaibab Ranger District covers 655,078 acres\\nimmediately north of Grand Canyon National Park. Like the\\nWilliams and Tusayan Ranger Districts, the North Kaibab\\nRanger District boasts diverse terrain and vegetation, as well\\nas sensitive animal species. Two federally listed endangered\\nspecies—the Mexican spotted owl and California condor—\\nlive in the District, which the U.S. Fish and Wildlife Service\\nhas designated as critical habitat for the spotted owl.\\n\\n B. The Travel Management Rule\\n\\n In 2005, the U.S. Department of Agriculture\\npromulgated a Travel Management Rule to “provide[] for a\\nsystem of National Forest System roads, National Forest\\nSystem trails, and areas on National Forest System lands that\\nare designated for motor vehicle use.” 36 C.F.R.\\n§ 212.50(a). 1 As part of this system, “[d]esignated roads,\\ntrails, and areas [are] identified on a motor vehicle map,”\\nwhich also “specif[ies] the classes of vehicles” and “the\\n 1\\n The Rule’s antecedents include executive orders issued by\\nPresidents Nixon and Carter that sought to limit the damage to federal\\npublic lands caused by off-road vehicles. See Utah Shared Access All.\\nv. Carpenter, 463 F.3d 1125, 1129–30 (10th Cir. 2006).\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 7\\n\\ntimes of year for which use is designated.” Id. § 212.56.\\nMotor vehicle use is prohibited on roads not so designated.\\nId. § 212.50(a). The regulations include a specific provision\\nconcerning the use of motor vehicles for dispersed camping\\nand big game retrieval, which allows that\\n\\n in designating routes, the responsible official\\n may include in the designation the limited use\\n of motor vehicles within a specified distance\\n of certain forest roads or trails where motor\\n vehicle use is allowed, and if appropriate\\n within specified time periods, solely for the\\n purposes of dispersed camping or retrieval of\\n a downed big game animal by an individual\\n who has legally taken that animal.\\n\\nId. § 212.51(b).\\n\\n The Forest Service’s Southwestern Regional Office\\nissued guidelines for implementation of the Travel\\nManagement Rule, including its motorized big game\\nretrieval provision. The guidelines noted that “National\\nForests in the Southwestern Region provide hunting\\nopportunities that are important to the public,” and directed\\nforests to identify designated routes for game retrieval “in\\nclose collaboration with the responsible State agency.” They\\nalso suggested, pursuant to discussions with the Arizona\\nGame and Fish Department, that motorized big game\\nretrieval be allowed “up to three miles from a designated\\nroute” for bison and “up to one mile from a designated route”\\nfor elk and mule deer.\\n\\f8 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n The Forest Service crafted travel management plans for\\neach of the three Ranger Districts in the Kaibab National\\nForest. It also prepared an Environmental Assessment (EA)\\nfor each plan to ascertain its environmental impact, but did\\nnot undertake a more rigorous Environmental Impact\\nStatement (EIS).\\n\\n i. The Williams Ranger District\\n\\n In July 2010, the Forest Service released the EA for the\\nWilliams Ranger District’s travel management plan, and\\nsubsequently issued a Decision Notice and Finding of No\\nSignificant Impact (DN/FONSI). The DN/FONSI generally\\n“prohibit[s] motorized travel off of designated routes on the\\nWilliams Ranger District,” but permits “the limited use of\\nmotor vehicles within one mile of all designated system\\nroads (except where prohibited) to retrieve a legally hunted\\nand tagged elk during all elk hunting seasons.” It allows\\nmotorized big game retrieval of elk (but not bison) up to one\\nmile off all designated open roads, so long as hunters make\\nonly “[o]ne trip that uses [the] most direct route and least\\nground disturbing.” The designated open road system\\nconsists of 1,114 miles of roadway, a reduction from\\nprevious motor vehicle activity, when 1,460 miles of roads\\nand 95 percent of the District were open to motor vehicle\\nuse. Several miles of the open roads pass through the spotted\\nowl critical habitat.\\n\\n ii. The Tusayan Ranger District\\n\\n Previously, the Tusayan Ranger District contained more\\nthan 700 miles of roads open to motor vehicles, and a vast\\nmajority of the District was open to cross-country motor\\nvehicle travel. The Forest Service’s final DN/FONSI for the\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 9\\n\\nDistrict 2 designated 566 miles of road open to motor\\nvehicles. The decision permits “[l]egally harvested elk [to]\\nbe retrieved during all legal elk hunting seasons” by motor\\nvehicles within one mile of designated roads. Motorized\\nretrieval of bison is not permitted, and the DN/FONSI limits\\nuse of motor vehicles when “conditions are such that travel\\nwould cause damage to natural and/or cultural resources,”\\nand mandated that “[m]otorized vehicles would not be\\npermitted to cross riparian areas, streams and rivers except\\nat hardened crossings or crossings with existing culverts.”\\n\\n iii. The North Kaibab Ranger District\\n\\n Prior to implementation of a new travel management\\nplan, 1,852 miles of road in the North Kaibab Ranger District\\nwere open to motor vehicle use, with 83 percent of the\\nDistrict open to cross-county travel. In September 2012, the\\nForest Service released an EA analyzing the District’s new\\nplan. Among other data, the EA noted that while “[c]ross-\\ncountry motorized travel, whether to retrieve game or for\\nother purposes, can adversely affect cultural resource sites if\\na vehicle is driven across a site,” only thirty-eight bison and\\nno elk were taken from the District in 2009.\\n\\n The Forest Service issued a DN/FONSI that designated\\n1,476 miles of open roads for motorized travel, including an\\nadditional 16 miles of unauthorized, user-created roads.\\nMotor vehicles can be used to retrieve elk or bison during\\nhunting seasons, under certain limiting conditions. Notably,\\n\\n 2\\n In April 2009, the Forest Service issued an initial EA that analyzed\\nthe impact of the new travel management plan, as well as a subsequent\\nDN/FONSI. In response to administrative appeals, that decision was\\nreversed, and a new environmental survey undertaken. The resulting\\nEA, issued in January 2011, reflected additional analysis and public\\ncomment.\\n\\f10 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nthe plan prohibits motorized retrieval of mule deer; the data\\nindicated that far more mule deer—1,020—were harvested\\nin the District in 2009 than bison or elk. The DN/FONSI\\nalso included guidance for monitoring and mitigation, as\\nwell as practices to limit the spread of invasive exotic weeds.\\n\\nII. Procedural Background\\n\\n The Districts’ travel management plans—their\\nmotorized big game retrieval provisions in particular—were\\nadministratively appealed, and the Regional Forester upheld\\nthem.\\n\\n Plaintiffs then filed a complaint for declaratory and\\ninjunctive relief in the district court. They challenged the\\ntravel management plans for each of the three Ranger\\nDistricts, alleging violations of the Travel Management\\nRule, the Administrative Procedure Act (APA), NEPA, and\\nthe NHPA. After the parties filed and briefed cross-motions\\nfor summary judgment, the district court granted the Forest\\nService’s motion and denied Plaintiffs’ motion. It further\\ndenied the motions for summary judgment filed by\\nIntervenors-Defendants-Appellees State of Arizona (the\\nState) and Safari Club International (Safari Club) as moot.\\nThis timely appeal followed.\\n\\n STANDARD OF REVIEW AND JURISDICTION\\n\\n We review de novo a district court’s order granting or\\ndenying a motion for summary judgment. Churchill County\\nv. Norton, 276 F.3d 1060, 1071 (9th Cir. 2001). Under the\\nAPA, agency action can be set aside if it is “arbitrary,\\ncapricious, an abuse of discretion, or otherwise not in\\naccordance with law.” 5 U.S.C. § 706(2)(A); see also\\nWildEarth Guardians v. Mont. Snowmobile Ass’n, 790 F.3d\\n920, 932 (9th Cir. 2015) (Travel Management Rule reviewed\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 11\\n\\nunder the APA); San Carlos Apache Tribe v. United States,\\n417 F.3d 1091, 1099 (9th Cir. 2005) (NHPA challenge\\nreviewed under the APA); Churchill County, 276 F.3d at\\n1071 (NEPA challenge reviewed under the APA).\\n\\n We have jurisdiction pursuant to 28 U.S.C. § 1291.\\n\\n ANALYSIS\\n\\nI. The Travel Management Rule\\n\\n Plaintiffs argue that the Forest Service violated the\\nTravel Management Rule by implementing plans that did not\\nsufficiently limit motorized big game retrieval in the\\nDistricts.\\n\\n Although the Travel Management Rule generally\\nprohibits off-road, motorized travel, it permits officials to\\ndesignate “the limited use of motor vehicles within a\\nspecified distance of certain forest roads . . . solely for the\\npurposes of dispersed camping or retrieval of a downed big\\ngame animal.” 36 C.F.R. § 212.51(b) (emphases added). As\\npart of this designation process, the Forest Service must\\nconsider various criteria, ranging from “public safety” and\\n“conflicts among uses of National Forest System lands” to\\n“[d]amage to soil, watershed, vegetation, and other forest\\nresources” and “[h]arassment of wildlife and significant\\ndisruption of wildlife habitats.” 36 C.F.R. § 212.55(a)–(b).\\nCompliance with these “minimization criteria” is\\nmandatory. Mont. Snowmobile, 790 F.3d at 929–32.\\n\\n Plaintiffs contend that the Forest Service violated the\\nTravel Management Rule by permitting off-road motorized\\nvehicle use to collect downed game within one mile of every\\nopen road in the Districts, in purported violation of the\\n\\f12 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nRule’s mandate that such activity be “limited” and only on\\n“certain” roads. We disagree.\\n\\n A. “Limited”\\n\\n First, Plaintiffs argue that “designating cross-country\\noff-road motor vehicle use for one mile off both sides of\\nevery single open road on each Ranger District is not a\\n‘limited’ designation as contemplated by the Travel\\nManagement Rule.” They note that the Forest Service’s own\\nregional office acknowledged as much when reviewing the\\nNorth Kaibab Ranger District’s plan proposal; the office\\ncommented, “Motorized [big game retrieval] is being\\nproposed on all system routes which is not consistent with\\nrule for ‘limited use.’” The assistant NEPA coordinator of\\nthat District also questioned the scope of the plan, noting,\\n“What I don’t see is how this got you to your proposed\\nactions, particularly the [motorized big game retrieval] on\\nALL roads with 1 mile corridor . . . . that’s huge.” In their\\nbrief, Plaintiffs include maps prepared by the Forest Service,\\nwhich illustrate the extent of areas (in light blue) where\\nmotorized big game retrieval is allowed in the Tusayan and\\nWilliams Ranger Districts. These maps indicate, as\\nPlaintiffs argue, that the plans permit motorized big game\\nretrieval across a vast swath of the Districts’ lands:\\n\\fWILDEARTH GUARDIANS V. PROVENCIO 13\\n\\f14 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n As the district court correctly noted, however, Plaintiffs’\\nfocus on only the spatial limitation of the Districts’ plans\\nignores the other restrictions on motorized big game\\nretrieval. The North Kaibab Ranger District plan illustrates\\nsome of these additional limitations: restricting retrieval to\\nlegally hunted elk and bison 3; permitting only one vehicle\\nper harvested animal; requiring hunters to “use the most\\ndirect and least ground disturbing route in and out of the area\\nto accomplish the retrieval”; and limiting the temporal\\nperiod to the “appropriate season as designated by the\\n[State], and for 24 hours following each season.” The\\nWilliams and Tusayan Ranger Districts further exclude the\\nretrieval of bison and require motor vehicles to cross streams\\nand rivers only at designated crossings. We agree with the\\ndistrict court that “these limitations are a significant\\ndeparture from the previous policy which did not limit the\\nnumber of trips . . . , did not limit the type of species which\\ncould be retrieved by motor vehicle, did not limit the\\ndistance traveled from system roads, and had no restrictions\\non seasons or weather conditions and no requirement for use\\nof a direct route.”\\n\\n Plaintiffs assert that “the nearly unlimited spatial\\nallowance . . . in and of itself[] violates the plain terms of the\\nTravel Management Rule,” but provide no authority for that\\nproposition. Because we find no support for it in either case\\nlaw or the applicable regulations, we conclude that the Forest\\nService did not abuse its discretion when it authorized plans\\nthat limit motorized big game retrieval based on factors other\\nthan geography. See 5 U.S.C. § 706(2)(A).\\n\\n\\n 3\\n The State notes that additional, non-qualifying big game species\\ncan also be found in the Kaibab National Forest, including mule deer,\\npronghorn, and black bears.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 15\\n\\n As for the regional office’s skepticism regarding the\\nscope of retrieval permitted under the plans, “the fact that a\\npreliminary determination by a local agency representative\\nis later overruled at a higher level within the agency does not\\nrender the decisionmaking process arbitrary and capricious,”\\nas agencies are “fully entitled” to “change[] their minds . . .\\nas long as the proper procedures were followed.” Nat’l\\nAss’n of Home Builders v. Defs. of Wildlife, 551 U.S. 644,\\n659 (2007). Such apparent inconsistencies might serve as\\nevidence of arbitrariness or capriciousness, see Barnes v.\\nU.S. Dep’t of Transp., 655 F.3d 1124, 1134 (9th Cir. 2011)\\n(noting that Home Builders “did not hold, however, that such\\npreliminary determinations are irrelevant in any context”),\\nbut absent other evidence that the plans violated the Travel\\nManagement Rule, we do not find these comments\\nparticularly enlightening, especially since the Regional\\nForester ultimately upheld the designation decisions as to\\neach Ranger District. 4\\n\\n Ultimately, as the district court concluded, “Plaintiffs\\nhave only identified dissatisfaction with the ultimate\\ndecisions made by the Forest Service in authorizing\\n[motorized big game retrieval] in the three ranger districts.”\\nAlthough it is not unreasonable to interpret “limited use”\\nspatially, Plaintiffs point to no statute, regulation, or ruling\\nthat requires a geographic limitation of this sort. The\\nDistricts’ plans limit motorized big game retrieval as to\\ntiming, qualified species, and number of vehicles, which is\\n\\n 4\\n The same can be said for any inconsistencies between the final\\nplans and the guidance provided by the Forest Service’s regional officer,\\nto which Plaintiffs point as evidence that the Rule was violated.\\nAlthough, for example, the guidance suggested that retrieval of elk not\\nbe allowed “between one hour before sunrise and 10:00 am,” that\\nlimitation was not a requirement, but merely a recommendation, and so\\na failure to implement it does not render the plans unlawful.\\n\\f16 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nboth a reasonable interpretation of the Rule and an indication\\nthat the Forest Service both considered and applied the\\nmandatory minimization criteria. See Mont. Snowmobile,\\n790 F.3d at 932. Given that much of the Districts’ land was\\nopen to cross-country motorized travel prior to\\nimplementation of the plans, we conclude that the new\\nrestrictions constitute a “limited” use of motorized vehicles.\\n\\n B. “Certain”\\n\\n Next, the Rule allows motorized retrieval on “certain\\nforest roads.” 36 C.F.R. § 212.51(b). Plaintiffs argue that\\nthe word “certain” requires that it only be allowed on “some,\\nbut not all” roads. Because the plans permit retrieval within\\none mile of all designated roads, Plaintiffs conclude that they\\nare unlawful. The Forest Service, however, correctly notes\\nthat while the word “certain” can mean “some, but not all,”\\nthe more common definition of the term is “definite” or\\n“fixed.” See, e.g., Webster’s Third New International\\nDictionary 367 (2002) (listing the primary definitions of\\n“certain” as “fixed, settled, stated” and “exact, precise”).\\n\\n Because the Forest Service limited motor vehicle use to\\na defined set of roads in each District, it complied with the\\nRule. Even if the proper interpretation of the word “certain”\\nwere ambiguous, the Forest Service’s definition of “fixed”\\nor “definite” is permissible, consistent with the text of the\\nRule, and entitled to deference. See Home Builders,\\n551 U.S. at 672 (“An agency’s interpretation of the meaning\\nof its own regulations is entitled to deference ‘unless plainly\\nerroneous or inconsistent with the regulation.’” (quoting\\nAuer v. Robbins, 519 U.S. 452, 461 (1997))).\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 17\\n\\n C. “Sparingly”\\n\\n Lastly, Plaintiffs contend that the Forest Service failed to\\nimplement motorized big game retrieval “sparingly.” Travel\\nManagement; Designated Routes and Areas for Motor\\nVehicle Use, 70 Fed. Reg. 68,264, 68,285 (Nov. 9, 2005)\\n(“The Department expects the Forest Service to apply this\\nprovision sparingly, on a local or State-wide basis, to avoid\\nundermining the purposes of the final rule and to promote\\nconsistency in implementation.” (emphasis added)). They\\nargue that “allowing motorized big game retrieval off of\\nevery single open road is not using [the Forest Service’s]\\nauthority sparingly,” and that the Forest Service erroneously\\nconcluded that the plans were sparing and limited because\\nthey permitted less motorized retrieval than under the prior\\npolicies that imposed no restrictions.\\n\\n This argument is unpersuasive. First, the word\\n“sparingly” does not appear in the Rule, but instead in its\\npreamble. We look to a preamble only when the regulation\\nitself is ambiguous. El Comité para el Bienstar de Earlimart\\nv. Warmerdam, 539 F.3d 1062, 1070 (9th Cir. 2008).\\nAccordingly, as the Forest Service correctly asserts, the\\npreamble does not “impose a duty above and beyond the\\nactual terms of the regulation.”\\n\\n Moreover, Plaintiffs’ argument fails for much the same\\nreason as their “limited” argument above: it relies on a\\nstrictly spatial conception of “sparingly,” while ignoring the\\nother restrictions that the plans impose on motorized big\\ngame retrieval. Absent authority requiring a strictly\\ngeographic interpretation of the words “limited” and\\n“sparingly,” we conclude that the Forest Service did not\\nviolate the plain terms of the Travel Management Rule.\\n\\f18 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nII. NEPA\\n\\n A. Standing\\n\\n As a preliminary matter, Safari Club challenges whether\\nPlaintiffs have standing to bring their NEPA claims. It\\nobserves that Plaintiffs’ “standing declarant, Kim Crumbo,\\nrevealed that his own activities in the forests cause the very\\nsame impact that [they] seek to attribute to motorized big\\ngame retrieval,” since Crumbo recounted “incidents during\\nwhich he, on his bicycle, interfered with young goshawk and\\na goshawk pursuing its prey.” Accordingly, Safari Club\\nsuggests that Plaintiffs cannot satisfy the redressability\\nrequirement for standing, see Salmon Spawning & Recovery\\nAll. v. Gutierrez, 545 F.3d 1220, 1226 (9th Cir. 2008),\\nbecause Plaintiffs “themselves are a source of” the negative\\neffects that they seek to analyze through an EIS, and so “[n]o\\nchange in process and no [EIS] analysis of motorized big\\ngame retrieval will prevent [Plaintiffs’] members from\\nengaging in conduct that is the source of effects” that they\\nseek to mitigate.\\n\\n Under Safari Club’s reasoning, a hypothetical plaintiff\\nchallenging an EPA decision on a CO2-emitting power plant\\nwould lack standing just because she also happens to exhale\\ncarbon dioxide. This result would not only be absurd, but\\nalso contrary to our prior precedent, for we have held that\\n\\n the mere existence of multiple causes of an\\n injury does not defeat redressability,\\n particularly for a procedural injury. So long\\n as a defendant is at least partially causing the\\n alleged injury, a plaintiff may sue that\\n defendant, even if the defendant is just one of\\n multiple causes of the plaintiff’s injury.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 19\\n\\nWildEarth Guardians v. U.S. Dep’t of Agric., 795 F.3d 1148,\\n1157 (9th Cir. 2015).\\n\\n Safari Club also challenges Plaintiffs’ NEPA standing\\nbased on their perceived motivation for bringing suit.\\nAsserting that Plaintiffs’ “goal in bringing this action was to\\nuse NEPA and the NHPA to force [the Forest Service] to\\nreverse [its] authorization of motorized big game retrieval,”\\nit contends that “NEPA does not provide a cause of action”\\nand that Plaintiffs lack prudential standing. Regardless of\\nPlaintiffs’ motivation in commencing this suit, however, if a\\ngroup is “trying to protect the environment,” then its “suit []\\nlies well within NEPA’s zone of interests.” Citizens for\\nBetter Forestry v. U.S. Dep’t of Agric., 341 F.3d 961, 976\\n(9th Cir. 2003); see also Havasupai Tribe v. Provencio,\\n906 F.3d 1155, 1166 (9th Cir. 2018). Accordingly, we\\nconclude that Plaintiffs have standing to bring their NEPA\\nclaims.\\n\\n B. Legal Framework\\n\\n NEPA requires federal agencies, including the Forest\\nService, to assess the environmental impact of proposed\\nactions that “significantly affect[] the quality of the human\\nenvironment.” 42 U.S.C. § 4332(C). It\\n\\n serves two fundamental objectives. First, it\\n “ensures that the agency, in reaching its\\n decision, will have available, and will\\n carefully consider, detailed information\\n concerning significant environmental\\n impacts.” And, second, it requires “that the\\n relevant information will be made available\\n to the larger audience that may also play a\\n role in both the decisionmaking process and\\n the implementation of that decision.”\\n\\f20 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nMont. Snowmobile, 790 F.3d at 924 (citation omitted)\\n(quoting Robertson v. Methow Valley Citizens Council,\\n490 U.S. 332, 349 (1989)). In short, “NEPA’s purpose is to\\nensure that ‘the agency will not act on incomplete\\ninformation, only to regret its decision after it is too late to\\ncorrect.’” Friends of Clearwater v. Dombeck, 222 F.3d 552,\\n557 (9th Cir. 2000) (quoting Marsh v. Or. Nat. Res. Council,\\n490 U.S. 360, 371 (1989)). We have admonished that “[w]e\\nmust . . . strictly interpret the procedural requirements in\\nNEPA . . . ‘to the fullest extent possible’ consistent with the\\npolicies embodied in NEPA. ‘[G]rudging, pro forma\\ncompliance will not do.’” Churchill County, 276 F.3d at\\n1072 (fourth alteration in original) (citation omitted)\\n(quoting Lathan v. Brinegar, 506 F.2d 677, 687, 693 (9th\\nCir. 1974) (en banc)). As part of this compliance, agencies\\nmust ensure “that environmental information is available to\\npublic officials and citizens before decisions are made and\\nbefore actions are taken. The information must be of high\\nquality. Accurate scientific analysis, expert agency\\ncomments, and public scrutiny are essential to implementing\\nNEPA.” 40 C.F.R. § 1500.1(b).\\n\\n The Forest Service must prepare an EIS—a more\\nthorough undertaking than an EA 5—if an action might\\nsignificantly affect environmental quality. As we have\\nexplained,\\n\\n\\n\\n 5\\n “Before deciding whether to complete an EIS, government\\nagencies may prepare a less formal EA which ‘briefly provides sufficient\\nevidence and analysis for determining whether to prepare an\\nenvironmental impact statement or a finding of no significant impact.’”\\nAnderson v. Evans, 371 F.3d 475, 488 (9th Cir. 2004) (quoting Tillamook\\nCounty v. U.S. Army Corps of Eng’rs, 288 F.3d 1140, 1144 (9th Cir.\\n2002)).\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 21\\n\\n An EIS must be prepared if “substantial\\n questions are raised as to whether a project\\n . . . may cause significant degradation of\\n some human environmental factor.” Thus, to\\n prevail on a claim that the Forest Service\\n violated its statutory duty to prepare an EIS,\\n a “plaintiff need not show that significant\\n effects will in fact occur.” It is enough for the\\n plaintiff to raise “substantial questions\\n whether a project may have a significant\\n effect” on the environment.\\n\\nBlue Mountains Biodiversity Project v. Blackwood, 161 F.3d\\n1208, 1212 (9th Cir. 1998) (alteration in original) (citations\\nomitted) (quoting Idaho Sporting Cong. v. Thomas, 137 F.3d\\n1146, 1149–50 (9th Cir. 1998)); see also Save the Yaak\\nComm. v. Block, 840 F.2d 714, 717 (9th Cir. 1988) (“[A]n\\nagency’s decision not to prepare an EIS will be considered\\nunreasonable if the agency fails to ‘supply a convincing\\nstatement of reasons why potential effects are\\ninsignificant.’” (quoting Steamboaters v. FERC, 759 F.2d\\n1382, 1393 (9th Cir. 1985))). The significance of an action\\ndepends on its context and intensity, the latter of which is\\nassessed using a list of criteria enumerated in the relevant\\nregulation. See 40 C.F.R. § 1508.27(a)–(b).\\n\\n C. Whether the Forest Service Needed to Prepare\\n EISs\\n\\n Plaintiffs contend that “the presence of several\\nsignificance factors indicating possible significant\\nenvironmental consequences of the proposed actions”\\nrequired the Forest Service to prepare EISs for each of the\\nDistricts’ travel management plans. We analyze in turn each\\nconsideration to which they point.\\n\\f22 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n i. Impacts\\n\\n The first enumerated consideration in evaluating an\\naction’s intensity concerns “[i]mpacts that may be both\\nbeneficial and adverse,” and notes that “[a] significant effect\\nmay exist even if the Federal agency believes that on balance\\nthe effect will be beneficial.” 40 C.F.R. § 1508.27(b)(1).\\n“Significance exists if it is reasonable to anticipate a\\ncumulatively significant impact on the environment.\\nSignificance cannot be avoided by terming an action\\ntemporary or by breaking it down into small component\\nparts.” Id. § 1508.27(b)(7).\\n\\n Plaintiffs suggest that “the travel management plans for\\neach Ranger District would have significant direct, indirect,\\nand cumulative impacts,” noting that both motorized vehicle\\nuse on open, designated roads and cross-country, off-road\\nmotorized vehicle use can have “significant detrimental\\neffects . . . on a variety of resources.”\\n\\n The assertion that motorized big game retrieval can have\\ndetrimental effects on the environment is consistent with the\\ninformation contained in the EAs prepared for the Williams\\nand Tusayan Ranger Districts. Those EAs noted that “[t]he\\nscientific literature documents a variety of negative effects\\nof roads and motorized travel on wildlife,” with\\n\\n [p]otential direct and indirect effects of roads\\n and motorized travel on wildlife includ[ing]\\n habitat loss, fragmentation, and degradation\\n caused by roads and cross country motorized\\n travel; roads can create barriers to\\n movements of certain species; animals can be\\n killed or injured as a result of being hit or run\\n over by motor vehicles; human disturbance\\n or harassment of animals caused by or\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 23\\n\\n facilitated by motorized travel; [and]\\n shooting or harvest of animals facilitated by\\n motor vehicle access to wildlife habitats.\\n\\nThe Williams Ranger District EA further indicated that off-\\nroad vehicle use “in areas with sensitive or moist soils can\\ncreate tracks, ruts and new user routes that may crush,\\ndisplace, and/or destroy cultural materials (i.e. artifacts,\\nfeatures, traditionally used plants), and damage significant\\ninformation that may contribute to our understanding of\\nhistory.” A particularly vexatious problem related to\\nmotorized vehicle use is the spread of invasive weeds. Each\\nof the three EAs noted that vehicles are a common cause of\\nweed introduction and spread, with the North Kaibab Ranger\\nDistrict EA reporting that “[t]he authorization of motorized\\nbig game retrieval will have an increased threat of invasive\\nspecies spread as every vehicle that travels cross-country has\\nthe ability to serve as a vector and create disturbance.” 6\\n\\n We do not disagree with Plaintiffs’ assertion that\\nmotorized big game retrieval can have a negative effect on\\nthe environment. But we nevertheless conclude that the\\nenvironmental impacts discussed in the EAs did not raise\\nsubstantial concerns that necessitated the preparation of\\nEISs. Plaintiffs might disagree with the Forest Service’s\\nsubstantive conclusions, but we see no indication that the\\nagency failed to satisfy NEPA’s procedural requirements—\\na crucial distinction that is lost in Plaintiffs’ formulation of\\nthe issue.\\n\\n\\n 6\\n This is because, as explained in the Williams Ranger District EA,\\n“[v]ehicles driving through populations of invasive plants often get\\nseed[s] entrapped in tire tread or undercarriages, move to another area\\nand then drop seeds into a previously uninfested area.”\\n\\f24 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n Plaintiffs’ treatment of the spread of invasive weeds is\\nillustrative. They tie the problem of weeds to another\\nenumerated NEPA consideration that implicates “[u]nique\\ncharacteristics of the geographic area such as proximity to\\nhistoric or cultural resources, park lands, prime farmlands,\\nwetlands, wild and scenic rivers, or ecologically critical\\nareas.” Id. § 1508.27(b)(3). They note that “both the North\\nKaibab and Tusayan Ranger Districts immediately abut\\nGrand Canyon National Park.” The EAs agreed: they\\ndescribed Grand Canyon National Park as “internationally\\nimportant,” and noted the Tusayan Ranger District’s “unique\\nlocation” as “a gateway to one of the most famous national\\nparks in the country.” In that District’s EA, the Forest\\nService acknowledged that recreation and hunting—\\nactivities facilitated by the travel management plan—“have\\nthe potential to introduce exotic plants” that “may then\\nspread to adjoining lands, including the Grand Canyon\\nNational Park.” The National Park Service, in a letter from\\nthe Acting Park Superintendent of Grand Canyon National\\nPark, advised the Forest Service to “institute a buffer zone\\nof 1-mile along the park boundary for any purpose including\\nbig-game retrieval, fuel-wood gathering, cross-county\\ntravel, etc.,” due to “increased pressure from motorized\\nvehicles at or near the southern park boundary over the past\\nseveral years”—a recommendation that was not adopted in\\nthe Tusayan Ranger District DN/FONSI.\\n\\n Clearly, the EAs demonstrated that motorized big game\\nretrieval risks the spread of invasive weeds, an undeniable\\nenvironmental impact. In response, the Forest Service relies\\nin part on questionable reasoning by focusing on the fact that\\nthe plans reduced negative impacts when compared with\\npre-plan activity. It notes that the plan eventually selected\\nfor the Tusayan Ranger District opted to limit the number of\\nroads open to the public, which, the EA noted, “reduces the\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 25\\n\\nnumber of opportunities for noxious and invasive exotic\\nweeds to be introduced and spread.” “Thus,” the Forest\\nService concludes, “contrary to [Plaintiffs’] argument, the\\nTusayan EA confirms that the Forest Service considered the\\nissue and reasonably concluded that the decision will reduce,\\nnot increase, the spread of exotic plants.”\\n\\n However, a conclusion, even a correct one, that a given\\naction might reduce a potential impact does not alone\\nindicate that the impact would not be significant. We have\\nnoted that the use of baselines is a helpful, and perhaps\\ninevitable, tool in conducting environmental surveys. See\\nOr. Nat. Desert Ass’n v. Jewell, 840 F.3d 562, 568 (9th Cir.\\n2016) (“The establishment of a ‘baseline is not an\\nindependent legal requirement, but rather, a practical\\nrequirement in environmental analysis often employed to\\nidentify the environmental consequences of a proposed\\nagency action.’” (quoting Am. Rivers v. FERC, 201 F.3d\\n1186, 1195 n.15 (9th Cir. 1999))). But we have also\\ndetermined that “simply because the Final Rule may be an\\nimprovement over the [previous] standard does not\\nnecessarily mean that it will not have a ‘significant effect’\\non the environment” where the agency “has not explained\\nwhy its rule will not have a significant effect.” Ctr. for\\nBiological Diversity v. NHTSA, 538 F.3d 1172, 1224 (9th\\nCir. 2008). The plans chosen by the Forest Service might\\nreduce (even substantially reduce) the spread of noxious\\nweeds in the Districts, but that alone does not address\\nwhether that reduced impact itself has a significant impact\\non the environment generally and Grand Canyon National\\nPark in particular. Therefore, the Forest Service cannot rely\\nsolely on the reduction of adverse impacts to demonstrate\\nthat those impacts are not significant.\\n\\f26 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n But that the Forest Service occasionally conflates\\nreduction with insignificance does not necessarily mean that\\nit violated NEPA. To demonstrate, we once again go into\\nthe weeds.\\n\\n The Forest Service acknowledged that the North Kaibab\\nRanger District contained “several species of invasive\\nweeds,” which “are spread [] via roads and forest visitors.”\\nThe plan that was eventually selected “reduce[d] the number\\nof roads that can be traveled on by 376 miles,” which, the\\nEA found, would “lower the amount of invasive species seed\\nintroduced or spread.” But notably, the EA continued:\\n\\n The authorization of motorized big game\\n retrieval will have an increased threat of\\n invasive species spread as every vehicle that\\n travels cross-country has the ability to serve\\n as a vector and create disturbance.\\n Alternative 2 [the selected plan] authorizes\\n motorized big game retrieval for only elk and\\n mule deer. This is expected to lead to only a\\n small increase in the potential for invasive\\n species spread and disturbance when\\n compared to Alternative 3 and should not\\n generate any realistic impacts.\\n\\nThis passage demonstrates that the Forest Service did not\\nmerely rely on the possibility of reduction, but also\\nconcluded that the plan would “not generate any realistic\\nimpacts”; in other words, that the effects would not be\\nsignificant. Furthermore, as the district court noted, the EA\\nmentioned that the North Kaibab Ranger District featured\\n“projects focus[ed] on treating known infestations across the\\nDistrict, prioritizing the species and locations that pose the\\ngreatest threats,” methods that had “proven successful in\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 27\\n\\neradicating or reducing potentially serious noxious species\\nthreats.” This language indicates that the Forest Service\\nacknowledged a potential environmental impact, and then\\ndetermined that, due to features of the travel management\\nplan and other remediation efforts, it was unlikely to be\\nsignificant. Contrary to Plaintiffs’ argument, the Forest\\nService did not merely determine that the problem would be\\nreduced; it also concluded that the impact would not be\\nsignificant. Plaintiffs disagree with the EA’s conclusion, but\\nthis is “a classic example of a factual dispute the resolution\\nof which implicates substantial agency expertise.” Marsh,\\n490 U.S. at 376. We agree with the district court: “Plaintiffs’\\nsingular and conclusory statement that exotic plants might\\nspread . . . does not raise substantial questions that would\\ntrigger the need for an EIS.”\\n\\n Similarly, although the Forest Service did not follow all\\nof the recommendations made by Grand Canyon National\\nPark’s Acting Park Superintendent, this fact does not mean\\nthat it ignored a significant environmental impact. Agencies\\ncan thoughtfully consider suggestions but ultimately decide\\nto reject them, and the presence of an articulated concern\\ndoes not alone trigger the need to conduct an EIS. See Native\\nEcosystems Council v. U.S. Forest Serv., 428 F.3d 1233,\\n1240 (9th Cir. 2005) (“NEPA permits a federal agency to\\ndisclose [] impacts without automatically triggering the\\n‘substantial questions’ threshold.”). Here, the record\\nindicates that the Forest Service explained why a buffer\\nwould not be employed (“We don’t expect to use a buffer\\nzone as many management actions will need to extend to the\\nForest Service-National Park boundary”), and further\\narticulated means of remedying the risk of illegal motor\\nvehicle use (such as “limit[ing] motorized big game retrieval\\nduring all elk seasons” and “work[ing] closely with Arizona\\nGame and Fish Department to monitor and enforce illegal\\n\\f28 WILDEARTH GUARDIANS V. PROVENCIO\\n\\ncross-country travel associated with hunting activities”). We\\nfind no indication in the record that the Forest Service did\\nnot adequately consider potential impacts, or that substantial\\nquestions remained that required the preparation of an EIS.\\n\\n The same conclusion ultimately applies to all of the\\nenvironmental impacts that Plaintiffs highlight in their\\nbriefs: although Plaintiffs disagree with the EAs’ factual\\nconclusions, the Forest Service nonetheless considered the\\nissues, gave them the requisite “hard look,” and thus fulfilled\\ntheir NEPA obligations. Save the Yaak, 840 F.2d at 717. In\\nreaching its conclusions that none of the impacts cited by\\nPlaintiffs were sufficiently significant to require the\\npreparation of EISs, the Forest Service did not “rel[y] on\\nfactors Congress did not intend it to consider, ‘entirely fail[]\\nto consider an important aspect of the problem,’ or offer[] an\\nexplanation ‘that runs counter to the evidence before the\\nagency or is so implausible that it could not be ascribed to a\\ndifference in view or the product of agency expertise.’”\\nLands Council v. McNair, 537 F.3d 981, 987 (9th Cir. 2008)\\n(en banc) (quoting Earth Island Inst. v. U.S. Forest Serv.,\\n442 F.3d 1147, 1157 (9th Cir. 2006)), overruled on other\\ngrounds by Winter v. Nat. Res. Def. Council, Inc., 555 U.S.\\n7 (2008). Instead, the evidence in the record indicates that,\\nalthough the EAs acknowledged that motorized big game\\nretrieval might have negative impacts on the environment,\\nthe Forest Service’s determination that these impacts would\\nnot be significant evinced “a rational connection between the\\nfacts found and the conclusions made.” Or. Nat. Res.\\nCouncil v. Lowe, 109 F.3d 521, 526 (9th Cir. 1997).\\nTherefore, its conclusions were not arbitrary and capricious,\\nand the Forest Service did not violate NEPA by declining to\\nprepare EISs based on the plans’ environmental impacts.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 29\\n\\n ii. Controversy and Uncertainty\\n\\n NEPA also requires the preparation of an EIS when an\\naction’s “effects on the quality of the human environment\\nare likely to be highly controversial,” and/or “are highly\\nuncertain or involve unique or unknown risks.” 40 C.F.R.\\n§ 1508.27(b)(4)–(5). “A project is ‘highly controversial’ if\\nthere is a ‘“substantial dispute [about] the size, nature, or\\neffect of the major Federal action rather than the existence\\nof opposition to a use.”’” Native Ecosystems Council,\\n428 F.3d at 1240 (alteration in original) (emphases added)\\n(quoting Blue Mountains, 161 F.3d at 1212); see also\\nWetlands Action Network v. U.S. Army Corps of Eng’rs,\\n222 F.3d 1105, 1122 (9th Cir. 2000) (“The existence of\\nopposition to a use, however, does not render an action\\ncontroversial.”), abrogated on other grounds by Wilderness\\nSoc’y v. U.S. Forest Serv., 630 F.3d 1173 (9th Cir. 2011).\\n“A substantial dispute exists when evidence, raised prior to\\nthe preparation of an EIS or FONSI casts serious doubt upon\\nthe reasonableness of an agency’s conclusions.” Nat’l Parks\\n& Conservation Ass’n v. Babbitt, 241 F.3d 722, 736 (9th Cir.\\n2001) (citation omitted), abrogated on other grounds by\\nMonsanto Co. v. Geertson Seed Farms, 561 U.S. 139 (2010).\\nAdditionally, because “[a]n agency must generally prepare\\nan EIS if the environmental effects of a proposed agency\\naction are highly uncertain,” such preparation “is mandated\\n‘where uncertainty may be resolved by further collection of\\ndata, or where the collection of such data may prevent\\nspeculation on potential . . . effects.’” Barnes, 655 F.3d at\\n1140 (second alteration in original) (quoting Native\\nEcosystems Council, 428 F.3d at 1240). However, NEPA\\nregulations “do not anticipate the need for an EIS anytime\\nthere is some uncertainty, but only if the effects of the project\\nare ‘highly’ uncertain.” Envtl. Prot. Info. Ctr. v. U.S. Forest\\nServ., 451 F.3d 1005, 1011 (9th Cir. 2006).\\n\\f30 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n Plaintiffs contend that “the travel management plans for\\nthe Tusayan, Williams, and North Kaibab Ranger Districts\\npresent highly controversial and highly uncertain effects that\\ninvolve unique or unknown risks,” because “significant\\ncontroversy exists as to the amount and type of motorized\\nrecreation that would be allowed across the three Ranger\\nDistricts.” But Plaintiffs point to nothing in the record\\nindicating the existence of a substantial dispute that casts\\ndoubt on the Forest Service’s conclusions about\\nenvironmental impacts. There may have been opposition to\\nthe plans, but mere opposition alone is insufficient to support\\na finding of controversy. The Forest Service “recognize[d]\\nthat elements of the Selected Alternative [] generated\\ncontroversy,” but concluded—apparently correctly—that\\nthere was “no substantiated scientific controversy over the\\neffects as described.”\\n\\n Plaintiffs assert that various questions raised during the\\nEA process revealed a high level of scientific uncertainty,\\nbut the record belies that assertion. For example, as to the\\npotential risks “based on the broad allowance of motorized\\nbig game retrieval,” the Forest Service attempted to estimate\\nhunting activity based on past data, and used this information\\nto conclude that the impacts of motorized big game retrieval\\nwould be limited. Plaintiffs identity nothing in the record to\\nsuggest that the Forest Service’s estimates were unduly\\nspeculative, or that it unreasonably relied upon these\\npredictions. Plaintiffs also argue that there was “uncertainty\\nregarding whether or not hunters will actually remove gut\\npiles” when retrieving carcasses, which they must do “to\\nprotect California condors from lead poisoning.” But\\nalthough the Forest Service acknowledged that this issue\\nmight present a problem, the record also indicates that it\\nconsidered the issue and reasonably concluded that it was\\nunlikely to significantly impact the North Kaibab Ranger\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 31\\n\\nDistrict’s condors because the Arizona Game and Fish\\nDepartment had provided to hunters, among other\\nincentives, lead-free ammunition. The Forest Service also\\nnoted that “there would be decreased risk of human\\ndisturbance of scavenging condors as a result of a reduced\\nopen road system and substantially restricted motorized\\ncross-country travel,” and concluded that the North Kaibab\\nRanger District plan “is not likely to jeopardize the\\ncontinued existence [of] California condors.” Plaintiffs\\nneither challenge nor address these conclusions, and instead\\nrely on the mere existence of potential problems as evidence\\nof significant uncertainty—a tactic that does not pass muster.\\nSee Native Ecosystems Council, 428 F.3d at 1240 (“Simply\\nbecause a challenger can cherry pick information and data\\nout of the administrative record to support its position does\\nnot mean that a project is highly controversial or highly\\nuncertain.”).\\n\\n iii. Precedent for Future Actions\\n\\n Another consideration for measuring an action’s\\nintensity for NEPA purposes is “[t]he degree to which the\\naction may establish a precedent for future actions with\\nsignificant effects or represents a decision in principle about\\na future consideration.” 40 C.F.R. § 1508.27(b)(6).\\nAlthough “EAs are usually highly specific to the project and\\nthe locale, thus creating no binding precedent,” In Def. of\\nAnimals v. U.S. Dep’t of Interior, 751 F.3d 1054, 1071 (9th\\nCir. 2014) (quoting Barnes, 655 F.3d at 1140), and the Forest\\nService here explicitly found that each of the three travel\\nmanagement plans was “not likely to establish a precedent\\nfor future actions with significant effects,” Plaintiffs assert\\nthat this consideration is nonetheless implicated because\\n“the Forest Service has [] made public statements indicating\\n\\f32 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nthat what the Kaibab National Forest does, so too will other\\nsouthwestern National Forests.”\\n\\n It is true that the record contains evidence to this effect—\\nincluding that the Coconino National Forest “will defer to\\nthe neighboring Kaibab National Forest’s policy for\\n[motorized big game retrieval] in units shared with the\\nWilliams Ranger District, regardless of how the Coconino\\nproposes to apply the Travel Management Rule,” and that\\nthe Prescott National Forest will “match them as best as we\\ncan”—but that does not mean that the Districts’ plans bind\\nor necessarily shape other forests’ plans in such a way that\\nthey should be considered precedential, especially since any\\nother forest’s plan would be subject to its own NEPA\\nanalysis. Cf. Sierra Club v. Marsh, 769 F.2d 868, 879 (1st\\nCir. 1985) (determining that an action was precedential\\nbecause “once Maine completes the causeway and port,\\npressure to develop the rest of the island could well prove\\nirreversible”). In Oregon Wild v. Bureau of Land\\nManagement, a district court dealt with a similar situation,\\nwhere a project was “part of a larger series of ‘pilot projects’\\naimed at ‘inform[ing] long-term planning’ for management\\nof [] lands in Oregon and California.” No. 6:14-CV-0110-\\nAA, 2015 WL 1190131, at *9 (D. Or. Mar. 14, 2015) (first\\nalteration in original). There, as here, the most that could be\\nconcluded from such a minor precedential effect is that this\\nconsideration “supports the conclusion that an EIS is\\nnecessary”—but “the precedential factor alone is not\\ndispositive.” Id.; see also Anderson v. Evans, 371 F.3d 475,\\n493 (9th Cir. 2004) (finding this factor “insufficient on its\\nown to demonstrate a significant environmental impact”\\nwhere an action is merely influential and not binding). Thus,\\nthis consideration alone did not require preparation of an\\nEIS.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 33\\n\\n iv. Threatened Species\\n\\n Finally, there is the issue of the Mexican spotted owl, a\\nthreatened species found in the Williams and North Kaibab\\nRanger Districts. The Forest Service must consider “[t]he\\ndegree to which the action may adversely affect an\\nendangered or threatened species or its habitat that has been\\ndetermined to be critical under the Endangered Species Act\\nof 1973.” 40 C.F.R. § 1508.27(b)(9). Plaintiffs point to\\nportions of the record that raise the specter of adverse\\neffects—including consultation letters between the Williams\\nand North Kaibab Ranger Districts and the U.S. Fish and\\nWildlife Service regarding the owls—but they again ignore\\nthe Forest Service’s overall conclusions. For example, the\\nWilliams Ranger District EA reported that some roads would\\npass through Mexican spotted owl critical habitats, which\\nmight result in “increased potential human disturbance.”\\nBut the EA ultimately concluded that the selected plan\\nwould be “primarily beneficial,” and would “not adversely\\naffect Mexican spotted owl or Mexican spotted owl\\ndesignated Critical Habitat.” In a separate biological\\nassessment, the Forest Service concluded that “[t]he effects\\ndetermination for Mexican spotted owl and Mexican spotted\\nowl critical habitat is may affect, not likely to adversely\\naffect,” based on the\\n\\n determination [] that potential effects of the\\n proposed action on the Mexican spotted owl\\n would be primarily beneficial. Closing roads\\n and restricting motorized cross-country\\n travel under the proposed action would result\\n in reduced motorized access to spotted owl\\n habitat compared to current management and\\n thus reduced risk of human disturbance to\\n spotted owls, reduced impacts to habitat of\\n\\f34 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n spotted owls and their small mammal prey\\n species, and reduced impacts to designated\\n critical habitat.\\n\\nNotably, the U.S. Fish and Wildlife Service concurred in this\\ndetermination.\\n\\n In short, although the Forest Service did not definitively\\nconclude that no Mexican spotted owls would be adversely\\naffected by the Districts’ travel management plans, the\\nrecord indicates that they sufficiently considered the issue\\nand arrived at a reasonable conclusion that the effects would\\nnot be significant, thus obviating the need for an EIS.\\n\\n v. Summation\\n\\n In the end, we conclude that the Forest Service’s\\ndetermination that no EISs were needed as to the Districts’\\ntravel management plans was reasonable. The plans might\\nhave some precedential effect, there is a possibility that\\nMexican spotted owls might be affected, and exotic weeds\\nmight be spread by motorized big game retrieval, but the\\nrecord ultimately supports the Forest Service’s conclusion\\nthat these concerns do not rise to the level of significance\\nthat would require EISs. The Forest Service gave the\\nrequisite hard look and made determinations that were\\nneither arbitrary nor capricious, and were consistent with the\\nevidence before it. Absent substantial questions that would\\nhave mandated EISs, the Forest Service did not violate\\nNEPA. 7\\n\\n\\n 7\\n Two other enumerated considerations are relevant in this case:\\n“[t]he degree to which the action may adversely affect . . . objects listed\\nin or eligible for listing in the National Register of Historic Places or may\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 35\\n\\nIII. NHPA\\n\\n Finally, Plaintiffs contend that the Forest Service\\nviolated the NHPA by failing to identify and evaluate the\\n“high density of cultural resources” that might be damaged\\nas a result of motorized travel in the Districts.\\n\\n The NHPA’s purpose is to “foster conditions under\\nwhich our modern society and our historic property can exist\\nin productive harmony,” 54 U.S.C. § 300101(1), and it\\nrequires federal agencies to “make a reasonable and good\\nfaith effort” to identify historic properties that might be\\naffected by an action, and to “take [those potential effects]\\ninto account.” 36 C.F.R. § 800.4(b)(1); see also 54 U.S.C.\\n§ 306108. 8 “Like NEPA, ‘[s]ection 106 of NHPA is a “stop,\\nlook, and listen” provision that requires each federal agency\\nto consider the effects of its programs.’” Te-Moak Tribe of\\nW. Shoshone of Nev. v. U.S. Dep’t of Interior, 608 F.3d 592,\\n\\n\\n\\n\\ncause loss or destruction of significant scientific, cultural, or historical\\nresources,” and “[w]hether the action threatens a violation of Federal,\\nState, or local law or requirements imposed for the protection of the\\nenvironment.” 40 C.F.R. § 1508.27(b)(8), (10). However, the former\\nconsideration essentially depends on whether the Forest Service\\ncomplied with the NHPA, while the latter hinges on the Forest Service’s\\ncompliance with both the NHPA and the Travel Management Rule.\\nBecause we conclude that the Forest Service complied with both, we also\\nconclude that no federal laws were violated, and no cultural resources\\nadversely affected, such that EISs were required based on these\\nconsiderations.\\n\\n 8\\n “Historic property means any prehistoric or historic district, site,\\nbuilding, structure, or object included in, or eligible for inclusion in, the\\nNational Register of Historic Places maintained by the Secretary of the\\nInterior.” 36 C.F.R. § 800.16(l)(1).\\n\\f36 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n607 (9th Cir. 2010) (quoting Muckleshoot Indian Tribe v.\\nU.S. Forest Serv., 177 F.3d 800, 805 (9th Cir. 1999)).\\n\\n Plaintiffs assert that the Forest Service violated the\\nNHPA in three ways: (1) by failing “to make a reasonable\\nand good faith effort to identify and evaluate cultural\\nproperties”; (2) by “erroneously determin[ing] that\\n‘Exemption Q’ excuses NHPA consultation for the Tusayan\\nand Williams Ranger Districts”; and (3) by arbitrarily\\nmaking “a ‘no adverse effect’ determination for the North\\nKaibab Ranger District after admitting that cross-country\\ntravel damages cultural resources.” We consider each\\nargument in turn.\\n\\n A. Identification of Cultural Properties\\n\\n Section 106 of the NHPA requires the Forest Service to\\n“make a reasonable and good faith effort to identify historic\\nproperties; determine whether identified properties are\\neligible for listing on the National Register . . .; [and] assess\\nthe effects of the undertaking on any eligible historic\\nproperties found.” Te-Moak Tribe, 608 F.3d at 607 (first\\nalteration in original) (quoting Muckleshoot Indian Tribe,\\n177 F.3d at 805). It must also engage in consultation with\\nthe State Historic Preservation Officer (SHPO) to\\n“[d]etermine and document the area of potential effects,”\\n“[g]ather information,” and “develop and evaluate\\nalternatives or modifications to the undertaking that could\\navoid, minimize, or mitigate adverse effects on historic\\nproperties.” 36 C.F.R. §§ 800.4(a), 800.6(a).\\n\\n Plaintiffs principally argue that the Forest Service failed\\nto “make a reasonable and good faith effort” because it did\\nnot complete 100 percent surveys of potentially affected\\nareas. The First Amended Programmatic Agreement\\nRegarding Historic Property Protection and Responsibilities\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 37\\n\\n(Programmatic Agreement), which the parties agree\\nprovided the relevant guidelines, mandated that “[t]he level\\nof need and extent of new field surveys or inspections will\\nbe proposed by a Professional Cultural Resource Specialist\\nand approved by the Forest Archaeologist based on the\\nguidelines provided in this section.” Those guidelines\\nincluded the use of “relevant information to assess the\\npotential to affect historic properties and the expected nature\\nand distribution of heritage properties that may be affected”;\\nnamely, “[t]he expected nature and severity of all associated\\nimpacts” and “[t]he expected nature and distribution of\\nheritage resources.” Based on the results of this “prefield\\nresearch,” the Resource Specialist and Forest Archeologist\\nwere directed to “determine the relative level of field survey\\nto be conducted.” The Programmatic Agreement called for\\n“100% surveys” where “site density is expected to be high”\\nand where “site densities are unknown and expected visitor\\nuse or impacts will be high.” By contrast, “areas may be\\nsurveyed at less than 100%” where “known site density is\\nlow.”\\n\\n Here, the record supports the Forest Service’s conclusion\\nthat the Programmatic Agreement did not require\\n100 percent surveys. At the time the travel management\\nplans were decided, the Forest Service had surveyed 42\\npercent of the Williams Ranger District, 23 percent of the\\nTusayan Ranger District, and 25 percent of the North Kaibab\\nRanger District. It concluded that no further surveying was\\nrequired, based on the expected density of cultural resources\\nand the nature and severity of impacts upon them.\\n\\n The record reinforces the determination that the expected\\nimpacts of the travel management plans would be low. After\\nconsidering historic hunting data, the EAs concluded that\\nless than 0.1 percent of each District’s acreage would be\\n\\f38 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nimpacted by motorized big game retrieval. However, the\\nfact that “[t]he expected nature and severity of all associated\\nimpacts” might have been low does not necessarily mean\\nthat the Forest Service did not need to conduct 100 percent\\nsurveys, for the Programmatic Agreement required\\n100 percent surveys for high-density sites regardless of\\nimpact level. Accordingly, the Forest Service’s obligation\\nrested on “[t]he expected nature and distribution of heritage\\nresources.”\\n\\n The record is somewhat unclear as to the density of\\nheritage resources in the areas open to motorized big game\\nretrieval. Plaintiffs note that many of the Districts’ cultural\\nresources are close to the roads, but that does not necessarily\\nspeak to the density of resources because the Programmatic\\nAgreement does not provide any benchmark or guidance as\\nto what constitutes a high-density site. Ultimately, the lack\\nof clarity weighs in the Forest Service’s favor: it was not\\nunreasonable for it to determine that the density was low or\\nunknown, either of which would have excused 100 percent\\nsurveys. 9 Thus, the Forest Service followed a reasonable\\ninterpretation of the Programmatic Agreement.\\n\\n B. Exemption Q\\n\\n The Programmatic Agreement’s Exemption Q provided\\nthat “[a]ctivities not involving ground or surface disturbance\\n(e.g., timber stand improvement and precommercial thinning\\nby hand)” are “exempt from further review and\\nconsultation.” Plaintiffs argue that the Forest Service\\n 9\\n The Programmatic Agreement required 100 percent surveys\\n“where site densities are unknown and expected visitor use or impacts\\nwill be high,” so it would have been reasonable for the Forest Service to\\nconclude that 100 percent surveys were not needed where densities were\\nunknown and use impacts were low.\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 39\\n\\narbitrarily relied on Exemption Q to excuse review of\\nmotorized big game retrieval in the Williams and Tusayan\\nRanger Districts.\\n\\n Once again, however, the record does not support\\nPlaintiffs’ assertions. We agree with Plaintiffs that invoking\\nExemption Q would have been inappropriate here. It is\\nclear, and the Forest Service does not dispute, that motorized\\ntravel causes surface disturbance. But although the Forest\\nService’s correspondence with non-party Center for\\nBiological Diversity suggested that it applied Exemption\\nQ, 10 the record as a whole supports a contrary conclusion.\\nThe Forest Service consulted with both the Arizona SHPO\\nand potentially affected tribes as to each District’s travel\\nmanagement plan—consultations that would not have been\\nrequired if Exemption Q had been applied. Moreover, the\\nForest Service made no other references to Exemption Q as\\npart of the travel management plan decision process, which\\nfurther supports the conclusion that it was not in fact\\ninvoked. Accordingly, the references to Exemption Q at\\n\\n 10\\n The Forest Service claims that Plaintiffs’ “brief does not identify\\nwhere or how the Tusayan and Williams decisions purportedly relied on\\nExemption Q,” and that their “responses to the non-party Center for\\nBiological Diversity’s administrative appeals of those decisions” contain\\nonly “an isolated reference to Exemption Q,” but these assertions are\\nmisleading. In its response to the appeal regarding the Tusayan Ranger\\nDistrict, the Forest Service wrote that it “determined that motorized big\\ngame retrieval fell under Exemption Q of the PA, ‘Activities not\\ninvolving ground or surface disturbance (e.g., timber stand improvement\\nand precommercial thinning by hand)’ and would have limited impacts\\nsimilar to the examples cited in the exemption in the” Programmatic\\nAgreement. That is more than a mere “isolated reference,” as it implies\\nthat the Forest Service actually applied Exemption Q to the Tusayan\\nRanger District. This same language invoking Exemption Q appeared in\\nthe Forest Service’s response regarding the Williams Ranger District.\\n\\f40 WILDEARTH GUARDIANS V. PROVENCIO\\n\\nmost amounted to harmless error, as they had no effect on\\nthe NHPA consultation process. See 5 U.S.C. § 706\\n(requiring the court to take “due account . . . of the rule of\\nprejudicial error” when conducting APA review); Tucson\\nHerpetological Soc’y v. Salazar, 566 F.3d 870, 880 (9th Cir.\\n2009) (“We have held that the harmless error doctrine ‘may\\nbe employed only “when a mistake of the administrative\\nbody is one that clearly had no bearing on the procedure\\nused or the substance of decision reached.”’” (quoting\\nGifford Pinchot Task Force v. U.S. Fish & Wildlife Serv.,\\n378 F.3d 1059, 1071 (9th Cir. 2004))). 11\\n\\n C. “No Adverse Effect” Determination\\n\\n Finally, Plaintiffs argue that the Forest Service’s\\nconclusion that motorized big game retrieval would have no\\nadverse effect on cultural resources was arbitrary. But to\\nbuttress this assertion, Plaintiffs again cherry pick isolated\\nsegments from the record without considering their broader\\ncontext. For example, the Cultural Resources Specialist\\nReport prepared for the North Kaibab Ranger District noted\\nthat “[c]ross country motorized travel, whether to retrieve\\ngame or for other purposes, can adversely affect cultural\\nresource sites if a vehicle is driven across a site,” since\\n“[v]ehicles can [] crush or displace artifacts and features\\n\\n 11\\n Curiously, although the Forest Service similarly argued in the\\ndistrict court that “the agency’s decisions did not rely on Exemption Q,”\\nwe note that the court concluded that Exemption Q did apply. On appeal,\\nthe Forest Service does not argue that the district court’s conclusion on\\nthis point was correct, and we can affirm the court’s ruling even though\\nwe agree with the parties that Exemption Q was not in fact applicable.\\nSee Applied Underwriters, Inc. v. Lichtenegger, 913 F.3d 884, 892 (9th\\nCir. 2019) (“We will affirm the district court’s correct legal results, even\\nif reached for the wrong reasons.” (quoting Alcaraz v. Block, 746 F.2d\\n593, 602 (9th Cir. 1984))).\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 41\\n\\nimpacting the physical integrity of the site and impairing or\\ndestroying scientific information that may contribute to the\\nunderstanding of the history and prehistory of an area.” But\\nthat same report also concluded as follows:\\n\\n Quantifying the potential for damage from\\n big game retrieval is difficult. The results\\n vary depending on the number of game\\n retrieval trips annually, the location of those\\n retrievals (high site probability areas versus\\n low probability), site types found in the area,\\n soil characteristics, routes used to access the\\n game and weather conditions at the time of\\n retrieval. However, the fewer number of\\n motorized trips that occur, the lower the\\n likelihood of encountering and impacting a\\n site. . . .\\n\\n Limiting cross-country travel will have a\\n beneficial effect on cultural resources by\\n reducing the potential for sites to be\\n damaged. This alternative would restrict\\n motorized big game retrieval to elk and\\n bison. In 2009, 38 buffalo and no elk were\\n taken . . . . While there is a possibility that\\n cross-country game retrieval of either of\\n these species could impact a cultural resource\\n site, given the low number of takes each year,\\n it is anticipated that the potential for adverse\\n effects to a site would be negligible:\\n 38 entries per year equates to less than\\n .0099% of the acreage on the NKRD. The\\n odds of adversely affecting a cultural\\n resource site under these conditions are\\n extremely low.\\n\\f42 WILDEARTH GUARDIANS V. PROVENCIO\\n\\n(emphases added). In light of this ultimate conclusion, as\\nwell as the implementation of the suggested mitigation\\nmeasures, it was not arbitrary for the Forest Service to\\nconclude, based on the evidence before it, that adverse\\neffects were unlikely. See Motor Vehicle Mfrs. Ass’n of\\nU.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43\\n(1983) (“[A]n agency rule would be arbitrary and capricious\\nif the agency . . . offered an explanation for its decision that\\nruns counter to the evidence before the agency.”).\\n\\n D. Summation\\n\\n The NHPA—and NEPA—“create obligations that are\\nchiefly procedural in nature.” San Carlos Apache Tribe,\\n417 F.3d at 1097 (quoting Pres. Coal., Inc. v. Pierce,\\n667 F.2d 851, 859 (9th Cir. 1982)). Even if cultural\\nresources might be harmed as a result of motorized big game\\nretrieval, that fact alone does not indicate that the Forest\\nService violated the NHPA. The Forest Service conducted\\nthe required prefield work, consulted with the appropriate\\nentities, and reached a determination consistent with the\\nevidence before it—in short, satisfied its procedural\\nobligations.\\n\\n CONCLUSION\\n\\n The travel management plans that the Forest Service\\nimplemented in the three Ranger Districts of the Kaibab\\nNational Forest limited motorized big game retrieval to\\ncertain roads and imposed additional restrictions to reduce\\nthe level and effect of motorized activity. In crafting the\\nplans, the Forest Service investigated potential impacts on\\nboth the environment and historic properties and reasonably\\ndetermined that no further action was needed. Accordingly,\\nwe conclude that the Forest Service followed the Travel\\n\\f WILDEARTH GUARDIANS V. PROVENCIO 43\\n\\nManagement Rule and fulfilled its procedural obligations\\nunder NEPA and the NHPA. 12\\n\\n AFFIRMED.\\n\\n\\n\\n\\n 12\\n Amicus Curiae Rocky Mountain Elk Foundation filed a motion\\nfor leave to file an amicus brief in support of the Forest Service. We\\nconclude that the proposed brief provides neither legal nor factual\\nsupport to help resolve the issues on appeal, and so deny the motion.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1284,"cells":{"id":{"kind":"number","value":4599150,"string":"4,599,150"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Mickey Fowler v. Tracy Guerin"},"decision_date":{"kind":"string","value":"2019-03-13"},"docket_number":{"kind":"string","value":"16-35052"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Ronald M. Gould and Sandra S. Ikuta, Circuit Judges, and John R. Tunheim, * Chief District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"Judge for the District of Minnesota, sitting by designation\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n MICKEY FOWLER; LEISA MAURER, No. 16-35052\\n and a class of similarly situated\\n individuals, D.C. No.\\n Plaintiffs-Appellants, 3:15-cv-05367-\\n BHS\\n v.\\n\\n TRACY GUERIN, Director of the ORDER\\n Washington State Department of\\n Retirement Systems,\\n Defendant-Appellee.\\n\\n\\n\\n Filed March 13, 2019\\n\\n Before: Ronald M. Gould and Sandra S. Ikuta, Circuit\\n Judges, and John R. Tunheim, * Chief District Judge.\\n\\n Order;\\n Dissent by Judge Bennett\\n\\n\\n\\n\\n *\\n The Honorable John R. Tunheim, Chief United States District\\nJudge for the District of Minnesota, sitting by designation.\\n\\n2 FOWLER V. GUERIN\\n\\n SUMMARY **\\n\\n\\n Civil Rights\\n\\n The panel denied a petition for panel rehearing and\\ndenied a petition for rehearing en banc on behalf of the court.\\n\\n In the underlying opinion, the panel reversed the district\\ncourt’s denial of a stipulated motion to certify a class and\\ndismissal, as prudentially unripe, of an action brought by\\nWashington public school teachers seeking an order that the\\nDirector of Washington State Department of Retirement\\nSystems return daily interest that was allegedly wrongfully\\nwithheld from plaintiffs’ state-managed retirement accounts.\\nThe panel held that the district court erred in dismissing\\nplaintiffs’ takings claim as prudentially unripe because the\\nwithholding of interest that had accrued on plaintiffs’\\naccounts constituted a per se taking, as to which the\\nprudential ripeness test did not apply. The panel further held\\nthat the plaintiffs’ claim could be certified for class treatment\\nunder Fed. R. Civ. P. 23(b)(2) because the relief sought of\\ncorrecting the records system for the class members’\\naccounts was in the nature of injunctive relief.\\n\\n Dissenting from the denial of rehearing en banc, Judge\\nBennett, joined by Judge R. Nelson as to Part III, stated that\\nthe merits panel wrongfully stripped the State of Washington\\nof its Eleventh Amendment immunity from suit by\\npermitting a damages claim to proceed against the State\\nunder the guise of an injunction. Judge Bennett further\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n FOWLER V. GUERIN 3\\n\\nstated the panel erred in concluding that Washington’s\\ndecision to abrogate the common law rule of daily interest\\nviolated the Takings Clause.\\n\\n\\n COUNSEL\\n\\nStephen K. Festor (argued), Stephen K. Strong, David F.\\nStobaugh, and Alexander F. Strong, Bendich Stobaugh &\\nStrong P.C., Seattle, Washington, for Plaintiffs-Appellants.\\n\\nJeffrey A.O. Freimund (argued) and Michael E. Tardif,\\nFreimund Jackson & Tardif PLLC, Olympia, Washington;\\nPeter Gonick, Deputy Solicitor General; Robert W.\\nFerguson, Attorney General; Office of the Attorney General,\\nOlympia, Washington; for Defendant-Appellee.\\n\\n\\n ORDER\\n\\n The panel, as constituted above, has unanimously voted\\nto deny the petition for panel rehearing. Judges Gould and\\nIkuta voted to deny the petition for rehearing en banc, and\\nJudge Tunheim has so recommended.\\n\\n The full court was advised of the petition for rehearing\\nen banc. A judge requested a vote on whether to rehear the\\nmatter en banc, and the matter failed to receive a majority of\\nthe votes of the nonrecused active judges in favor of en banc\\nconsideration. Fed. R. App. P. 35. The petition for panel\\nrehearing and the petition for rehearing en banc are\\nDENIED.\\n\\n4 FOWLER V. GUERIN\\n\\nBENNETT, Circuit Judge, with whom R. NELSON, Circuit\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4376403/\", \"author_raw\": \"Judge for the District of Minnesota, sitting by designation\"}, {\"author\": \"Judge, joins as to Part III, dissenting\", \"type\": \"dissent\", \"text\": \"Judge, joins as to Part III, dissenting from the denial of\\nrehearing en banc:\\n\\n I respectfully dissent from our decision not to rehear this\\ncase en banc. I believe that the panel made two fundamental\\nerrors of enormous scope, both of which we should have\\ncorrected en banc.\\n\\n First, the panel has wrongfully stripped the State of\\nWashington of its Eleventh Amendment immunity from suit\\nby permitting a damages claim to proceed against the State\\nunder the guise of an injunction requiring the State to return\\nto Plaintiffs “their” property. The property was never\\nPlaintiffs’, and, in any case, is simply money—uncredited\\ninterest that will now be paid to Plaintiffs from the State’s\\ntreasury. That decision, which contravenes clear Supreme\\nCourt and Ninth Circuit precedent and creates a circuit split,\\nstrips the Eleventh Amendment of much of its vitality. It\\ntakes little in the way of imagination to foresee future\\nplaintiffs recasting their otherwise-barred claims for money\\ndamages against a state as injunctive relief claims for return\\nof what is supposedly their property.\\n\\n Having bypassed Washington’s immunity from suit, the\\npanel then created a Fifth Amendment property right no\\ncourt has ever recognized. According to the panel, when a\\nstate chooses to hold individuals’ funds in an interest-\\nbearing account, that account must, constitutionally, accrue\\ninterest day-to-day, because that was the way the common\\nlaw worked in centuries past:\\n\\n Because the right to daily interest is deeply\\n ingrained in our common law tradition, this\\n property interest is protected by the Takings\\n Clause regardless of whether a state\\n\\n FOWLER V. GUERIN 5\\n\\n legislature purports to authorize a state\\n officer to abrogate the common law.\\n\\nFowler v. Guerin, 899 F.3d 1112, 1119 (9th Cir. 2018).\\n\\n In other words, neither the Washington legislature, nor\\nthe legislatures of its sister states, nor even Congress, may\\nconstitutionally allow interest to accrue weekly, monthly, or\\nannually on retirement (or other) accounts they establish by\\nstatute. The panel’s decision is wholly untethered to the text\\nof the Fifth Amendment and unsupported by any case. Many\\nstates and the United States currently have retirement\\nsystems with interest-bearing accounts that, just like\\nWashington’s, do not accrue interest daily. If the panel is\\ncorrect, these states and the United States are all currently\\nviolating the Fifth Amendment and have been for decades.\\n\\n Both of the panel’s errors—stripping Washington of its\\nconstitutional immunity from suit and creating a never-\\nbefore-recognized constitutional right—independently\\nwarrant rehearing en banc. Thus I respectfully dissent.\\n\\n I\\n\\n I start with a bit of background. Washington State public\\nschool teachers participate in the Teachers Retirement\\nSystem, which is a part of the Public Employees Retirements\\nSystem (“PERS”). This case concerns PERS Plan II, a\\ndefined benefit retirement plan. Wash. Rev. Code\\n§ 41.32.760. “A defined-benefit plan gives current and\\nformer employees property interests in their pension benefits\\nbut not in the assets held by the trust.” Johnson v. Ga.-Pac.\\nCorp., 19 F.3d 1184, 1189 (7th Cir. 1994).\\n\\n To fund Plan II benefits, participants and their employers\\nmake monthly contributions throughout their employment,\\n\\n6 FOWLER V. GUERIN\\n\\nand their individual accounts reflect those contributions. Id.\\n§ 41.45.050. However, a state agency maintains the funds in\\na comingled account that is not itself interest-bearing. Id.\\n§§ 41.50.077, 080. Rather, the State invests the funds, and\\nthose investments have a return of about eight percent\\nannually. The State uses contributions and investment\\nreturns to pay benefits to participants upon retirement.\\n\\n Washington law requires the Director of the Department\\nof Retirement Systems (the “Director”) to “make an\\nallowance of regular interest” on the participants’ PERS Plan\\nII contributions, Wash. Rev. Code § 41.50.215, and defines\\n“regular interest” as “such rate as the director may\\ndetermine,” id. § 41.32.010(38). The Washington legislature\\nexpressly “affirms that the authority of the director . . .\\nincludes the authority and responsibility to establish the\\namount and all conditions for regular interest, if any.” Id.\\n§ 41.50.033(3). The Director thus has complete statutory\\n“authority to determine how interest is earned.” Probst v.\\nState Dep’t of Ret. Sys., 271 P.3d 966, 970 (Wash. Ct. App.\\n2012) (emphasis added). For more than forty years, PERS\\nPlan II accounts earned interest quarterly, and “do[] not\\n‘earn’ or accrue regular interest on a day by day basis.”\\nWash. Admin. Code § 415-02-150(5). Where a withdrawal\\nor transfer of a participant’s funds takes place mid-quarter,\\nno interest accrues on the funds between the end of the\\nprevious quarter and the date of the withdrawal or transfer.\\nThe Washington Court of Appeals has stated as a definitive\\nmatter of state law: “The legislature’s intent to abrogate the\\ndaily interest rule . . . is plainly evident.” Probst, 271 P.3d\\nat 971.\\n\\n Because only tenure and yearly compensation define\\nPlan II participants’ retirement benefits, the amount of\\nmoney in an individual participant account becomes\\n\\n FOWLER V. GUERIN 7\\n\\nimmaterial upon the participant’s retirement and eligibility\\nfor benefits. But if a participant leaves service early and\\nwithdraws his or her contributions, or transfers them to a\\ndifferent retirement fund, the participant receives (or\\ntransfers) the amount shown in the individual account.\\nOtherwise, “PERS . . . employees have no claim on the fund\\nuntil they complete their term of employment and qualify for\\na pension.” Bowles v. Wash. Dep’t of Ret. Sys., 847 P.2d 440,\\n454 (Wash. 1993) (en banc).\\n\\n Here, Plaintiffs are Washington teachers who\\nparticipated in PERS II before transferring their PERS II\\naccounts into a new plan where the accounts became seed\\nmoney for an employee investment account. Because\\nPlaintiffs’ account transfers took place mid-quarter, their\\naccounts did not earn any interest between the end of the\\nprevious quarter and the date of transfer. Plaintiffs’ lawsuit\\nseeks to recover that purportedly “taken” interest.\\n\\n With this background in mind, I turn to discuss the two\\nareas of the panel’s opinion that I believe should have been\\naddressed en banc.\\n\\n II\\n\\n “The Eleventh Amendment confirms that the\\nfundamental principle of sovereign immunity limits the\\ngrant of judicial authority in Art. III.” Green v. Mansour,\\n474 U.S. 64, 68 (1985) (internal quotation marks omitted).\\nWhere, as here, a plaintiff sues a state official in his or her\\nofficial capacity, sovereign immunity bars the claim. See\\nPennhurst St. Sch. & Hosp. v. Halderman, 465 U.S. 89, 101\\n(1989) (“The Eleventh Amendment bars a suit against state\\nofficials when the state is the real, substantial party in\\ninterest.” (internal quotation marks omitted)); see also Hafer\\nv. Melo, 502 U.S. 21, 25 (1991) (“Suits against state officials\\n\\n8 FOWLER V. GUERIN\\n\\nin their official capacity . . . should be treated as suits against\\nthe State.”).\\n\\n A claim under the Takings Clause seeks “not just\\ncompensation per se but rather damages for the\\nunconstitutional denial of such compensation.” City of\\nMonterey v. Del Monte Dunes at Monterey, Ltd., 526 U.S.\\n687, 710 (1999). Thus “the Eleventh Amendment bars\\nreverse condemnation actions brought in federal court\\nagainst state officials in their official capacities.” Seven Up\\nPete Venture v. Schweitzer, 523 F.3d 948, 956 (9th Cir.\\n2008). Our holding in Seven Up Pete Venture, which is in\\nagreement with every court of appeals to consider the issue,1\\nshould have ended the panel’s analysis. Plaintiffs here are\\nsuing the Director in her official capacity for violations of\\nthe Takings Clause—precisely the sort of claim that we, and\\neach of our sister circuits to consider the issue, have held\\nviolates the Eleventh Amendment immunity that the states\\nenjoy. By permitting the Plaintiffs’ claims to proceed, the\\npanel departs from this long and heretofore unbroken line of\\nauthority.\\n\\n A\\n\\n By construing Plaintiffs’ claim as seeking an\\n“injunction,” the panel tries to shoehorn the claim into Ex\\nparte Young’s narrow Eleventh Amendment exception for “a\\nsuit for prospective relief against a state official in his\\nofficial capacity” to correct an ongoing violation of the\\n\\n 1\\n See, e.g., Hutto v. S.C. Ret. Sys., 773 F.3d 536, 552 (4th Cir. 2014);\\nDLX, Inc. v. Kentucky, 381 F.3d 511, 526 (6th Cir. 2004); John G. &\\nMarie Stella Kenedy Mem. Found. v. Mauro, 21 F.3d 667, 674 (5th Cir.\\n1994); Robinson v. Ga. Dep’t of Transp., 966 F.2d 637, 638–39 (11th\\nCir. 1992); Garrett v. Illinois, 612 F.2d 1038, 1040 (7th Cir. 1980);\\nCitadel Corp. v. P.R. Highway Auth., 695 F.2d 31, 33 n.4 (1st Cir. 1982).\\n\\n FOWLER V. GUERIN 9\\n\\nConstitution. Cardenas v. Anzai, 311 F.3d 929, 934–35 (9th\\nCir. 2002) (citing Ex parte Young, 209 U.S. 123, 159–60\\n(1908)). Edelman v. Jordan, 415 U.S. 651 (1974) taught us\\nlong ago that plaintiffs cannot sidestep the Eleventh\\nAmendment merely by using forward-looking labels to\\nachieve what is, in essence, a backwards-looking result.\\n\\n Ex parte Young is inapplicable where the relief sought\\n“is measured in terms of a monetary loss resulting from a\\npast breach of a legal duty on the part of the defendant state\\nofficials,” Edelman, 415 U.S. at 668, or where “the state is\\nthe real, substantial party in interest . . . as when the\\njudgment sought would expend itself on the public treasury\\nor domain, or interfere with public administration,” Va.\\nOffice for Protection & Advocacy v. Stewart, 563 U.S. 247,\\n255 (2011) (internal citation and quotation marks omitted).\\nUnder Edelman and Stewart, Plaintiffs’ claims clearly do not\\nfall within the Ex parte Young exception. Regardless of the\\nprospective label that Plaintiffs give their claim, it is\\nfunctionally retrospective, and the Supreme Court\\ncommands us to treat it that way. Edelman, 415 U.S. at 668.\\n\\n 1\\n\\n The panel holds that the relief sought is prospective\\nbecause the Plaintiffs are merely seeking an injunction for\\nthe return of money that the Director “skimmed” from their\\naccounts. Fowler, 899 F.3d at 1120. This characterization\\nincorrectly assumes that the accounts in fact accrued interest\\nthat the Director then took from the Plaintiffs. As discussed\\nsupra p. 6, though, the Plaintiffs’ accounts “do[] not ‘earn’\\nor accrue regular interest on a day by day basis.” Wash.\\nAdmin. Code § 415-02-150(5). Because the interest never\\nexisted until credited by the Director (and here the Plaintiffs’\\nactual claimed constitutional violation is the failure to\\ncredit), Plaintiffs cannot claim that the Director wrongly\\n\\n10 FOWLER V. GUERIN\\n\\ntook it from them. Properly understood, the Plaintiffs’ claims\\nare for money supposedly owed to them, not money actually\\ntaken from them—a critical distinction for Eleventh\\nAmendment purposes. See Edelman, 415 U.S. at 668 (stating\\nthat where “equitable restitution” “is measured in terms of a\\nmonetary loss resulting from a past breach of a legal duty on\\nthe part of the . . . state,” “it is in practical effect\\nindistinguishable in many aspects from an award of damages\\nagainst the State”).\\n\\n By asking the district court to order the state to pay\\nmoney it allegedly owes but withheld from them, Plaintiffs\\nseek a purely retrospective damages award.\\n\\n 2\\n\\n Ex parte Young is also inapplicable here because the\\nState, not the Director, is the real party in interest. As in\\nEdelman, the “restitution award” “will to a virtual certainty\\nbe paid from state funds, and not from the pockets of the\\nindividual state officials who were the defendants in this\\naction.” 415 U.S. at 668. The Director of the Washington\\nDRS is, of course, not personally liable—the money at issue\\nwill have to come from the State.\\n\\n And although the panel opinion hardly addresses the\\nState treasury’s liability—“the most salient factor in\\nEleventh Amendment determinations,” Hess v. Port\\nAuthority Trans-Hudson, 513 U.S. 30, 48 (1994)—the\\nrecord here shows clearly that the state treasury will be liable\\nfor any award to the Plaintiffs, whether or not the court calls\\nthe award an injunction.\\n\\n Plaintiffs’ employers (local school districts) make\\nemployer contributions to the PERS II fund, and those\\ndistricts receive their funding for employee benefits directly\\n\\n FOWLER V. GUERIN 11\\n\\nfrom the State. See Bowles, 847 P.2d at 450 (noting that\\nwhere a state retirement plan has a defined-benefits\\nstructure, “employer contributions must be increased to\\nwhatever level becomes necessary to fund the statutorily\\ndefined benefits” and thus “all risk of a shortfall rests on state\\nand local government employers and ultimately, on\\ntaxpayers”). If Plaintiffs get their “injunction” and receive\\nmoney from the PERS II fund, someone (the State) will have\\nto provide the money needed to replenish the fund.\\n\\n The panel says that the State treasury will be safe from a\\njudgment in Plaintiffs’ favor because the relief that Plaintiffs\\nseek is simply interest that accrued on Plaintiffs’ accounts\\nbut that the State did not credit to them. Fowler, 899 F.3d\\nat 1120. Again, this misstates Washington law: the “taken”\\ninterest is not in the PERS II fund because it never came into\\nexistence to begin with. But even if that were not the case\\nand Plaintiffs sought their own money that sits in the wrong\\nretirement fund, that money is being used to fund PERS II\\nretirement benefits, and the State, to meet its PERS II\\nobligations, would still have to replace the amounts\\ntransferred with money from the treasury.\\n\\n Plaintiffs’ contention that the State treasury will not be\\nthe immediate source of funding for a judgment in their favor\\nmisses the mark. We have found sovereign immunity to\\napply even where “the state is not directly liable for a\\njudgment against [the named defendant].” Alaska Cargo\\nTransp., Inc. v. Alaska R.R. Corp., 5 F.3d 378, 381 (9th Cir.\\n1993) (barring, on sovereign immunity grounds, a suit\\nagainst a partially state-funded railroad because “state law\\nprovides to [the railroad] a financial safety net of broad\\ndimension”); see also Morris v. Wash. Metro. Area Transit\\nAuth., 781 F.2d 218, 225–26 (D.C. Cir. 1986) (“Given the\\npracticalities of Maryland and Virginia’s financial\\n\\n12 FOWLER V. GUERIN\\n\\ncommitments to WMATA, a judgment against WMATA\\nwould directly affect the treasuries of Maryland and\\nVirginia.”). Likewise here, the State is statutorily obligated\\nto adequately fund the retirement accounts at issue, and a\\njudgment in Plaintiffs’ favor that requires a debit from the\\nPERS II account would clearly require the State to expend\\nadditional funds to cover the difference.\\n\\n B\\n\\n The sole case on which the panel relies to hold that the\\nPlaintiffs’ claim falls outside the Eleventh Amendment’s\\nambit is Taylor v. Westly, 402 F.3d 924 (9th Cir. 2005). But\\nTaylor involved an escheat statute whereby supposedly\\nabandoned property was seized by the State of California\\nand held in express trust for the property’s owners. Id. at\\n931–32. We ultimately allowed only the plaintiffs’ due\\nprocess claims to proceed, reasoning that “[m]oney that the\\nstate holds for the benefit of private individuals is not the\\nstate’s money, any more than towed cars are the state’s cars.”\\nId. at 932.\\n\\n In the years since we decided Taylor, we have essentially\\nlimited its application to escheat statutes. See N.E. Med.\\nServs., Inc. v. Cal. Dep’t of Health Care Servs., 712 F.3d\\n461, 469 (9th Cir. 2013). To paraphrase Northeast Medical\\nServices, here, unlike in Taylor, the State “did not receive\\nthe [interest] pursuant to a unique statutory scheme. There is\\nno [Washington] law requiring the state to hold the [interest]\\nin a custodial trust. Any monetary award to the [Plaintiffs]\\nwould necessarily come from the state treasury.” Id. 2 Taylor\\n\\n 2\\n Undergirding the panel’s discussion of Taylor is the apparent\\nassumption that the allegedly missing interest is held in trust for the\\nPlaintiffs’ benefit. See Fowler, 899 F.3d at 1120 (“Washington’s\\n\\n FOWLER V. GUERIN 13\\n\\nsimply does not shoulder the weight that the panel places\\nupon it.\\n\\n * * *\\n\\n The ruling here strikes at the very heart of the federalism\\ninterests the Eleventh Amendment was designed to protect.\\nNot just Washington, but its sister states as well, will no\\ndoubt read this decision for what it is—an invitation to\\nplaintiffs with money claims against states to press those\\nclaims in federal court, the Eleventh Amendment\\nnotwithstanding. We should have taken this case en banc to\\nwithdraw that invitation.\\n\\n III\\n\\n The panel erred in concluding that the Washington\\nlegislature’s unremarkable decision to abrogate the common\\nlaw rule of daily interest violated the Takings Clause. This\\ndecision has far-reaching consequences for other\\ngovernment pension plans, like those established by the\\nUnited States and states in and outside the Ninth Circuit that\\ncredit interest less frequently than daily.\\n\\n A\\n\\n The panel held that the Plaintiffs have a constitutionally\\nprotected property interest in daily interest earnings,\\n\\n\\n\\nsovereign immunity [does not] shield[] investment funds held for the\\nbenefit of its employees.”). Not so. The relevant Washington retirement\\naccounts “are not trusts.” Retired Pub. Emp. Council of Wash. v. Charles,\\n62 P.3d 470, 481 (Wash. 2003) (en banc).\\n\\n14 FOWLER V. GUERIN\\n\\nnotwithstanding clear state law to the contrary. 3 This holding\\nis unprecedented. As far as I know, no court has held that\\nwhen a state establishes and holds a retirement (or other\\naccount) for someone, and chooses to pay interest on that\\naccount, the owner of the funds has a constitutional right to\\ndaily interest that a state cannot abrogate. In reaching this\\nconclusion the panel ignored Supreme Court guidance\\npermitting states “great latitude” in awarding interest,\\nmisapplied the “interest follows principal” rule, and\\nimproperly created a new property right to daily interest.\\n\\n 1\\n\\n Assuming the Plaintiffs have an ownership interest in the\\nprincipal in their individual accounts, it does not follow they\\nare entitled to daily interest. The Supreme Court has\\nrecognized that state governments have “great latitude in\\nregulating the circumstances under which interest may be\\nearned.” Phillips v. Wash. Legal Found., 524 U.S. 156, 168\\n(1998). The panel disregards the traditional discretion\\nafforded to the states and holds, instead, that when a state\\nawards interest, it must do so on a daily basis. As explained\\nbelow, though, neither the panel nor the Plaintiffs identify\\nany authority for the proposition that when a state decides to\\nprovide some amount of interest, it must, as a matter of\\nconstitutional law, do so daily.\\n\\n In this case, the Director has done what the Court has\\npermitted: “regulating the circumstances under which\\ninterest may be earned.” Id. Pursuant to her statutory\\n 3\\n It is an odd constitutional right the panel creates. Even were the\\npanel correct, there is no right to interest at any particular rate. So,\\nWashington could, for example, even under the panel’s view of the law,\\nprovide for 0.01% interest, compounded daily, but not for ten percent\\ninterest, compounded monthly, quarterly, or annually.\\n\\n FOWLER V. GUERIN 15\\n\\nauthority (Wash. Rev. Code § 41.50.033(3)), the Director\\nhas defined the Teachers’ property rights with respect to\\ninterest in the Plan II account: “if the amount in your\\nindividual account on the last day of a quarter is more than\\nzero dollars, the department will calculate an amount of\\nregular interest to be credited to your account”; that account\\n“does not ‘earn’ or accrue regular interest on a day by day\\nbasis.” Wash. Admin. Code § 415-02-150(3), (5). 4\\n\\n Because the Supreme Court has preserved a state’s right\\nto define how it pays interest, and by extension, the property\\nrights related to how interest is earned, the panel erred in\\nconcluding that the Plaintiffs state a claim under the Fifth\\nAmendment.\\n\\n 2\\n\\n The panel did not expressly invoke the “interest follows\\nprincipal” rule discussed in Schneider v. California\\nDepartment of Corrections, 151 F.3d 1194, 1199 (9th Cir.\\n1998), but “clarified” Schneider’s holding to conclude the\\nPlaintiffs were entitled to daily interest. Fowler, 899 F.3d\\nat 1118. The panel’s understanding of the interest-follows-\\nprincipal rule, though, is deeply flawed. The rule that the\\nprincipal’s owner is entitled to interest earned thereon does\\nnot mean that all funds in a state account must earn interest,\\nand by extension cannot require a state voluntarily awarding\\ninterest to do so daily.\\n\\n In this case, the Plaintiffs are seeking additional interest\\nearned on a state-held account that the State pools with other\\n\\n 4\\n And again, under Washington law, the statutory right to access an\\nindividual account only accrues if an individual seeks a refund or transfer\\nof contributions—otherwise, the individual’s right is only to a pension.\\n\\n16 FOWLER V. GUERIN\\n\\nindividual accounts in a non-interest bearing fund. The facts\\nof this case therefore fundamentally differ from Schneider,\\nWebb’s Fabulous Pharmacies v. Beckwith, 449 U.S. 155\\n(1980), Phillips, and Brown v. Legal Foundation of\\nWashington, 538 U.S. 216 (2003), all of which involved\\nclaims for the return of interest actually generated in a third-\\nparty bank account. In Schneider, for example, the\\nDepartment of Corrections placed inmate funds into an\\naccount maintained by a third-party, and the inmate’s claims\\ncould proceed only to the extent that those third-party\\naccounts actually bore interest. See 151 F.3d at 1201 (“On\\nremand, the district court shall permit discovery to determine\\nwhether or not interest actually accrues on the prisoners’\\nITA funds.”).\\n\\n The same is true of Phillips and Brown—both cases\\ninvolved Interest on Lawyers Trust Account (“IOLTA”)\\nprograms, and considered whether the state committed a\\ntaking by mandating that interest actually generated on a\\nlawyer’s client trust fund (which was generated in a bank or\\nother financial institution) be used for charitable purposes.\\nLikewise, Webb’s involved interest actually earned in an\\naccount maintained at a local bank by the clerk of court.\\n449 U.S. at 157 n.1. In all three cases interest actually\\naccrued to accounts because of the contractual relationship\\nbetween the depositor and the financial institution that held\\nthe principal for the attorney. See Phillips, 524 U.S. at 159\\n(assessing ownership of interest generated in an IOLTA fund\\nheld by a bank); Brown, 538 U.S. at 228–29 (determining\\nwhether an attorney and clients stated a takings claim for\\nIOLTA interest); Tex. State Bank v. United States, 423 F.3d\\n1370, 1379 (Fed. Cir. 2005) (“In contrast to Webb’s,\\nPhillips, and Brown, where the deposited funds were held by\\nthird party banks, here Texas State did not provide funds to\\n\\n FOWLER V. GUERIN 17\\n\\na third party that were then deposited in an interest-bearing\\naccount in a private bank[.]”).\\n\\n Because the pooled PERS funds here do not bear interest,\\nthe individual account holders cannot use the “interest\\nfollows principal” rule to claim a constitutional right to a\\nshare of that non-existent interest. Rather, their entitlement\\nto interest arises entirely from Washington law.\\n\\n Under Washington law, the individual accounts, which\\nemployees can access to withdraw or transfer funds, bear\\ninterest (at a rate of 5.5% with the accrual and compounding\\nrules set by statute and the Director). But the “interest\\nfollows principal” cases neither hold, nor suggest, that where\\na state has discretion whether to award interest on a\\nretirement account, and chooses to do so, it offends the\\nTakings Clause by doing so less frequently than daily.\\n\\n 3\\n\\n Even if the panel was correct in holding that (1) the\\ncontributions in the Plan II account can form the basis for an\\nindependent claim on the earnings of that account; and\\n(2) the Plaintiffs had a property interest in the Plan II\\naccount, the panel was still wrong to hold the State could not\\nstatutorily modify the common law daily-interest rule, as the\\nProbst court found the State had, based on seventy-five years\\nof Washington statutes.\\n\\n The only basis the panel provided for its holding that\\ninterest must accrue daily is the “impressive common law\\npedigree” of the daily interest rule. Fowler, 899 F.3d at 1118.\\nIt may be that interest de die in diem was the default at\\ncommon law, but states are free to modify common law\\ndefault rules, and the panel never explains why this rule is\\nany different.\\n\\n18 FOWLER V. GUERIN\\n\\n “Property interests . . . are created and their dimensions\\nare defined by existing rules or understandings . . . that\\nsecure certain benefits and support certain claims of\\nentitlement to those benefits.” Bd. of Regents of State\\nColleges v. Roth, 408 U.S. 564, 577 (1972). “But not all\\neconomic interests are ‘property rights’; only those\\neconomic advantages are ‘rights’ which have the law back\\nof them, and only when they are so recognized may courts\\ncompel others to forbear from interfering with them or to\\ncompensate for their invasion.” United States v. Willow\\nRiver Power Co., 324 U.S. 499, 502 (1945).\\n\\n Even assuming that the panel correctly identified a\\ncommon law rule favoring daily interest, it does not remotely\\nfollow that the rule is immutable and immune from\\nlegislative modification. At common law, the entitlement to\\na proportionate share of an annual rate of payment was\\nhighly dependent on context. Annuities, for example, were\\nearned and paid annually and not apportioned if the\\nannuitant died before the day payment was due. See In re\\nBailey’s Estate, 23 Pa. C. C. 139, 142 (Pa. Orphans’ Ct.\\n1899). Dividends for share of stock in corporations and rent\\nwere similarly not subject to apportionment. See Mann v.\\nAnderson, 32 S.E. 870, 871 (Ga. 1899); Bank of Pa. v. Wise,\\n3 Watts 394, 403 (Pa. 1834).\\n\\n I am unaware of any court to hold that a state violates the\\nFifth Amendment by statutorily modifying any of these\\ncommon law rules. To the contrary, cases cited in the panel\\nopinion suggest that a state could permissibly do so. See\\nMann, 32 S.E. at 871 (“Interest was apportionable at\\ncommon law because it was held to accrue de die in diem,\\nand therefore to be susceptible of intermediate division. This\\nis the rule of the common law, and there is no statutory force\\nof law in this state which changes this rule in reference to\\n\\n FOWLER V. GUERIN 19\\n\\ndividends declared on stock in corporations.” (emphasis\\nadded)); see also Nehls v. Sauer, 93 N.W. 346, 347 (Iowa\\n1903) (observing that Iowa modified by statute the common\\nlaw rule against apportionment in the case of life tenancies\\nbut not annuities); Edwin A. Howes, Jr., The American Law\\nRelating to Income and Principal, 73–74 (Little, Brown, &\\nCo. 1905) (identifying states that have, by statute, modified\\nthe rule against apportionment of annuities).\\n\\n It is therefore not enough that the panel identify a\\ncommon law rule that might otherwise govern in the absence\\nof contrary state legislation. The panel must also\\ndemonstrate why the common law rule that interest is\\napportioned daily is so much a fixture of the legal landscape\\nthat the Plaintiffs “have more than an abstract need or desire\\n[or] a unilateral expectation of it,” Roth, 408 U.S. at 577, to\\njustify setting aside otherwise lawful state modification of\\nthe rule. And the fact that no court has, before now, held that\\nstate governments cannot modify the daily interest rule when\\nthey hold cash strongly suggests that the rule is not so deeply\\ningrained in our tradition that states may not modify it\\nwithout running afoul of the Takings Clause.\\n\\n B\\n\\n Rehearing en banc is also warranted here because of the\\ntremendous potential impact of the panel’s incorrect\\ndecision. It is no small thing to hold that a significant aspect\\nof a State’s retirement system is unconstitutional,\\nparticularly when the state has used that system, in some\\nform, since the 1930s. See Probst, 271 P.3d at 972 (citing\\nWashington Laws of 1937, ch. 221 § 1(22)). The impact of\\nthe panel’s decision, though, will be felt well beyond\\nWashington’s borders.\\n\\n20 FOWLER V. GUERIN\\n\\n The panel’s holding will cast significant doubt on the\\nlegitimacy of retirement systems administered by numerous\\nstates and the federal government that apportion interest less\\nfrequently than daily. 5 Congress and the administrators of\\nthe Federal Employees Retirement System (“FERS”) will, I\\nimagine, be very surprised to discover that they are\\ncommitting an unconstitutional taking by failing to pay daily\\ninterest on refunds of employees’ contributions to FERS\\ndefined-benefit plans. 6\\n\\n In addition to Washington and the United States, public\\nemployee retirement systems in Alabama, Alaska,\\nConnecticut, Kansas, Kentucky, South Dakota, Virginia, and\\nWisconsin all apportion interest on retirement account funds\\nless frequently than daily. 7 By the panel’s logic, these states\\n\\n 5\\n Or, for that matter, any account that a private party maintains with\\na state.\\n\\n 6\\n See 5 U.S.C. § 8401(19)(D)(ii); 5 C.F.R. § 841.605(b)(1) (interest\\nbased on number of full months); CSRS and FERS Handbook, Chapter\\n32, § 32B1.1-3(H), p. 28 (available at https://www.opm.gov/retirement-\\nservices/publications-forms/csrsfers-handbook/c032.pdf) (“No interest\\nis paid on a refund of FERS contributions: For a fractional part of a\\nmonth.”).\\n\\n 7\\n See Alaska Public Employees Retirement System Information\\nHandbook, at 6 (available at http://doa.alaska.gov/drb/pdf/pers/handboo\\nk/2011/PERS_handbook_2011_web.pdf) (semi-annual); Alabama\\nEmployees’ Retirement System, Members Handbook, at 9 (available at\\nhttps://www.rsa-al.gov/uploads/files/ERS_Member_Handbook_T2_bo\\nokmarked.pdf) (interest based on previous year’s average balance);\\nConn. Gen. Stat. § 5-166(b)(1) (monthly); Kansas Public Employees\\nRetirement System Application for Withdrawal of Contributions\\n(available at https://www.kpers.org/forms/k13.pdf) (annually or\\nquarterly, depending on plan); Kentucky Employees Retirement System\\nComprehensive Annual Financial Report, 2017, at 39–40 (available at\\nhttps://kyret.ky.gov/Publications/Books/2017%20CAFR%20(Compreh\\n\\n FOWLER V. GUERIN 21\\n\\nare committing an unconstitutional taking, and I have little\\ndoubt that lawyers in these jurisdictions will use the panel’s\\nopinion as a basis for Takings Clause challenges to these\\nretirement plans.\\n\\n IV\\n\\n If the Eleventh Amendment is to continue to have\\nmeaningful force, we cannot permit plaintiffs to attain\\notherwise prohibited retrospective relief against a state’s\\ntreasury simply by describing that relief in terms of an\\ninjunction or other equitable remedy. Nor should we as a\\ncourt create a property right to daily interest when nothing\\nin the precedents of the Supreme Court or this court have\\never even suggested that when a state awards interest, it must\\ndo so daily. The effects of the panel’s novel holding will be\\nfelt around the country in the form of legal challenges to state\\nand federal retirement plans that similarly award interest less\\nfrequently than daily. We should have taken this case en\\nbanc to correct our errors.\\n\\n I respectfully dissent.\\n\\n\\n\\n\\nensive%20Annual%20Financial%20Report).pdf) (annually); S.D.\\nCodified Laws § 3-12-47.8 (annually); Va. Code Ann. § 51.1-147(C)\\n(annually); Wisc. Stat. § 40.04(4)(a)(2), (3) (interest based on previous\\nyear’s closing balance).\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4376403/\", \"author_raw\": \"Judge, joins as to Part III, dissenting\"}]}"},"judge1":{"kind":"string","value":"RONALD M GOULD"},"judge2":{"kind":"string","value":"SANDRA S IKUTA"},"judge3":{"kind":"string","value":"JOHN R TUNHEIM"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":2,"string":"2"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":1,"string":"1"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4599150/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Mickey Fowler v. Tracy Guerin\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-13\nOpinion type: majority\nHas dissenting opinion: Yes\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n MICKEY FOWLER; LEISA MAURER, No. 16-35052\n and a class of similarly situated\n individuals, D.C. No.\n Plaintiffs-Appellants, 3:15-cv-05367-\n BHS\n v.\n\n TRACY GUERIN, Director of the ORDER\n Washington State Department of\n Retirement Systems,\n Defendant-Appellee.\n\n\n\n Filed March 13, 2019\n\n Before: Ronald M. Gould and Sandra S. Ikuta, Circuit\n Judges, and John R. Tunheim, * Chief District Judge.\n\n Order;\n Dissent by Judge Bennett\n\n\n\n\n *\n The Honorable John R. Tunheim, Chief United States District\nJudge for the District of Minnesota, sitting by designation.\n\n2 FOWLER V. GUERIN\n\n SUMMARY **\n\n\n Civil Rights\n\n The panel denied a petition for panel rehearing and\ndenied a petition for rehearing en banc on behalf of the court.\n\n In the underlying opinion, the panel reversed the district\ncourt’s denial of a stipulated motion to certify a class and\ndismissal, as prudentially unripe, of an action brought by\nWashington public school teachers seeking an order that the\nDirector of Washington State Department of Retirement\nSystems return daily interest that was allegedly wrongfully\nwithheld from plaintiffs’ state-managed retirement accounts.\nThe panel held that the district court erred in dismissing\nplaintiffs’ takings claim as prudentially unripe because the\nwithholding of interest that had accrued on plaintiffs’\naccounts constituted a per se taking, as to which the\nprudential ripeness test did not apply. The panel further held\nthat the plaintiffs’ claim could be certified for class treatment\nunder Fed. R. Civ. P. 23(b)(2) because the relief sought of\ncorrecting the records system for the class members’\naccounts was in the nature of injunctive relief.\n\n Dissenting from the denial of rehearing en banc, Judge\nBennett, joined by Judge R. Nelson as to Part III, stated that\nthe merits panel wrongfully stripped the State of Washington\nof its Eleventh Amendment immunity from suit by\npermitting a damages claim to proceed against the State\nunder the guise of an injunction. Judge Bennett further\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\n FOWLER V. GUERIN 3\n\nstated the panel erred in concluding that Washington’s\ndecision to abrogate the common law rule of daily interest\nviolated the Takings Clause.\n\n\n COUNSEL\n\nStephen K. Festor (argued), Stephen K. Strong, David F.\nStobaugh, and Alexander F. Strong, Bendich Stobaugh &\nStrong P.C., Seattle, Washington, for Plaintiffs-Appellants.\n\nJeffrey A.O. Freimund (argued) and Michael E. Tardif,\nFreimund Jackson & Tardif PLLC, Olympia, Washington;\nPeter Gonick, Deputy Solicitor General; Robert W.\nFerguson, Attorney General; Office of the Attorney General,\nOlympia, Washington; for Defendant-Appellee.\n\n\n ORDER\n\n The panel, as constituted above, has unanimously voted\nto deny the petition for panel rehearing. Judges Gould and\nIkuta voted to deny the petition for rehearing en banc, and\nJudge Tunheim has so recommended.\n\n The full court was advised of the petition for rehearing\nen banc. A judge requested a vote on whether to rehear the\nmatter en banc, and the matter failed to receive a majority of\nthe votes of the nonrecused active judges in favor of en banc\nconsideration. Fed. R. App. P. 35. The petition for panel\nrehearing and the petition for rehearing en banc are\nDENIED.\n\n4 FOWLER V. GUERIN\n\nBENNETT, Circuit Judge, with whom R. NELSON, Circuit\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Mickey Fowler v. Tracy Guerin\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-13\\nOpinion type: majority\\nHas dissenting opinion: Yes\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n MICKEY FOWLER; LEISA MAURER, No. 16-35052\\n and a class of similarly situated\\n individuals, D.C. No.\\n Plaintiffs-Appellants, 3:15-cv-05367-\\n BHS\\n v.\\n\\n TRACY GUERIN, Director of the ORDER\\n Washington State Department of\\n Retirement Systems,\\n Defendant-Appellee.\\n\\n\\n\\n Filed March 13, 2019\\n\\n Before: Ronald M. Gould and Sandra S. Ikuta, Circuit\\n Judges, and John R. Tunheim, * Chief District Judge.\\n\\n Order;\\n Dissent by Judge Bennett\\n\\n\\n\\n\\n *\\n The Honorable John R. Tunheim, Chief United States District\\nJudge for the District of Minnesota, sitting by designation.\\n\\n2 FOWLER V. GUERIN\\n\\n SUMMARY **\\n\\n\\n Civil Rights\\n\\n The panel denied a petition for panel rehearing and\\ndenied a petition for rehearing en banc on behalf of the court.\\n\\n In the underlying opinion, the panel reversed the district\\ncourt’s denial of a stipulated motion to certify a class and\\ndismissal, as prudentially unripe, of an action brought by\\nWashington public school teachers seeking an order that the\\nDirector of Washington State Department of Retirement\\nSystems return daily interest that was allegedly wrongfully\\nwithheld from plaintiffs’ state-managed retirement accounts.\\nThe panel held that the district court erred in dismissing\\nplaintiffs’ takings claim as prudentially unripe because the\\nwithholding of interest that had accrued on plaintiffs’\\naccounts constituted a per se taking, as to which the\\nprudential ripeness test did not apply. The panel further held\\nthat the plaintiffs’ claim could be certified for class treatment\\nunder Fed. R. Civ. P. 23(b)(2) because the relief sought of\\ncorrecting the records system for the class members’\\naccounts was in the nature of injunctive relief.\\n\\n Dissenting from the denial of rehearing en banc, Judge\\nBennett, joined by Judge R. Nelson as to Part III, stated that\\nthe merits panel wrongfully stripped the State of Washington\\nof its Eleventh Amendment immunity from suit by\\npermitting a damages claim to proceed against the State\\nunder the guise of an injunction. Judge Bennett further\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n FOWLER V. GUERIN 3\\n\\nstated the panel erred in concluding that Washington’s\\ndecision to abrogate the common law rule of daily interest\\nviolated the Takings Clause.\\n\\n\\n COUNSEL\\n\\nStephen K. Festor (argued), Stephen K. Strong, David F.\\nStobaugh, and Alexander F. Strong, Bendich Stobaugh &\\nStrong P.C., Seattle, Washington, for Plaintiffs-Appellants.\\n\\nJeffrey A.O. Freimund (argued) and Michael E. Tardif,\\nFreimund Jackson & Tardif PLLC, Olympia, Washington;\\nPeter Gonick, Deputy Solicitor General; Robert W.\\nFerguson, Attorney General; Office of the Attorney General,\\nOlympia, Washington; for Defendant-Appellee.\\n\\n\\n ORDER\\n\\n The panel, as constituted above, has unanimously voted\\nto deny the petition for panel rehearing. Judges Gould and\\nIkuta voted to deny the petition for rehearing en banc, and\\nJudge Tunheim has so recommended.\\n\\n The full court was advised of the petition for rehearing\\nen banc. A judge requested a vote on whether to rehear the\\nmatter en banc, and the matter failed to receive a majority of\\nthe votes of the nonrecused active judges in favor of en banc\\nconsideration. Fed. R. App. P. 35. The petition for panel\\nrehearing and the petition for rehearing en banc are\\nDENIED.\\n\\n4 FOWLER V. GUERIN\\n\\nBENNETT, Circuit Judge, with whom R. NELSON, Circuit\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1285,"cells":{"id":{"kind":"number","value":4599151,"string":"4,599,151"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Duanna Knighton v. Cedarville Rancheria of Npi"},"decision_date":{"kind":"string","value":"2019-03-13"},"docket_number":{"kind":"string","value":"17-15515"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: A. Wallace Tashima and Milan D. Smith, Jr., Circuit Judges, and Lawrence L. Piersol, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"Opinion by Judge Piersol\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n DUANNA KNIGHTON, No. 17-15515\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:16-cv-02438-\\n WHO\\n CEDARVILLE RANCHERIA OF\\n NORTHERN PAIUTE INDIANS;\\n CEDARVILLE RANCHERIA TRIBAL OPINION\\n COURT; PATRICIA R. LENZI, in her\\n capacity as Chief Judge of the\\n Cedarville Rancheria Tribal Court,\\n Defendants-Appellees.\\n\\n Appeal from the United States District Court\\n for the Eastern District of California\\n William Horsley Orrick, District Judge, Presiding\\n\\n Argued and Submitted November 16, 2018\\n San Francisco, California\\n\\n Filed March 13, 2019\\n\\n Before: A. Wallace Tashima and Milan D. Smith, Jr.,\\n Circuit Judges, and Lawrence L. Piersol, * District Judge.\\n\\n Opinion by Judge Piersol\\n\\n *\\n The Honorable Lawrence L. Piersol, United States District Judge\\nfor the District of South Dakota, sitting by designation.\\n\\f2 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\n SUMMARY **\\n\\n\\n Tribal Jurisdiction\\n\\n The panel affirmed the district court’s dismissal of an\\naction challenging a tribal court’s subject matter jurisdiction\\nover tort claims brought by the tribe against a nonmember\\nemployee.\\n\\n The tort claims arose from conduct committed by the\\nnonmember on tribal lands during the scope of her\\nemployment. The panel held that a tribe’s regulatory power\\nover nonmembers on tribal land derives both from the tribe’s\\ninherent sovereign power to exclude nonmembers from\\ntribal land and from the tribe’s inherent sovereign power to\\nprotect self-government and control internal relations.\\n\\n The panel held that the tribe had authority to regulate the\\nnonmember employee’s conduct at issue pursuant to its\\nexclusionary power. Alternatively, the tribe had regulatory\\nauthority under both Montana exceptions, which allow a\\ntribe (1) to regulate the activities of nonmembers who enter\\nconsensual relationships with the tribe or its members and\\n(2) to exercise civil authority over the conduct of\\nnonmembers on fee lands within its reservation when that\\nconduct threatens or directly affects the political integrity,\\nthe economic security, or the health or welfare of the tribe.\\nGiven the existence of regulatory authority, the sovereign\\ninterests at stake, and the congressional interest in promoting\\n\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 3\\n\\nself-government, the tribal court had jurisdiction over the\\ntribe’s tort claims.\\n\\n\\n COUNSEL\\n\\nPatrick L. Deedon (argued), Maire & Deedon, Redding,\\nCalifornia, for Plaintiff-Appellant.\\n\\nJack Duran, Jr., Esq. (argued), Duran Law Office, Roseville,\\nCalifornia, for Defendants-Appellees.\\n\\n\\n OPINION\\n\\nPIERSOL, Senior District Judge:\\n\\n This case concerns the sources and scope of an Indian\\ntribe’s jurisdiction over tort claims brought by the tribe\\nagainst a nonmember employed by the tribe. The tort claims\\narose from conduct committed by the nonmember on tribal\\nlands during the scope of her employment. The question\\npresented is whether the tribal court has jurisdiction to\\nadjudicate tribal claims against its nonmember employee,\\nwhere the tribe’s personnel policies and procedures manual\\nregulated the nonmember’s conduct at issue and provided\\nthat the tribal council would address violations by the\\nnonmember during the course of her employment, and the\\ntribal court and tribal judicial code were established and\\nenacted after the nonmember left her employment with the\\ntribe.\\n\\n We previously held that a tribe’s inherent sovereign\\npower to exclude nonmembers from tribal land is an\\nindependent source of regulatory power over nonmember\\n\\f4 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nconduct on tribal land. See Water Wheel Camp Recreational\\nArea, Inc. v. LaRance, 642 F.3d 802, 814 (9th Cir. 2011) (per\\ncuriam) (stating that where the nonmember activity occurred\\non tribal land, and when there are no competing state\\ninterests at play, “the tribe’s status as landowner is enough\\nto support regulatory jurisdiction without considering\\nMontana [v. United States, 450 U.S. 544 (1981)]”). Today\\nwe also observe that a tribe’s regulatory power over\\nnonmembers on tribal land does not solely derive from an\\nIndian tribe’s exclusionary power, but also derives\\nseparately from its inherent sovereign power to protect self-\\ngovernment and control internal relations. See Montana,\\n450 U.S. at 564 (stating that Indian tribes retain their\\ninherent sovereign power to protect tribal self-government\\nand to control internal relations); see also Merrion v.\\nJicarilla Apache Tribe, 455 U.S. 130, 144–45 (1982)\\n(holding that the tribe’s authority to tax nonmember mining\\nand drilling on tribal land derived from its inherent power to\\ngovern and pay for the costs of self-government and stating\\nthat such regulations were also within the tribe’s inherent\\npower to condition the continued presence of nonmembers\\non tribal land).\\n\\n Accordingly, we now hold that under the circumstances\\npresented here, the tribe has authority to regulate the\\nnonmember employee’s conduct at issue pursuant to its\\ninherent power to exclude nonmembers from tribal lands.\\nWe also hold, in the alternative, that the tribe has regulatory\\nauthority over the nonmember employee’s conduct under\\nboth Montana exceptions. Given the existence of regulatory\\nauthority, the sovereign interests at stake, and the\\ncongressional interest in promoting tribal self-government,\\nwe conclude that the tribal court has jurisdiction over the\\ntribe’s claims in this case.\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 5\\n\\n FACTUAL AND PROCEDURAL BACKGROUND\\n\\nI. Factual Background\\n\\n A. The Cedarville Rancheria Tribe\\n\\n The Cedarville Rancheria of Northern Paiute Indians\\n(“the Tribe”) is a federally recognized Indian tribe that has\\napproximately twelve voting members and operates a\\n17-acre Rancheria in Cedarville, California (“the\\nRancheria”). The Rancheria is held in trust for the Tribe by\\nthe United States government. During the latter part of\\nevents at issue in this case, the Tribe’s administrative offices\\nwere relocated from the Rancheria to land held in fee 1 by the\\nTribe in Alturas, California.\\n\\n The Tribe’s governing body is the Community Council,\\nwhich is composed of all qualified voters of the Rancheria\\nwho are 18 years of age or older. Every three years, the\\nCommunity Council elects three of its members to serve on\\nthe Executive Committee—the Tribal Chairperson, Vice\\nChairperson, and Secretary. The Executive Committee\\nenforces the Community Council’s ordinances and other\\nenactments and represents the Tribe in negotiations with\\ntribal, federal, state, and local governments.\\n\\n B. Knighton’s Employment with the Tribe\\n\\n Duanna Knighton (“Knighton”) was employed by the\\nTribe from July 1996 until she resigned in March 2013.\\nKnighton is not a member of the Tribe and had never resided\\non or owned land within the Rancheria. At the time of her\\n\\n 1\\n Pending with the Bureau of Indian Affairs is a petition by the Tribe\\nto place the property on which the Tribe’s administrative offices are now\\nlocated in trust with the United States government.\\n\\f6 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nresignation, Knighton’s position was that of Tribal\\nAdministrator. As Tribal Administrator, she oversaw the\\nday-to-day management of the Rancheria, its personnel, and\\nmany aspects of its finances.\\n\\n During Knighton’s employment, the Tribe regulated its\\nemployees pursuant to the Cedarville Rancheria Personnel\\nPolicies and Procedures Manual (“the Personnel Manual”).\\nThe Personnel Manual regulated employee conduct\\nincluding, but not limited to: misfeasance and malfeasance\\nin the performance of duty, incompetency in the\\nperformance of job duties, theft, carelessness or negligence\\nwith the monies or property of the Rancheria, inducement of\\nan employee to act in violation of Rancheria regulations, and\\nviolation of personnel rules. Disciplinary actions for an\\nemployee’s breach of rules and standards of conduct in the\\ncourse of employment specified in the Personnel Manual\\nincluded a verbal warning, written reprimand, suspension\\nwithout pay, demotion, and involuntary termination.\\n\\n The Personnel Manual provided that where the Tribal\\nAdministrator was the subject of disciplinary action, the\\nCommunity Council directly oversaw the disciplinary\\nprocess.\\n\\n C. Knighton’s Employment with RISE\\n\\n From 2009 until at least 2016, in addition to her position\\nas Tribal Administrator, Knighton was also serving as an\\nemployee or officer of Resources for Student Education\\n(“RISE”), a California nonprofit, that provides education\\nservices and programs to Indian children. RISE is not a\\ntribally created or licensed business entity, and it receives\\nthe majority of its funding from state and federal grants and\\nprivate donations.\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 7\\n\\n D. The Tribe’s Purchase of RISE Property\\n\\n In mid-2009, Knighton, acting in her capacity as Tribal\\nAdministrator, negotiated the Tribe’s purchase from RISE of\\na building in Alturas, California, where the Tribe’s\\nadministrative offices are now located. During this time,\\nKnighton was also an employee or agent of RISE.\\n\\n Knighton initially recommended that the Tribe purchase\\nthe building for $350,000, allegedly representing that such a\\nprice was below market value even though she had not\\nreceived a professional appraisal of the property. The Tribe\\nlater discovered that the $350,000 purchase price\\nrecommended by Knighton was $200,000 above market\\nvalue. Knighton also represented to the Tribe that it could\\npay off its building loan within five years after the purchase\\nand that RISE would pay rent to the Tribe for its occupancy\\nuntil the note on the building was paid off.\\n\\n The Tribe asserts that at no time during the purchase\\nnegotiations did Knighton disclose she had a conflict of\\ninterest representing both RISE and the Tribe in the sale, that\\nRISE was close to insolvency, or that she had an agreement\\nwith RISE to split the proceeds of the building sale. The\\nparties settled on a purchase price of $300,000. Within\\ntwelve months of the sale, RISE moved its business\\noperations out of the building.\\n\\n E. Knighton’s Resignation\\n\\n Before Knighton resigned in March 2013 as Tribal\\nAdministrator, she allegedly cashed out $29,925 in vacation\\nand sick pay in violation of the Tribe’s policies and\\nprocedures. The Tribe issued a check in the amount of\\n$29,925, payable to RISE on Knighton’s behalf. The Tribal\\nVice Chairman approved Knighton’s request to cash out\\n\\f8 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nbased on her representation that her request had been\\napproved by the Tribal Chairperson, when in fact, the Tribal\\nChairperson had denied Knighton’s request.\\n\\n When Knighton resigned in March 2013, she took with\\nher all files, including files belonging to the Tribe, room\\nfurnishings, and a computer, representing to the Tribe that\\nthe property removed belonged to RISE.\\n\\n In late 2013, the Tribe wrote a letter to RISE demanding\\nthe return of the $29,925 and any and all tribal property,\\nincluding the computer. Both RISE and Knighton refused\\nthrough their counsel to return the funds or any of the\\nproperty.\\n\\n F. Creation of Constitution, Tribal Judicial Code,\\n and Tribal Court\\n\\n In February 2011, while Knighton was still employed by\\nthe Tribe, the Tribe’s voting membership adopted the\\nConstitution and Bylaws of the Cedarville Rancheria, which\\nwas approved by the Regional Director of the Bureau of\\nIndian Affairs. Article II of the Tribe’s Constitution\\nprovides that the “jurisdiction of [the Tribe] shall extend to\\nthe land now within the confines of the [] Rancheria and to\\nsuch other lands as may thereafter be added thereto.”\\n\\n In December 2013, nine months after Knighton’s\\nresignation, the Tribe enacted the Cedarville Rancheria\\nJudicial Code (“the Tribal Judicial Code”) and established\\nthe Cedarville Rancheria Tribal Court, which consists of a\\ntribal court (“the Tribal Court”) and a tribal court of appeals\\n(“the Tribal Court of Appeals”). All judges must be lawyers\\nexperienced in the practice of tribal and federal Indian law\\nand licensed to practice in the highest court of any state.\\nJudges cannot be the Tribal Administrator, Assistant Clerks,\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 9\\n\\nor members of the Executive Committee. The Tribal\\nJudicial Code provides that the Tribal Court and Tribal Court\\nof Appeals have jurisdiction over all civil causes of action\\nthat arise within the boundaries of the Rancheria. Pursuant\\nto the Tribal Judicial Code, the Tribal Court has the power\\nto issue orders and judgments and to award limited money\\ndamages.\\n\\n G. The Tribe’s Audit Findings\\n\\n In early 2014, after Knighton resigned, the Tribe\\nconducted a forensic accounting of the Tribe’s financial\\nposition. The Tribe alleges that the forensic accounting\\ncame about after the former Tribal Chairperson shot and\\nkilled four tribal members at an Executive Committee\\nmeeting on February 20, 2014. The Tribal Chairman was a\\nvocal critic of Knighton’s performance. He was among\\nthose killed by his sister. During this accounting, the Tribe\\nreviewed its annual audit reports dating back to 2005 and\\nfound that the reports detailed several material weakness\\nfindings by the auditor. The auditor’s findings noted major\\ndeficiencies in the accounting of the Tribe’s finances, which\\nKnighton oversaw, and noted that the Tribe had not adopted\\na policy regarding the investment of tribal funds. The Tribe\\nalso discovered that in 2008, an annual audit of the Tribe’s\\nfinances showed that $3.07 million of the Tribe’s money had\\nbeen invested by Knighton in high-risk investments, which\\nhad declined in value by more than $1.2 million by the end\\nof 2008. The Tribe also discovered that tribal funds\\nbelonging to the Tribe’s children had been co-mingled with\\nfunds invested on behalf of adults, resulting in improper\\ntaxation.\\n\\n The Tribe asserts that the annual audit reports, and the\\nmaterial weakness findings and investment losses detailed\\ntherein, had not been provided by Knighton to the Tribe and\\n\\f10 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nwere only discovered by the Tribe after Knighton’s\\nresignation.\\n\\nII. Procedural Background\\n\\n The Tribe filed a complaint in the Tribal Court against\\nKnighton, RISE, and Oppenheimer Funds, Inc. 2 The\\ncomplaint included claims for fraud and deceit, recovery of\\nunauthorized and excessive pension payments, breach of\\nfiduciary duty, aiding and abetting breach of fiduciary duty,\\nand unjust enrichment, common count-account stated, and\\ncommon count-money had and received. In support of its\\nclaims, the Tribe alleged that Knighton improperly\\nmanipulated the Tribe’s policies and procedures to provide\\nher salary and fringe benefits, including a pension in excess\\nof what would normally be paid to a Tribal Administrator for\\na like-sized tribe. The Tribe also alleged that Knighton\\ninvested its money in high-risk investments without the\\nappropriate authority, and attempted to enter financial\\nagreements without appropriate authorization or waivers of\\ntribal sovereign immunity.\\n\\n Knighton responded by filing a motion to dismiss,\\nclaiming, in relevant part, that the Tribal Court lacked\\nsubject matter jurisdiction under Montana v. United States.\\n\\n The Tribal Court denied Knighton’s motion to dismiss,\\nfinding that it had jurisdiction over the Tribe’s claims against\\nKnighton under both Montana exceptions because Knighton\\nentered into a consensual relationship with the Tribe, by\\nvirtue of her employment with the Tribe, and because\\nKnighton’s conduct threatened or had a direct effect on the\\n\\n 2\\n RISE and Oppenheimer Funds, Inc. are no longer parties in this\\nlawsuit.\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 11\\n\\npolitical integrity, economic security, and health and welfare\\nof the Tribe. The Tribal Court’s decision was affirmed by\\nthe Tribal Court of Appeals, but the case was remanded to\\nthe Tribal Court to determine whether RISE was an\\nindispensable party to the suit, following a finding that the\\nissue had not been raised in the Tribal Court.\\n\\n On remand, pursuant to a stipulation between the parties,\\nthe Tribal Court stayed the case to allow Knighton to contest\\nin federal district court the Tribal Court’s jurisdiction over\\nthe Tribe’s asserted claims. As a result of the stay, there is\\nno Tribal Court exhaustion issue in this case.\\n\\n Knighton filed a lawsuit in federal district court seeking,\\namong other things, a declaratory judgment that the Tribal\\nCourt lacks subject matter jurisdiction over the Tribe’s\\nclaims against Knighton under both Montana exceptions,\\nand a permanent injunction against further proceedings in\\nthe Tribal Court. The defendants moved to dismiss\\nKnighton’s complaint on the basis that the Tribal Court’s\\njurisdiction over the Tribe’s claims was proper under both\\nMontana exceptions.\\n\\n The district court ruled that the Tribal Court has subject\\nmatter jurisdiction over the Tribe’s claims, and granted\\ndefendants’ motion to dismiss. The district court declined to\\napply Montana in its jurisdictional analysis based on its\\nfinding that Knighton’s alleged conduct occurred either on\\ntribal land within the Rancheria’s borders or was closely\\nrelated to tribal land. The district court stated that under\\nWater Wheel, the Montana framework did not apply to\\njurisdictional issues involving nonmember conduct on tribal\\nland. The district court concluded that the Tribe had\\nauthority to regulate Knighton’s conduct because\\n“Knighton’s employment activities directly affected the\\nTribe’s inherent powers to protect the welfare of its members\\n\\f12 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nand preserve the integrity of its government” and because\\n“her conduct threatened the Tribe’s very economic\\nsurvival,” and held that the Tribal Court had jurisdiction to\\nadjudicate the Tribe’s claims.\\n\\n Knighton appealed.\\n\\n JURISDICTION AND STANDARD OF REVIEW\\n\\n We have jurisdiction under 28 U.S.C. § 1291. The\\nquestion of tribal court jurisdiction is a question of federal\\nlaw, which we review de novo, with factual findings\\nreviewed for clear error. Smith v. Salish Kootenai Coll.,\\n434 F.3d 1127, 1130 (9th Cir. 2006) (en banc).\\n\\n ANALYSIS\\n\\n “To exercise its inherent civil authority over a defendant,\\na tribal court must have both subject matter jurisdiction—\\nconsisting of regulatory and adjudicative jurisdiction—and\\npersonal jurisdiction.” Water Wheel, 642 F.3d at 809. At\\nissue in this case is whether the Tribal Court has subject\\nmatter jurisdiction over the Tribe’s claims against Knighton.\\n\\nI. Regulatory Jurisdiction\\n\\n A. Legal Precedent and This Case\\n\\n “The sovereignty that the Indian tribes retain is of a\\nunique and limited character. It exists only at the sufferance\\nof Congress and is subject to complete defeasance. But until\\nCongress acts, the tribes retain their existing sovereign\\npowers.” United States v. Wheeler, 435 U.S. 313, 323\\n(1978). “Indian tribes still possess those aspects of\\nsovereignty not withdrawn by treaty or statute, or by\\nimplication as a necessary result of their dependent status.”\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 13\\n\\nId. In National Farmers Union Insurance Cos. v. Crow\\nTribe of Indians, the Court recognized that “[t]he tribes also\\nretain some of the inherent powers of the self-governing\\npolitical communities that were formed long before\\nEuropeans first settled in North America.” 471 U.S. 845,\\n851 (1985) (citing White Mountain Apache Tribe v. Bracker,\\n448 U.S. 136, 142 (1980); Santa Clara Pueblo v. Martinez,\\n436 U.S. 49 55–56 (1978)). The Court went on to say that\\nunder 28 U.S.C. § 1331, a federal court may determine\\n“whether a tribal court has exceeded the lawful limits of its\\njurisdiction.” Id. at 853. Thus, the outer boundaries of tribal\\ncourt jurisdiction are a matter of federal common law.\\n\\n We have noted that the Court has long recognized that as\\npart of their residual sovereignty, tribes retain the inherent\\npower to exclude nonmembers from tribal lands. See Water\\nWheel, 642 F.3d at 808; see also New Mexico v. Mescalero\\nApache Tribe, 462 U.S. 324, 333 (1983) (“A tribe’s power\\nto exclude nonmembers entirely or to condition their\\npresence on [tribal land] is [] well established.”). “From a\\ntribe’s inherent sovereign powers flow lesser powers,\\nincluding the power to regulate [nonmembers] on tribal\\nland.” Water Wheel, 642 F.3d at 808–09 (citing South\\nDakota v. Bourland, 508 U.S. 679, 689 (1993)).\\n\\n The Court has made clear, however, “that once tribal\\nland is converted into fee simple [land], the tribe loses\\nplenary jurisdiction over it . . . . As a general rule, then ‘the\\ntribe has no authority itself, by way of tribal ordinance or\\nactions in the tribal courts, to regulate the use of fee land.’”\\nPlains Commerce Bank v. Long Family Land & Cattle Co.,\\n554 U.S. 316, 328–29 (2008) (quoting Brendale v.\\nConfederated Tribes & Bands of Yakima Indian Nation,\\n492 U.S. 408, 430 (1989) (plurality opinion)). In Montana\\nv. United States, the Court recognized two exceptions to this\\n\\f14 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\ngeneral rule. First, “[a] tribe may regulate, through taxation,\\nlicensing, or other means, the activities of nonmembers who\\nenter consensual relationships with the tribe or its members,\\nthrough commercial dealing, contracts, leases, or other\\narrangements.” Montana, 450 U.S. at 565. Second, a tribe\\nmay exercise civil authority over the conduct of\\nnonmembers on fee lands within its reservation when “that\\nconduct threatens or has some direct effect on the political\\nintegrity, the economic security, or the health or welfare of\\nthe tribe.” Id. at 566.\\n\\n “Since deciding Montana, the Supreme Court has\\napplied those exceptions almost exclusively to questions of\\njurisdiction arising on [non-tribal] land.” Water Wheel,\\n642 F.3d at 809. The exception is Nevada v. Hicks, 533 U.S.\\n353 (2001). Water Wheel, 642 F.3d at 809. In Hicks, the\\nCourt addressed a tribal court’s jurisdiction over claims\\nagainst state officers arising from the execution of a search\\nwarrant on tribal land for alleged violations of state poaching\\nlaws—specifically, the killing of bighorn sheep off the\\nreservation. 533 U.S. at 356–57. Both the state court and\\nthen the tribal court issued search warrants. Id. at 356. The\\nCourt stated that although ownership status of the land “may\\nsometimes be a dispositive factor” in determining a tribe’s\\nauthority to regulate nonmember activity on tribal land, the\\ntribe’s power to exclude nonmembers from tribal land was\\n“not alone enough to support” the tribe’s regulatory\\njurisdiction over the state officers’ activities when the state\\nhad a competing interest in executing a warrant for an off-\\nreservation crime. Id. at 360. The Court applied Montana\\nand concluded that “tribal authority to regulate state officers\\nin executing process related to the violation, off reservation,\\nof state laws is not essential to tribal self-government or\\ninternal relations” while “[t]he State’s interest in execution\\nof process is considerable.” Id. at 364.\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 15\\n\\n Although some jurisdictions have interpreted Hicks as\\neliminating the right-to-exclude framework as an\\nindependent source of regulatory power over nonmember\\nconduct on tribal land, we have declined to do so. In Water\\nWheel, we observed that Hicks “expressly limited its holding\\nto ‘the question of tribal-court jurisdiction over state officers\\nenforcing state law.’” Water Wheel, 642 F.3d at 813\\n(quoting Hicks, 533 U.S. at 358 n.2). Indeed, the Hicks\\nCourt specifically “le[ft] open the question of tribal-court\\njurisdiction over nonmember defendants in general.” Hicks,\\n533 U.S. at 358 n.2. In Water Wheel, we held that a “tribe’s\\nstatus as landowner is enough to support regulatory\\njurisdiction” except “when the specific concerns at issue [in\\nHicks] exist.” 642 F.3d at 813. “Doing otherwise would\\nimpermissibly broaden Montana’s scope beyond what any\\nprecedent requires and restrain tribal sovereign authority\\ndespite Congress’s clearly stated federal interest in\\npromoting tribal self-government.” Id. at 813.\\n\\n In Hicks, the defendants were state officers enforcing a\\nstate-court-issued search warrant, so there was a significant\\nstate interest at stake. By contrast, the present case involves\\na private, consensual employment relationship between\\nKnighton and the Tribe, which occurred primarily on tribal\\nland. There are no significant competing state interests, as\\nin Hicks. Accordingly, our Water Wheel precedent compels\\nthe conclusion that the Tribe possesses regulatory\\njurisdiction over its claims against Knighton.\\n\\n Since Hicks’s limited holding, the Court in Plains\\nCommerce Bank held that a tribal court did not have\\njurisdiction to adjudicate a discrimination claim concerning\\na non-Indian defendant’s sale of fee land. Plains Commerce\\nBank, 554 U.S. at 323, 340–41. The land in question was\\nsold as part of the 1908 Allotment Act and was owned by a\\n\\f16 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nnon-Indian party for at least 50 years. Id. at 331, 341. The\\nCourt found that the discrimination law that the plaintiffs\\nwere attempting to enforce operated as a restraint on\\nalienation and had the effect of regulating the substantive\\nterms on which the non-Indian bank was able to offer its fee\\nland for sale. Id. at 331. The Court stated that while\\n“Montana and its progeny permit tribal regulation of\\nnonmember conduct inside the reservation that implicates\\nthe tribe’s sovereign interests,” that case “does not permit\\nIndian tribes to regulate the sale of non-Indian fee land,” as\\nneither of the Montana exceptions applies. Id. at 332. By\\ncontrast, in the present case the nonmember defendant while\\non tribal land allegedly used her position as Tribal\\nAdministrator to violate the terms of her employment in a\\nwide variety of ways that were significantly detrimental to\\nthe management and financial security of the Tribe.\\n\\n B. Appellant’s Arguments\\n\\n Knighton argues that treating ownership status of the\\nland as a dispositive factor in upholding a tribe’s power to\\nregulate nonmember conduct on tribal land (unless, as in\\nHicks, there are significant state interests present) is contrary\\nto our prior rulings in McDonald v. Means, 309 F.3d 530\\n(9th Cir. 2002), and Smith v. Salish Kootenai College. We\\ndisagree. In McDonald, we specifically recognized that a\\ntribe’s jurisdiction over civil claims against nonmembers\\narising on tribal land is limited under Hicks only in cases\\nwhere significant state interests are present. See 309 F.3d\\nat 540. And in Window Rock Unified School District v.\\nReeves, 861 F.3d 894, 902 n.9 (9th Cir. 2017), we concluded\\nthat Smith did not limit a tribe’s jurisdiction over civil claims\\nagainst nonmembers bearing a direct connection to tribal\\nland. We concluded that Smith was distinguishable because\\nit involved a nonmember plaintiff, as opposed to a\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 17\\n\\nnonmember defendant, who had entered into a consensual\\nrelationship with the tribe by filing his action in tribal court.\\nId.\\n\\n Knighton’s argument that a tribe’s regulatory power over\\nnonmember conduct on tribal land is limited to conduct that\\ndirectly interferes with a tribe’s inherent powers to exclude\\nand manage its own lands is also unavailing. In Window\\nRock, we concluded that the tribal court’s jurisdiction over\\nemployment-related claims that did not involve access to\\ntribal land was plausible; accordingly, we held that the\\nnonmember defendants were required to exhaust their tribal\\ncourt remedies before proceeding in federal court. Id. at 896,\\n906. Moreover, limiting a tribe’s regulatory power over\\nnonmember conduct to that which directly interferes with a\\ntribe’s inherent powers to exclude and manage its own lands,\\nas Knighton suggests, would restrict tribal sovereignty\\nabsent explicit authorization from Congress—an approach\\nwe specifically rejected in Water Wheel. See 642 F.3d at 812\\n(stating that the tribe’s right to exclude nonmembers from\\ntribal land includes the power to regulate them “unless\\nCongress has said otherwise, or unless the Supreme Court\\nhas recognized that such power conflicts with federal\\ninterests promoting tribal self government”).\\n\\n Knighton also argues that under the facts of this case,\\nWater Wheel’s right-to-exclude framework is inapplicable\\nbecause some of her alleged misconduct occurred off tribal\\nland, after the tribal administrative offices were relocated to\\nfee land owned by the Tribe. Although the Tribe’s\\ncomplaint does not allege precisely where the conduct at\\nissue occurred, most of the claims alleged against Knighton\\ninvolve conduct that took place on tribal land, before the\\nTribe’s administrative offices were moved in mid-2009 to\\nthe RISE building in Alturas, California. Moreover, the facts\\n\\f18 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nof this case are unique in that any claims that may have\\narisen outside tribal land are based on alleged misconduct\\nand misrepresentations made by Knighton on tribal land.\\nSee Smith, 434 F.3d at 1135 (stating that jurisdictional\\ninquiry is not limited to deciding precisely when and where\\nthe claim arose, but whether it bears some direct connection\\nto tribal lands). For example, the $29,925 overpayment for\\nunused vacation and sick leave that the Tribe seeks to\\nrecover stems from misrepresentations that Knighton\\nallegedly made throughout the course of her employment,\\nbefore the Tribe’s administrative offices relocated. In\\naddition, the relocation of the Tribe’s administrative offices\\nfrom tribal land to the RISE building on tribal fee land was\\nallegedly due to misrepresentations by Knighton.\\n\\n Knighton further argues that even if the Tribe had the\\npower to regulate her conduct on tribal land during the\\ncourse of her employment under Water Wheel’s right-to-\\nexclude framework, the Tribe’s authority is limited to the\\nregulations that were in place during her employment—\\nwhich is to say, those provided for in the Personnel Manual.\\nKnighton contends that the Tribe is attempting to impose\\nnew regulations on her through tort law after she left her\\nemployment with the Tribe.\\n\\n A tribe’s power to exclude nonmembers from tribal lands\\npermits a tribe to condition a nonmember’s entry or\\ncontinued presence on tribal land, see Merrion, 455 U.S. at\\n144–45, but this inherent power does not permit the Tribe to\\nimpose new regulations upon Knighton’s conduct\\nretroactively when she is no longer present on tribal land.\\nHowever, we agree with the district court that Knighton’s\\nalleged conduct violated the Tribe’s regulations that were in\\nplace at the time of her employment. The Personnel Manual\\nregulated employee conduct including, but not limited to,\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 19\\n\\nmisfeasance or malfeasance in the performance of duty,\\nincompetency in the performance of job duties, theft,\\ncarelessness or negligence with the monies or property of the\\nRancheria, inducement of an employee to act in violation of\\nRancheria regulations, and violation of personnel rules—all\\nconduct that forms the basis of the Tribe’s claims against\\nKnighton.\\n\\n C. Sources of Authority\\n\\n In Water Wheel, we concluded that a tribe’s inherent\\nsovereign power to exclude nonmembers from tribal land\\nprovides an independent basis upon which a tribe may\\nregulate the conduct of nonmembers on tribal land. But, a\\ntribe’s power to exclude is not the only source of its\\nregulatory authority over nonmembers on tribal land. See\\nBrendale, 492 U.S. at 425 (“An Indian tribe’s [] power to\\nexclude nonmembers of the tribe from its lands is not the\\nonly source of Indian regulatory authority.”). “[T]ribes have\\ninherent sovereignty independent of that authority arising\\nfrom their power to exclude.” Id. (citing Merrion, 455 U.S.\\nat 141); see also Babbitt Ford, Inc. v. Navajo Indian Tribe,\\n710 F.2d 587, 592 (9th Cir. 1983) (“The power to exercise\\ntribal civil authority over [nonmembers] derives not only\\nfrom the tribe’s inherent powers necessary to self-\\ngovernment and territorial management, but also from the\\npower to exclude nonmembers from tribal land.” (citing\\nMerrion, 455 U.S. at 141–44)).\\n\\n In addition to the power to exclude, we have the\\nMontana Court’s acknowledgment that Indian tribes retain\\ntheir inherent sovereign power to protect tribal self-\\ngovernment and to control internal relations. 450 U.S.\\nat 564. “[I]n accordance with that right tribes ‘may regulate\\nnonmember behavior that implicates [these sovereign\\ninterests].’” Attorney’s Process & Investigation Servs., Inc.\\n\\f20 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nv. Sac & Fox Tribe of Miss. in Iowa, 609 F.3d 927, 936\\n(8th Cir. 2010) (quoting Plains Commerce Bank, 554 U.S. at\\n335).\\n\\n Subsequent to Montana, in Merrion, the Court affirmed\\nthat Indian tribes have inherent sovereign power to regulate\\nnonmember conduct on tribal land independent of that\\nauthority arising from their power to exclude. Merrion,\\n455 U.S. at 144. The Court in Merrion concluded that a\\ntribe’s power to tax nonmember mining and drilling on tribal\\nland derived from its inherent “power to govern and to pay\\nfor the costs of self-government,” and concluded that such\\nregulatory authority was also within the tribe’s inherent\\npower to condition the continued presence of nonmembers\\non tribal land. Id. at 144–45. These varied sources of tribal\\nregulatory power over nonmember conduct on the\\nreservation were affirmed by the Court in Plains Commerce\\nBank. 554 U.S. at 337 (“[T]he regulation must stem from\\nthe tribe’s inherent sovereign authority to set conditions on\\nentry, preserve tribal self-government, or control internal\\nrelations.”).\\n\\n While the district court believed that our caselaw\\nprohibited the application of the Montana framework to\\ntribal jurisdictional issues involving nonmember conduct on\\ntribal land, it also recognized that a tribe’s regulatory power\\nover nonmembers on tribal land does not solely derive from\\nits power to set conditions on entry or continued presence.\\nAccordingly, it concluded that the Tribe had regulatory\\njurisdiction over Knighton’s conduct because “Knighton’s\\nemployment activities directly affected the Tribe’s inherent\\npowers to protect the welfare of its members and preserve\\nthe integrity of its government,” and because “her conduct\\nthreatened the Tribe’s very economic survival.”\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 21\\n\\n We now clarify Water Wheel and our subsequent cases\\ninvolving tribal jurisdictional issues on tribal land do not\\nexclude Montana as a source of tribal regulatory authority\\nover nonmember conduct on tribal land. Rather, our caselaw\\nstates that an Indian tribe has power to regulate nonmember\\nconduct on tribal land incident to its sovereign power to\\nexclude nonmembers from tribal land, regardless of whether\\neither of the Montana exceptions is satisfied. See Water\\nWheel, 642 F.3d at 814 (“[T]he tribe’s status as landowner is\\nenough to support regulatory jurisdiction without\\nconsidering Montana.” (emphasis added)); Grand Canyon\\nSkywalk Dev., LLC v. ‘Sa’ Nyu Wa Inc., 715 F.3d 1196, 1204\\n(9th Cir. 2013) (“[A] tribe’s inherent authority over tribal\\nland may provide for regulatory authority over\\n[nonmembers] on that land without the need to consider\\nMontana.” (emphasis added)); Window Rock, 861 F.3d at\\n902 (“[I]n civil cases involving nonmember conduct on\\ntribal land, we have held that tribal courts have jurisdiction\\nunless a treaty or federal statute provides otherwise—\\nregardless of whether the Montana exceptions would be\\nsatisfied.” (emphasis added)). Certainly, as our caselaw has\\ndiscussed at length, without evidence of a contrary intent by\\nCongress, a tribe’s power to regulate nonmember conduct on\\ntribal land flows from its inherent power to exclude and is\\ncircumscribed only to the limited extent that the\\ncircumstances in Hicks—significant state interests—are\\npresent. See Water Wheel, 642 F.3d at 813; Grand Canyon,\\n715 F.3d at 1205; Window Rock, 861 F.3d at 902. However,\\nthe Court has made clear that a tribe also has sovereign\\nauthority to regulate nonmember conduct on tribal lands\\nindependent of its authority to exclude if that conduct\\nintrudes on a tribe’s inherent sovereign power to preserve\\nself-government or control internal relations. The Montana\\nexceptions are “rooted” in the tribes’ inherent power to\\nregulate nonmember behavior that implicates these\\n\\f22 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nsovereign interests. Attorney’s Process, 609 F.3d at 936\\n(citing Plains Commerce Bank, 554 U.S. at 335).\\n\\n Accordingly, although we conclude that the Tribe had\\nauthority to regulate Knighton’s conduct on tribal land\\npursuant to its sovereign exclusionary powers, a separate\\nquestion remains as to whether the Tribe also had regulatory\\nauthority over Knighton’s conduct pursuant to Montana.\\n\\n i. First Montana Exception\\n\\n Montana’s consensual relationship exception recognizes\\nthat tribes have jurisdiction to regulate consensual relations\\n“through taxation, licensing, or other means.” 450 U.S.\\nat 565. Courts have recognized that tort law, under which\\nthe Tribe’s claims against Knighton arise, constitutes a form\\nof regulation. See Attorney’s Process, 609 F.3d at 938\\n(stating that if a tribe retains the power under Montana to\\nregulate nonmember conduct, it does not make any\\ndifference whether it does so through precisely tailored\\nregulations or through tort claims). However, Montana’s\\nconsensual relationship exception requires that “the\\nregulation imposed by the Indian tribe have a nexus to the\\nconsensual relationship itself.” Atkinson Trading Co. v.\\nShirley, 532 U.S. 645, 656 (2001). “A nonmember’s\\nconsensual relationship in one area thus does not trigger\\ntribal civil authority in another.” Id.\\n\\n Examining the facts of this case, we conclude that the\\nTribe has regulatory authority over Knighton’s conduct in\\nthis case under Montana’s consensual relationship\\nexception. The conduct that the Tribe seeks to regulate\\nthrough tort law arises directly out of the consensual\\nemployment relationship between the Tribe and Knighton.\\nMoreover, given the circumstances, Knighton should have\\nreasonably anticipated that her conduct might “trigger” tribal\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 23\\n\\nauthority. Water Wheel, 642 F.3d at 818 (quoting Plains\\nCommerce Bank, 554 U.S. at 338). Knighton is no stranger\\nto the Tribe’s governance and laws. She had been an\\nemployee of the Tribe for approximately sixteen years and,\\nas Tribal Administrator, was responsible for the overall\\nsupervision and management of tribal operations and\\ncarrying out tribal projects consistent with the Tribal\\nConstitution. The Tribal Constitution, adopted\\napproximately two years before Knighton resigned as Tribal\\nAdministrator, specifically provided that the “jurisdiction of\\n[the Tribe] shall extend to land now within the confines of\\nthe [Rancheria] and to such other lands as may thereafter be\\nadded thereto.” We conclude that given these\\ncircumstances, Knighton should reasonably have anticipated\\nthat her conduct on tribal land would fall within the Tribe’s\\nregulatory jurisdiction.\\n\\n ii. Second Montana Exception\\n\\n In determining whether Knighton’s conduct threatens or\\nhas some direct effect on the political integrity, the economic\\nsecurity, or the health or welfare of the tribe—the second\\nMontana exception, 450 U.S. at 566—we find instructive the\\nEighth Circuit’s decision in Attorney’s Process. In that case,\\nAPI, a nonmember corporation, was hired by a tribal\\ngovernment leader who refused to step down from\\nleadership after he lost in a special tribal election. 609 F.3d\\nat 932. Under their contract, API agreed to perform services\\nrelating to “the investigation of a takeover by dissidents at\\nthe Tribe’s facility located on the Tribe’s reservation lands.”\\nId. As the newly elected tribal council occupied the casino\\nand tribal government offices, approximately thirty API\\nagents forced their way into both buildings, which were\\nlocated on tribal land. Id. The agents were armed with\\nbatons, at least one carried a firearm, and they seized\\n\\f24 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nconfidential information from both facilities related to the\\ntribe’s gaming operations and finances. Id. In addition to\\nthe wrongfully seized confidential information, the agents\\ncaused approximately $7,000 in property damage and\\ncommitted various intentional torts against tribal members.\\nId. The tribe filed suit in tribal court for trespass to tribal\\nland and chattels, misappropriation of trade secrets, and\\nother claims. Id.\\n\\n API argued that tort claims do not in the ordinary course\\nthreaten the political integrity, economic security, or the\\nhealth or welfare of the tribe and thus the tribal court had no\\njurisdiction over the tribe’s claims under Montana’s second\\nexception. Id. at 937. Relying on Plains Commerce Bank,\\nthe court in Attorney’s Process stated that courts “should not\\nsimply consider the abstract elements of the tribal claim at\\nissue, but must focus on the specific nonmember conduct\\nalleged, taking a functional view of the regulatory effect of\\nthe claim on the nonmember.” Id. at 938. The court\\nconcluded that API’s raid on the casino and government\\noffices, leading to the claims for trespass to land, trespass to\\nchattels, and misappropriation of tribal trade secrets,\\n“menace[d] the ‘political integrity, the economic security,\\n[and] the health [and] welfare’ of the Tribe to such a degree\\nthat it ‘imperil[ed] the subsistence’ of the tribal community”\\nand that the tribe therefore retained the inherent power under\\nthe second Montana exception to regulate that conduct. Id.\\nat 939 (alterations in original) (quoting Plains Commerce\\nBank, 554 U.S. at 341).\\n\\n While Knighton’s conduct constitutes a different type of\\nviolation, it was of long duration and had a great impact upon\\nthe Tribe, and so we conclude that the alleged harm to the\\nTribe caused by her conduct “‘imperil[ed] the subsistence’\\nof the tribal community.” Evans v. Shoshone-Bannock Land\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 25\\n\\nUse Policy Comm’n, 736 F.3d 1298, 1306 (9th Cir. 2013)\\n(quoting Plains Commerce Bank, 554 U.S. at 341). Among\\nthe tribe’s many claims are allegations that Knighton\\ninvested the Tribe’s money without appropriate authority,\\nconcealed investment documents and audit reports from the\\nTribe, and attempted to enter financial agreements without\\nthe appropriate authorization or waiver of tribal sovereign\\nimmunity. The Tribe also alleges that Knighton made\\nunreasonably risky investments that led to investment losses\\nin excess of $1.2 million, excess transaction fees, and state\\nand federal tax exposure, and that she breached her fiduciary\\nduty and deceived the Tribe, causing it to pay $300,000,\\n$150,000 above market value, for the RISE building\\npurchase. Finally, the Tribe alleges that when she resigned\\nher employment with the Tribe, Knighton took all files,\\nincluding files belonging to the Tribe, room furnishings, and\\na computer, representing to the Tribe that the property\\nremoved belonged to RISE. We conclude that this conduct\\nthreatened the Tribe’s very subsistence and that the Tribe\\ntherefore retains the inherent power under the second\\nMontana exception to regulate this conduct.\\n\\nII. Adjudicatory Jurisdiction\\n\\n Knighton also contends that the Tribe is seeking to\\nexercise greater adjudicative authority over her than it was\\ncapable of at the time of her employment. She argues that\\nthe adjudicatory authority of the Tribe is limited to the\\ndisciplinary procedures provided for in the Personnel\\nManual. At the time of her employment, the disciplinary\\nactions detailed in the Personnel Manual for an employee’s\\nbreach of rules and standards of conduct in the course of\\nemployment included a verbal warning, written reprimand,\\nsuspension without pay, demotion, and involuntary\\ntermination. The Personnel Manual provided that when the\\n\\f26 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nTribal Administrator was the subject of disciplinary action,\\nthe Community Council directly oversaw the disciplinary\\nand grievance procedures.\\n\\n We hold that the Tribe has the power to regulate\\nKnighton’s conduct incident to its sovereign powers to\\nexclude nonmembers from tribal land, and also, in the\\nalternative, under both Montana exceptions. “[W]here tribes\\npossess authority to regulate the activities of nonmembers,\\n‘[c]ivil jurisdiction over [disputes arising out of] such\\nactivities presumptively lies in the tribal courts.’” Strate v.\\nA-1 Contractors, 520 U.S. 438, 453 (1997) (second and third\\nalterations in original) (quoting Iowa Mut. Ins. Co. v.\\nLaPlante, 480 U.S. 9, 18 (1987)). However, a tribe’s\\nadjudicative authority over nonmembers may not exceed its\\nregulatory authority. Id.\\n\\n We conclude that under the facts of this case, the Tribe’s\\nadjudicatory authority does not exceed the regulatory\\nauthority it had over Knighton’s conduct during her\\nemployment under Water Wheel’s right-to-exclude\\nframework. As discussed above, the Personnel Manual\\nregulated the conduct that forms the basis of the Tribe’s\\nclaims against Knighton and conferred jurisdiction over her\\nconduct as Tribal Administrator on the Community Council.\\nThe fact that the Tribe now seeks to adjudicate these claims\\nin the Tribal Court does not undermine its jurisdiction over\\nthe Tribe’s claims.\\n\\n Likewise, examining the Tribe’s adjudicative authority\\nover Knighton’s conduct under Montana, we return to the\\nilluminating Eighth Circuit opinion in Attorney’s Process.\\nSimilar to this case, in Attorney’s Process, the tribal court\\nsystem was established after the tort claims against API\\narose. 609 F.3d at 933. API argued that the tribe lacked\\njurisdiction over its claims because there were no written\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 27\\n\\nregulations in place at the time which prohibited the tortious\\nconduct that API was alleged to have committed. Id. at 938.\\nThe court stated that “[i]f the Tribe retains the power under\\nMontana to regulate such conduct, we fail to see how it\\nmakes any difference whether it does so through precisely\\ntailored regulations or through tort claims such as those at\\nissue [in the case].” Id. The court concluded that because\\nAPI’s intervention onto tribal land threatened the “‘political\\nintegrity, the economic security, [and] the health [and]\\nwelfare’ of the Tribe,” the tribe had the authority to regulate\\nand adjudicate such conduct under Montana, as well as\\nincident to its sovereign right to exclude nonmembers from\\ntribal land. Id. at 940 (alterations in original) (quoting\\nMontana, 450 U.S. at 566).\\n\\n As the court in Attorney’s Process recognized, our task\\nis to outline the boundaries of the inherent sovereign power\\nretained by the Indian tribes. “Those boundaries are\\nestablished by federal law, a source of law external to the\\ntribes.” Id. at 938 (citing Nat’l Farmers Union, 471 U.S.\\nat 852). In contrast, “positive tribal law,” the court stated,\\n“is internal to the tribes.” Id. “It is a manifestation of tribal\\npower, and as such it does not contribute to the external\\nlimitations which concern us here. Once it is determined that\\ncertain conduct is within the scope of a tribe’s power as a\\nmatter of federal law, our inquiry is at an end.” Id.\\n\\n In the present case, the Tribe’s authority to regulate\\nKnighton’s conduct derived not only from its sovereign\\npower to exclude nonmembers from tribal lands, but also\\nfrom its inherent sovereign power to regulate consensual\\nrelations with nonmembers “through taxation, licensing, or\\nother means,” and to protect the “political integrity, the\\neconomic security, [and] the health [and] welfare” of the\\nTribe. Montana, 450 U.S. at 565–66.\\n\\f28 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\n Once the authority to regulate nonmember conduct\\nexists, whether from Water Wheel or from Montana, then the\\nobservation from the court in Attorney’s Process persuades\\nus that it makes no difference whether the Tribe adjudicates\\nKnighton’s conduct through the Personnel Manual or\\nthrough tort law.\\n\\n CONCLUSION\\n\\n There is no general rule as to the extent of a tribe’s\\nadjudicative jurisdiction over non-Indians on tribal land, but\\n“it is clear that the general rule announced in Strate, and\\nconfirmed in Hicks and Plains Commerce Bank, that\\nadjudicative jurisdiction is confined by the bounds of a\\ntribe’s regulatory jurisdiction” applies. Water Wheel,\\n642 F.3d at 814. Given the existence of regulatory authority,\\nthe sovereign interests at stake, and the congressional\\ninterest in promoting tribal self-government, we conclude\\nthat the Tribal Court has jurisdiction over the Tribe’s claims\\nin this case.\\n\\n AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4376404/\", \"author_raw\": \"Opinion by Judge Piersol\"}]}"},"judge1":{"kind":"string","value":"A WALLACE TASHIMA"},"judge2":{"kind":"string","value":"MILAN D SMITH JR"},"judge3":{"kind":"string","value":"LAWRENCE L PIERSOL"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4599151/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Duanna Knighton v. Cedarville Rancheria of Npi\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-13\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n DUANNA KNIGHTON, No. 17-15515\n Plaintiff-Appellant,\n D.C. No.\n v. 2:16-cv-02438-\n WHO\n CEDARVILLE RANCHERIA OF\n NORTHERN PAIUTE INDIANS;\n CEDARVILLE RANCHERIA TRIBAL OPINION\n COURT; PATRICIA R. LENZI, in her\n capacity as Chief Judge of the\n Cedarville Rancheria Tribal Court,\n Defendants-Appellees.\n\n Appeal from the United States District Court\n for the Eastern District of California\n William Horsley Orrick, District Judge, Presiding\n\n Argued and Submitted November 16, 2018\n San Francisco, California\n\n Filed March 13, 2019\n\n Before: A. Wallace Tashima and Milan D. Smith, Jr.,\n Circuit Judges, and Lawrence L. Piersol, * District Judge.\n\n Opinion by Judge Piersol\n\n *\n The Honorable Lawrence L. Piersol, United States District Judge\nfor the District of South Dakota, sitting by designation.\n\f2 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\n SUMMARY **\n\n\n Tribal Jurisdiction\n\n The panel affirmed the district court’s dismissal of an\naction challenging a tribal court’s subject matter jurisdiction\nover tort claims brought by the tribe against a nonmember\nemployee.\n\n The tort claims arose from conduct committed by the\nnonmember on tribal lands during the scope of her\nemployment. The panel held that a tribe’s regulatory power\nover nonmembers on tribal land derives both from the tribe’s\ninherent sovereign power to exclude nonmembers from\ntribal land and from the tribe’s inherent sovereign power to\nprotect self-government and control internal relations.\n\n The panel held that the tribe had authority to regulate the\nnonmember employee’s conduct at issue pursuant to its\nexclusionary power. Alternatively, the tribe had regulatory\nauthority under both Montana exceptions, which allow a\ntribe (1) to regulate the activities of nonmembers who enter\nconsensual relationships with the tribe or its members and\n(2) to exercise civil authority over the conduct of\nnonmembers on fee lands within its reservation when that\nconduct threatens or directly affects the political integrity,\nthe economic security, or the health or welfare of the tribe.\nGiven the existence of regulatory authority, the sovereign\ninterests at stake, and the congressional interest in promoting\n\n\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 3\n\nself-government, the tribal court had jurisdiction over the\ntribe’s tort claims.\n\n\n COUNSEL\n\nPatrick L. Deedon (argued), Maire & Deedon, Redding,\nCalifornia, for Plaintiff-Appellant.\n\nJack Duran, Jr., Esq. (argued), Duran Law Office, Roseville,\nCalifornia, for Defendants-Appellees.\n\n\n OPINION\n\nPIERSOL, Senior District Judge:\n\n This case concerns the sources and scope of an Indian\ntribe’s jurisdiction over tort claims brought by the tribe\nagainst a nonmember employed by the tribe. The tort claims\narose from conduct committed by the nonmember on tribal\nlands during the scope of her employment. The question\npresented is whether the tribal court has jurisdiction to\nadjudicate tribal claims against its nonmember employee,\nwhere the tribe’s personnel policies and procedures manual\nregulated the nonmember’s conduct at issue and provided\nthat the tribal council would address violations by the\nnonmember during the course of her employment, and the\ntribal court and tribal judicial code were established and\nenacted after the nonmember left her employment with the\ntribe.\n\n We previously held that a tribe’s inherent sovereign\npower to exclude nonmembers from tribal land is an\nindependent source of regulatory power over nonmember\n\f4 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\nconduct on tribal land. See Water Wheel Camp Recreational\nArea, Inc. v. LaRance, 642 F.3d 802, 814 (9th Cir. 2011) (per\ncuriam) (stating that where the nonmember activity occurred\non tribal land, and when there are no competing state\ninterests at play, “the tribe’s status as landowner is enough\nto support regulatory jurisdiction without considering\nMontana [v. United States, 450 U.S. 544 (1981)]”). Today\nwe also observe that a tribe’s regulatory power over\nnonmembers on tribal land does not solely derive from an\nIndian tribe’s exclusionary power, but also derives\nseparately from its inherent sovereign power to protect self-\ngovernment and control internal relations. See Montana,\n450 U.S. at 564 (stating that Indian tribes retain their\ninherent sovereign power to protect tribal self-government\nand to control internal relations); see also Merrion v.\nJicarilla Apache Tribe, 455 U.S. 130, 144–45 (1982)\n(holding that the tribe’s authority to tax nonmember mining\nand drilling on tribal land derived from its inherent power to\ngovern and pay for the costs of self-government and stating\nthat such regulations were also within the tribe’s inherent\npower to condition the continued presence of nonmembers\non tribal land).\n\n Accordingly, we now hold that under the circumstances\npresented here, the tribe has authority to regulate the\nnonmember employee’s conduct at issue pursuant to its\ninherent power to exclude nonmembers from tribal lands.\nWe also hold, in the alternative, that the tribe has regulatory\nauthority over the nonmember employee’s conduct under\nboth Montana exceptions. Given the existence of regulatory\nauthority, the sovereign interests at stake, and the\ncongressional interest in promoting tribal self-government,\nwe conclude that the tribal court has jurisdiction over the\ntribe’s claims in this case.\n\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 5\n\n FACTUAL AND PROCEDURAL BACKGROUND\n\nI. Factual Background\n\n A. The Cedarville Rancheria Tribe\n\n The Cedarville Rancheria of Northern Paiute Indians\n(“the Tribe”) is a federally recognized Indian tribe that has\napproximately twelve voting members and operates a\n17-acre Rancheria in Cedarville, California (“the\nRancheria”). The Rancheria is held in trust for the Tribe by\nthe United States government. During the latter part of\nevents at issue in this case, the Tribe’s administrative offices\nwere relocated from the Rancheria to land held in fee 1 by the\nTribe in Alturas, California.\n\n The Tribe’s governing body is the Community Council,\nwhich is composed of all qualified voters of the Rancheria\nwho are 18 years of age or older. Every three years, the\nCommunity Council elects three of its members to serve on\nthe Executive Committee—the Tribal Chairperson, Vice\nChairperson, and Secretary. The Executive Committee\nenforces the Community Council’s ordinances and other\nenactments and represents the Tribe in negotiations with\ntribal, federal, state, and local governments.\n\n B. Knighton’s Employment with the Tribe\n\n Duanna Knighton (“Knighton”) was employed by the\nTribe from July 1996 until she resigned in March 2013.\nKnighton is not a member of the Tribe and had never resided\non or owned land within the Rancheria. At the time of her\n\n 1\n Pending with the Bureau of Indian Affairs is a petition by the Tribe\nto place the property on which the Tribe’s administrative offices are now\nlocated in trust with the United States government.\n\f6 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\nresignation, Knighton’s position was that of Tribal\nAdministrator. As Tribal Administrator, she oversaw the\nday-to-day management of the Rancheria, its personnel, and\nmany aspects of its finances.\n\n During Knighton’s employment, the Tribe regulated its\nemployees pursuant to the Cedarville Rancheria Personnel\nPolicies and Procedures Manual (“the Personnel Manual”).\nThe Personnel Manual regulated employee conduct\nincluding, but not limited to: misfeasance and malfeasance\nin the performance of duty, incompetency in the\nperformance of job duties, theft, carelessness or negligence\nwith the monies or property of the Rancheria, inducement of\nan employee to act in violation of Rancheria regulations, and\nviolation of personnel rules. Disciplinary actions for an\nemployee’s breach of rules and standards of conduct in the\ncourse of employment specified in the Personnel Manual\nincluded a verbal warning, written reprimand, suspension\nwithout pay, demotion, and involuntary termination.\n\n The Personnel Manual provided that where the Tribal\nAdministrator was the subject of disciplinary action, the\nCommunity Council directly oversaw the disciplinary\nprocess.\n\n C. Knighton’s Employment with RISE\n\n From 2009 until at least 2016, in addition to her position\nas Tribal Administrator, Knighton was also serving as an\nemployee or officer of Resources for Student Education\n(“RISE”), a California nonprofit, that provides education\nservices and programs to Indian children. RISE is not a\ntribally created or licensed business entity, and it receives\nthe majority of its funding from state and federal grants and\nprivate donations.\n\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 7\n\n D. The Tribe’s Purchase of RISE Property\n\n In mid-2009, Knighton, acting in her capacity as Tribal\nAdministrator, negotiated the Tribe’s purchase from RISE of\na building in Alturas, California, where the Tribe’s\nadministrative offices are now located. During this time,\nKnighton was also an employee or agent of RISE.\n\n Knighton initially recommended that the Tribe purchase\nthe building for $350,000, allegedly representing that such a\nprice was below market value even though she had not\nreceived a professional appraisal of the property. The Tribe\nlater discovered that the $350,000 purchase price\nrecommended by Knighton was $200,000 above market\nvalue. Knighton also represented to the Tribe that it could\npay off its building loan within five years after the purchase\nand that RISE would pay rent to the Tribe for its occupancy\nuntil the note on the building was paid off.\n\n The Tribe asserts that at no time during the purchase\nnegotiations did Knighton disclose she had a conflict of\ninterest representing both RISE and the Tribe in the sale, that\nRISE was close to insolvency, or that she had an agreement\nwith RISE to split the proceeds of the building sale. The\nparties settled on a purchase price of $300,000. Within\ntwelve months of the sale, RISE moved its business\noperations out of the building.\n\n E. Knighton’s Resignation\n\n Before Knighton resigned in March 2013 as Tribal\nAdministrator, she allegedly cashed out $29,925 in vacation\nand sick pay in violation of the Tribe’s policies and\nprocedures. The Tribe issued a check in the amount of\n$29,925, payable to RISE on Knighton’s behalf. The Tribal\nVice Chairman approved Knighton’s request to cash out\n\f8 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\nbased on her representation that her request had been\napproved by the Tribal Chairperson, when in fact, the Tribal\nChairperson had denied Knighton’s request.\n\n When Knighton resigned in March 2013, she took with\nher all files, including files belonging to the Tribe, room\nfurnishings, and a computer, representing to the Tribe that\nthe property removed belonged to RISE.\n\n In late 2013, the Tribe wrote a letter to RISE demanding\nthe return of the $29,925 and any and all tribal property,\nincluding the computer. Both RISE and Knighton refused\nthrough their counsel to return the funds or any of the\nproperty.\n\n F. Creation of Constitution, Tribal Judicial Code,\n and Tribal Court\n\n In February 2011, while Knighton was still employed by\nthe Tribe, the Tribe’s voting membership adopted the\nConstitution and Bylaws of the Cedarville Rancheria, which\nwas approved by the Regional Director of the Bureau of\nIndian Affairs. Article II of the Tribe’s Constitution\nprovides that the “jurisdiction of [the Tribe] shall extend to\nthe land now within the confines of the [] Rancheria and to\nsuch other lands as may thereafter be added thereto.”\n\n In December 2013, nine months after Knighton’s\nresignation, the Tribe enacted the Cedarville Rancheria\nJudicial Code (“the Tribal Judicial Code”) and established\nthe Cedarville Rancheria Tribal Court, which consists of a\ntribal court (“the Tribal Court”) and a tribal court of appeals\n(“the Tribal Court of Appeals”). All judges must be lawyers\nexperienced in the practice of tribal and federal Indian law\nand licensed to practice in the highest court of any state.\nJudges cannot be the Tribal Administrator, Assistant Clerks,\n\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 9\n\nor members of the Executive Committee. The Tribal\nJudicial Code provides that the Tribal Court and Tribal Court\nof Appeals have jurisdiction over all civil causes of action\nthat arise within the boundaries of the Rancheria. Pursuant\nto the Tribal Judicial Code, the Tribal Court has the power\nto issue orders and judgments and to award limited money\ndamages.\n\n G. The Tribe’s Audit Findings\n\n In early 2014, after Knighton resigned, the Tribe\nconducted a forensic accounting of the Tribe’s financial\nposition. The Tribe alleges that the forensic accounting\ncame about after the former Tribal Chairperson shot and\nkilled four tribal members at an Executive Committee\nmeeting on February 20, 2014. The Tribal Chairman was a\nvocal critic of Knighton’s performance. He was among\nthose killed by his sister. During this accounting, the Tribe\nreviewed its annual audit reports dating back to 2005 and\nfound that the reports detailed several material weakness\nfindings by the auditor. The auditor’s findings noted major\ndeficiencies in the accounting of the Tribe’s finances, which\nKnighton oversaw, and noted that the Tribe had not adopted\na policy regarding the investment of tribal funds. The Tribe\nalso discovered that in 2008, an annual audit of the Tribe’s\nfinances showed that $3.07 million of the Tribe’s money had\nbeen invested by Knighton in high-risk investments, which\nhad declined in value by more than $1.2 million by the end\nof 2008. The Tribe also discovered that tribal funds\nbelonging to the Tribe’s children had been co-mingled with\nfunds invested on behalf of adults, resulting in improper\ntaxation.\n\n The Tribe asserts that the annual audit reports, and the\nmaterial weakness findings and investment losses detailed\ntherein, had not been provided by Knighton to the Tribe and\n\f10 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\nwere only discovered by the Tribe after Knighton’s\nresignation.\n\nII. Procedural Background\n\n The Tribe filed a complaint in the Tribal Court against\nKnighton, RISE, and Oppenheimer Funds, Inc. 2 The\ncomplaint included claims for fraud and deceit, recovery of\nunauthorized and excessive pension payments, breach of\nfiduciary duty, aiding and abetting breach of fiduciary duty,\nand unjust enrichment, common count-account stated, and\ncommon count-money had and received. In support of its\nclaims, the Tribe alleged that Knighton improperly\nmanipulated the Tribe’s policies and procedures to provide\nher salary and fringe benefits, including a pension in excess\nof what would normally be paid to a Tribal Administrator for\na like-sized tribe. The Tribe also alleged that Knighton\ninvested its money in high-risk investments without the\nappropriate authority, and attempted to enter financial\nagreements without appropriate authorization or waivers of\ntribal sovereign immunity.\n\n Knighton responded by filing a motion to dismiss,\nclaiming, in relevant part, that the Tribal Court lacked\nsubject matter jurisdiction under Montana v. United States.\n\n The Tribal Court denied Knighton’s motion to dismiss,\nfinding that it had jurisdiction over the Tribe’s claims against\nKnighton under both Montana exceptions because Knighton\nentered into a consensual relationship with the Tribe, by\nvirtue of her employment with the Tribe, and because\nKnighton’s conduct threatened or had a direct effect on the\n\n 2\n RISE and Oppenheimer Funds, Inc. are no longer parties in this\nlawsuit.\n\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 11\n\npolitical integrity, economic security, and health and welfare\nof the Tribe. The Tribal Court’s decision was affirmed by\nthe Tribal Court of Appeals, but the case was remanded to\nthe Tribal Court to determine whether RISE was an\nindispensable party to the suit, following a finding that the\nissue had not been raised in the Tribal Court.\n\n On remand, pursuant to a stipulation between the parties,\nthe Tribal Court stayed the case to allow Knighton to contest\nin federal district court the Tribal Court’s jurisdiction over\nthe Tribe’s asserted claims. As a result of the stay, there is\nno Tribal Court exhaustion issue in this case.\n\n Knighton filed a lawsuit in federal district court seeking,\namong other things, a declaratory judgment that the Tribal\nCourt lacks subject matter jurisdiction over the Tribe’s\nclaims against Knighton under both Montana exceptions,\nand a permanent injunction against further proceedings in\nthe Tribal Court. The defendants moved to dismiss\nKnighton’s complaint on the basis that the Tribal Court’s\njurisdiction over the Tribe’s claims was proper under both\nMontana exceptions.\n\n The district court ruled that the Tribal Court has subject\nmatter jurisdiction over the Tribe’s claims, and granted\ndefendants’ motion to dismiss. The district court declined to\napply Montana in its jurisdictional analysis based on its\nfinding that Knighton’s alleged conduct occurred either on\ntribal land within the Rancheria’s borders or was closely\nrelated to tribal land. The district court stated that under\nWater Wheel, the Montana framework did not apply to\njurisdictional issues involving nonmember conduct on tribal\nland. The district court concluded that the Tribe had\nauthority to regulate Knighton’s conduct because\n“Knighton’s employment activities directly affected the\nTribe’s inherent powers to protect the welfare of its members\n\f12 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\nand preserve the integrity of its government” and because\n“her conduct threatened the Tribe’s very economic\nsurvival,” and held that the Tribal Court had jurisdiction to\nadjudicate the Tribe’s claims.\n\n Knighton appealed.\n\n JURISDICTION AND STANDARD OF REVIEW\n\n We have jurisdiction under 28 U.S.C. § 1291. The\nquestion of tribal court jurisdiction is a question of federal\nlaw, which we review de novo, with factual findings\nreviewed for clear error. Smith v. Salish Kootenai Coll.,\n434 F.3d 1127, 1130 (9th Cir. 2006) (en banc).\n\n ANALYSIS\n\n “To exercise its inherent civil authority over a defendant,\na tribal court must have both subject matter jurisdiction—\nconsisting of regulatory and adjudicative jurisdiction—and\npersonal jurisdiction.” Water Wheel, 642 F.3d at 809. At\nissue in this case is whether the Tribal Court has subject\nmatter jurisdiction over the Tribe’s claims against Knighton.\n\nI. Regulatory Jurisdiction\n\n A. Legal Precedent and This Case\n\n “The sovereignty that the Indian tribes retain is of a\nunique and limited character. It exists only at the sufferance\nof Congress and is subject to complete defeasance. But until\nCongress acts, the tribes retain their existing sovereign\npowers.” United States v. Wheeler, 435 U.S. 313, 323\n(1978). “Indian tribes still possess those aspects of\nsovereignty not withdrawn by treaty or statute, or by\nimplication as a necessary result of their dependent status.”\n\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 13\n\nId. In National Farmers Union Insurance Cos. v. Crow\nTribe of Indians, the Court recognized that “[t]he tribes also\nretain some of the inherent powers of the self-governing\npolitical communities that were formed long before\nEuropeans first settled in North America.” 471 U.S. 845,\n851 (1985) (citing White Mountain Apache Tribe v. Bracker,\n448 U.S. 136, 142 (1980); Santa Clara Pueblo v. Martinez,\n436 U.S. 49 55–56 (1978)). The Court went on to say that\nunder 28 U.S.C. § 1331, a federal court may determine\n“whether a tribal court has exceeded the lawful limits of its\njurisdiction.” Id. at 853. Thus, the outer boundaries of tribal\ncourt jurisdiction are a matter of federal common law.\n\n We have noted that the Court has long recognized that as\npart of their residual sovereignty, tribes retain the inherent\npower to exclude nonmembers from tribal lands. See Water\nWheel, 642 F.3d at 808; see also New Mexico v. Mescalero\nApache Tribe, 462 U.S. 324, 333 (1983) (“A tribe’s power\nto exclude nonmembers entirely or to condition their\npresence on [tribal land] is [] well established.”). “From a\ntribe’s inherent sovereign powers flow lesser powers,\nincluding the power to regulate [nonmembers] on tribal\nland.” Water Wheel, 642 F.3d at 808–09 (citing South\nDakota v. Bourland, 508 U.S. 679, 689 (1993)).\n\n The Court has made clear, however, “that once tribal\nland is converted into fee simple [land], the tribe loses\nplenary jurisdiction over it . . . . As a general rule, then ‘the\ntribe has no authority itself, by way of tribal ordinance or\nactions in the tribal courts, to regulate the use of fee land.’”\nPlains Commerce Bank v. Long Family Land & Cattle Co.,\n554 U.S. 316, 328–29 (2008) (quoting Brendale v.\nConfederated Tribes & Bands of Yakima Indian Nation,\n492 U.S. 408, 430 (1989) (plurality opinion)). In Montana\nv. United States, the Court recognized two exceptions to this\n\f14 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\ngeneral rule. First, “[a] tribe may regulate, through taxation,\nlicensing, or other means, the activities of nonmembers who\nenter consensual relationships with the tribe or its members,\nthrough commercial dealing, contracts, leases, or other\narrangements.” Montana, 450 U.S. at 565. Second, a tribe\nmay exercise civil authority over the conduct of\nnonmembers on fee lands within its reservation when “that\nconduct threatens or has some direct effect on the political\nintegrity, the economic security, or the health or welfare of\nthe tribe.” Id. at 566.\n\n “Since deciding Montana, the Supreme Court has\napplied those exceptions almost exclusively to questions of\njurisdiction arising on [non-tribal] land.” Water Wheel,\n642 F.3d at 809. The exception is Nevada v. Hicks, 533 U.S.\n353 (2001). Water Wheel, 642 F.3d at 809. In Hicks, the\nCourt addressed a tribal court’s jurisdiction over claims\nagainst state officers arising from the execution of a search\nwarrant on tribal land for alleged violations of state poaching\nlaws—specifically, the killing of bighorn sheep off the\nreservation. 533 U.S. at 356–57. Both the state court and\nthen the tribal court issued search warrants. Id. at 356. The\nCourt stated that although ownership status of the land “may\nsometimes be a dispositive factor” in determining a tribe’s\nauthority to regulate nonmember activity on tribal land, the\ntribe’s power to exclude nonmembers from tribal land was\n“not alone enough to support” the tribe’s regulatory\njurisdiction over the state officers’ activities when the state\nhad a competing interest in executing a warrant for an off-\nreservation crime. Id. at 360. The Court applied Montana\nand concluded that “tribal authority to regulate state officers\nin executing process related to the violation, off reservation,\nof state laws is not essential to tribal self-government or\ninternal relations” while “[t]he State’s interest in execution\nof process is considerable.” Id. at 364.\n\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 15\n\n Although some jurisdictions have interpreted Hicks as\neliminating the right-to-exclude framework as an\nindependent source of regulatory power over nonmember\nconduct on tribal land, we have declined to do so. In Water\nWheel, we observed that Hicks “expressly limited its holding\nto ‘the question of tribal-court jurisdiction over state officers\nenforcing state law.’” Water Wheel, 642 F.3d at 813\n(quoting Hicks, 533 U.S. at 358 n.2). Indeed, the Hicks\nCourt specifically “le[ft] open the question of tribal-court\njurisdiction over nonmember defendants in general.” Hicks,\n533 U.S. at 358 n.2. In Water Wheel, we held that a “tribe’s\nstatus as landowner is enough to support regulatory\njurisdiction” except “when the specific concerns at issue [in\nHicks] exist.” 642 F.3d at 813. “Doing otherwise would\nimpermissibly broaden Montana’s scope beyond what any\nprecedent requires and restrain tribal sovereign authority\ndespite Congress’s clearly stated federal interest in\npromoting tribal self-government.” Id. at 813.\n\n In Hicks, the defendants were state officers enforcing a\nstate-court-issued search warrant, so there was a significant\nstate interest at stake. By contrast, the present case involves\na private, consensual employment relationship between\nKnighton and the Tribe, which occurred primarily on tribal\nland. There are no significant competing state interests, as\nin Hicks. Accordingly, our Water Wheel precedent compels\nthe conclusion that the Tribe possesses regulatory\njurisdiction over its claims against Knighton.\n\n Since Hicks’s limited holding, the Court in Plains\nCommerce Bank held that a tribal court did not have\njurisdiction to adjudicate a discrimination claim concerning\na non-Indian defendant’s sale of fee land. Plains Commerce\nBank, 554 U.S. at 323, 340–41. The land in question was\nsold as part of the 1908 Allotment Act and was owned by a\n\f16 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\nnon-Indian party for at least 50 years. Id. at 331, 341. The\nCourt found that the discrimination law that the plaintiffs\nwere attempting to enforce operated as a restraint on\nalienation and had the effect of regulating the substantive\nterms on which the non-Indian bank was able to offer its fee\nland for sale. Id. at 331. The Court stated that while\n“Montana and its progeny permit tribal regulation of\nnonmember conduct inside the reservation that implicates\nthe tribe’s sovereign interests,” that case “does not permit\nIndian tribes to regulate the sale of non-Indian fee land,” as\nneither of the Montana exceptions applies. Id. at 332. By\ncontrast, in the present case the nonmember defendant while\non tribal land allegedly used her position as Tribal\nAdministrator to violate the terms of her employment in a\nwide variety of ways that were significantly detrimental to\nthe management and financial security of the Tribe.\n\n B. Appellant’s Arguments\n\n Knighton argues that treating ownership status of the\nland as a dispositive factor in upholding a tribe’s power to\nregulate nonmember conduct on tribal land (unless, as in\nHicks, there are significant state interests present) is contrary\nto our prior rulings in McDonald v. Means, 309 F.3d 530\n(9th Cir. 2002), and Smith v. Salish Kootenai College. We\ndisagree. In McDonald, we specifically recognized that a\ntribe’s jurisdiction over civil claims against nonmembers\narising on tribal land is limited under Hicks only in cases\nwhere significant state interests are present. See 309 F.3d\nat 540. And in Window Rock Unified School District v.\nReeves, 861 F.3d 894, 902 n.9 (9th Cir. 2017), we concluded\nthat Smith did not limit a tribe’s jurisdiction over civil claims\nagainst nonmembers bearing a direct connection to tribal\nland. We concluded that Smith was distinguishable because\nit involved a nonmember plaintiff, as opposed to a\n\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 17\n\nnonmember defendant, who had entered into a consensual\nrelationship with the tribe by filing his action in tribal court.\nId.\n\n Knighton’s argument that a tribe’s regulatory power over\nnonmember conduct on tribal land is limited to conduct that\ndirectly interferes with a tribe’s inherent powers to exclude\nand manage its own lands is also unavailing. In Window\nRock, we concluded that the tribal court’s jurisdiction over\nemployment-related claims that did not involve access to\ntribal land was plausible; accordingly, we held that the\nnonmember defendants were required to exhaust their tribal\ncourt remedies before proceeding in federal court. Id. at 896,\n906. Moreover, limiting a tribe’s regulatory power over\nnonmember conduct to that which directly interferes with a\ntribe’s inherent powers to exclude and manage its own lands,\nas Knighton suggests, would restrict tribal sovereignty\nabsent explicit authorization from Congress—an approach\nwe specifically rejected in Water Wheel. See 642 F.3d at 812\n(stating that the tribe’s right to exclude nonmembers from\ntribal land includes the power to regulate them “unless\nCongress has said otherwise, or unless the Supreme Court\nhas recognized that such power conflicts with federal\ninterests promoting tribal self government”).\n\n Knighton also argues that under the facts of this case,\nWater Wheel’s right-to-exclude framework is inapplicable\nbecause some of her alleged misconduct occurred off tribal\nland, after the tribal administrative offices were relocated to\nfee land owned by the Tribe. Although the Tribe’s\ncomplaint does not allege precisely where the conduct at\nissue occurred, most of the claims alleged against Knighton\ninvolve conduct that took place on tribal land, before the\nTribe’s administrative offices were moved in mid-2009 to\nthe RISE building in Alturas, California. Moreover, the facts\n\f18 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\nof this case are unique in that any claims that may have\narisen outside tribal land are based on alleged misconduct\nand misrepresentations made by Knighton on tribal land.\nSee Smith, 434 F.3d at 1135 (stating that jurisdictional\ninquiry is not limited to deciding precisely when and where\nthe claim arose, but whether it bears some direct connection\nto tribal lands). For example, the $29,925 overpayment for\nunused vacation and sick leave that the Tribe seeks to\nrecover stems from misrepresentations that Knighton\nallegedly made throughout the course of her employment,\nbefore the Tribe’s administrative offices relocated. In\naddition, the relocation of the Tribe’s administrative offices\nfrom tribal land to the RISE building on tribal fee land was\nallegedly due to misrepresentations by Knighton.\n\n Knighton further argues that even if the Tribe had the\npower to regulate her conduct on tribal land during the\ncourse of her employment under Water Wheel’s right-to-\nexclude framework, the Tribe’s authority is limited to the\nregulations that were in place during her employment—\nwhich is to say, those provided for in the Personnel Manual.\nKnighton contends that the Tribe is attempting to impose\nnew regulations on her through tort law after she left her\nemployment with the Tribe.\n\n A tribe’s power to exclude nonmembers from tribal lands\npermits a tribe to condition a nonmember’s entry or\ncontinued presence on tribal land, see Merrion, 455 U.S. at\n144–45, but this inherent power does not permit the Tribe to\nimpose new regulations upon Knighton’s conduct\nretroactively when she is no longer present on tribal land.\nHowever, we agree with the district court that Knighton’s\nalleged conduct violated the Tribe’s regulations that were in\nplace at the time of her employment. The Personnel Manual\nregulated employee conduct including, but not limited to,\n\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 19\n\nmisfeasance or malfeasance in the performance of duty,\nincompetency in the performance of job duties, theft,\ncarelessness or negligence with the monies or property of the\nRancheria, inducement of an employee to act in violation of\nRancheria regulations, and violation of personnel rules—all\nconduct that forms the basis of the Tribe’s claims against\nKnighton.\n\n C. Sources of Authority\n\n In Water Wheel, we concluded that a tribe’s inherent\nsovereign power to exclude nonmembers from tribal land\nprovides an independent basis upon which a tribe may\nregulate the conduct of nonmembers on tribal land. But, a\ntribe’s power to exclude is not the only source of its\nregulatory authority over nonmembers on tribal land. See\nBrendale, 492 U.S. at 425 (“An Indian tribe’s [] power to\nexclude nonmembers of the tribe from its lands is not the\nonly source of Indian regulatory authority.”). “[T]ribes have\ninherent sovereignty independent of that authority arising\nfrom their power to exclude.” Id. (citing Merrion, 455 U.S.\nat 141); see also Babbitt Ford, Inc. v. Navajo Indian Tribe,\n710 F.2d 587, 592 (9th Cir. 1983) (“The power to exercise\ntribal civil authority over [nonmembers] derives not only\nfrom the tribe’s inherent powers necessary to self-\ngovernment and territorial management, but also from the\npower to exclude nonmembers from tribal land.” (citing\nMerrion, 455 U.S. at 141–44)).\n\n In addition to the power to exclude, we have the\nMontana Court’s acknowledgment that Indian tribes retain\ntheir inherent sovereign power to protect tribal self-\ngovernment and to control internal relations. 450 U.S.\nat 564. “[I]n accordance with that right tribes ‘may regulate\nnonmember behavior that implicates [these sovereign\ninterests].’” Attorney’s Process & Investigation Servs., Inc.\n\f20 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\nv. Sac & Fox Tribe of Miss. in Iowa, 609 F.3d 927, 936\n(8th Cir. 2010) (quoting Plains Commerce Bank, 554 U.S. at\n335).\n\n Subsequent to Montana, in Merrion, the Court affirmed\nthat Indian tribes have inherent sovereign power to regulate\nnonmember conduct on tribal land independent of that\nauthority arising from their power to exclude. Merrion,\n455 U.S. at 144. The Court in Merrion concluded that a\ntribe’s power to tax nonmember mining and drilling on tribal\nland derived from its inherent “power to govern and to pay\nfor the costs of self-government,” and concluded that such\nregulatory authority was also within the tribe’s inherent\npower to condition the continued presence of nonmembers\non tribal land. Id. at 144–45. These varied sources of tribal\nregulatory power over nonmember conduct on the\nreservation were affirmed by the Court in Plains Commerce\nBank. 554 U.S. at 337 (“[T]he regulation must stem from\nthe tribe’s inherent sovereign authority to set conditions on\nentry, preserve tribal self-government, or control internal\nrelations.”).\n\n While the district court believed that our caselaw\nprohibited the application of the Montana framework to\ntribal jurisdictional issues involving nonmember conduct on\ntribal land, it also recognized that a tribe’s regulatory power\nover nonmembers on tribal land does not solely derive from\nits power to set conditions on entry or continued presence.\nAccordingly, it concluded that the Tribe had regulatory\njurisdiction over Knighton’s conduct because “Knighton’s\nemployment activities directly affected the Tribe’s inherent\npowers to protect the welfare of its members and preserve\nthe integrity of its government,” and because “her conduct\nthreatened the Tribe’s very economic survival.”\n\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 21\n\n We now clarify Water Wheel and our subsequent cases\ninvolving tribal jurisdictional issues on tribal land do not\nexclude Montana as a source of tribal regulatory authority\nover nonmember conduct on tribal land. Rather, our caselaw\nstates that an Indian tribe has power to regulate nonmember\nconduct on tribal land incident to its sovereign power to\nexclude nonmembers from tribal land, regardless of whether\neither of the Montana exceptions is satisfied. See Water\nWheel, 642 F.3d at 814 (“[T]he tribe’s status as landowner is\nenough to support regulatory jurisdiction without\nconsidering Montana.” (emphasis added)); Grand Canyon\nSkywalk Dev., LLC v. ‘Sa’ Nyu Wa Inc., 715 F.3d 1196, 1204\n(9th Cir. 2013) (“[A] tribe’s inherent authority over tribal\nland may provide for regulatory authority over\n[nonmembers] on that land without the need to consider\nMontana.” (emphasis added)); Window Rock, 861 F.3d at\n902 (“[I]n civil cases involving nonmember conduct on\ntribal land, we have held that tribal courts have jurisdiction\nunless a treaty or federal statute provides otherwise—\nregardless of whether the Montana exceptions would be\nsatisfied.” (emphasis added)). Certainly, as our caselaw has\ndiscussed at length, without evidence of a contrary intent by\nCongress, a tribe’s power to regulate nonmember conduct on\ntribal land flows from its inherent power to exclude and is\ncircumscribed only to the limited extent that the\ncircumstances in Hicks—significant state interests—are\npresent. See Water Wheel, 642 F.3d at 813; Grand Canyon,\n715 F.3d at 1205; Window Rock, 861 F.3d at 902. However,\nthe Court has made clear that a tribe also has sovereign\nauthority to regulate nonmember conduct on tribal lands\nindependent of its authority to exclude if that conduct\nintrudes on a tribe’s inherent sovereign power to preserve\nself-government or control internal relations. The Montana\nexceptions are “rooted” in the tribes’ inherent power to\nregulate nonmember behavior that implicates these\n\f22 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\nsovereign interests. Attorney’s Process, 609 F.3d at 936\n(citing Plains Commerce Bank, 554 U.S. at 335).\n\n Accordingly, although we conclude that the Tribe had\nauthority to regulate Knighton’s conduct on tribal land\npursuant to its sovereign exclusionary powers, a separate\nquestion remains as to whether the Tribe also had regulatory\nauthority over Knighton’s conduct pursuant to Montana.\n\n i. First Montana Exception\n\n Montana’s consensual relationship exception recognizes\nthat tribes have jurisdiction to regulate consensual relations\n“through taxation, licensing, or other means.” 450 U.S.\nat 565. Courts have recognized that tort law, under which\nthe Tribe’s claims against Knighton arise, constitutes a form\nof regulation. See Attorney’s Process, 609 F.3d at 938\n(stating that if a tribe retains the power under Montana to\nregulate nonmember conduct, it does not make any\ndifference whether it does so through precisely tailored\nregulations or through tort claims). However, Montana’s\nconsensual relationship exception requires that “the\nregulation imposed by the Indian tribe have a nexus to the\nconsensual relationship itself.” Atkinson Trading Co. v.\nShirley, 532 U.S. 645, 656 (2001). “A nonmember’s\nconsensual relationship in one area thus does not trigger\ntribal civil authority in another.” Id.\n\n Examining the facts of this case, we conclude that the\nTribe has regulatory authority over Knighton’s conduct in\nthis case under Montana’s consensual relationship\nexception. The conduct that the Tribe seeks to regulate\nthrough tort law arises directly out of the consensual\nemployment relationship between the Tribe and Knighton.\nMoreover, given the circumstances, Knighton should have\nreasonably anticipated that her conduct might “trigger” tribal\n\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 23\n\nauthority. Water Wheel, 642 F.3d at 818 (quoting Plains\nCommerce Bank, 554 U.S. at 338). Knighton is no stranger\nto the Tribe’s governance and laws. She had been an\nemployee of the Tribe for approximately sixteen years and,\nas Tribal Administrator, was responsible for the overall\nsupervision and management of tribal operations and\ncarrying out tribal projects consistent with the Tribal\nConstitution. The Tribal Constitution, adopted\napproximately two years before Knighton resigned as Tribal\nAdministrator, specifically provided that the “jurisdiction of\n[the Tribe] shall extend to land now within the confines of\nthe [Rancheria] and to such other lands as may thereafter be\nadded thereto.” We conclude that given these\ncircumstances, Knighton should reasonably have anticipated\nthat her conduct on tribal land would fall within the Tribe’s\nregulatory jurisdiction.\n\n ii. Second Montana Exception\n\n In determining whether Knighton’s conduct threatens or\nhas some direct effect on the political integrity, the economic\nsecurity, or the health or welfare of the tribe—the second\nMontana exception, 450 U.S. at 566—we find instructive the\nEighth Circuit’s decision in Attorney’s Process. In that case,\nAPI, a nonmember corporation, was hired by a tribal\ngovernment leader who refused to step down from\nleadership after he lost in a special tribal election. 609 F.3d\nat 932. Under their contract, API agreed to perform services\nrelating to “the investigation of a takeover by dissidents at\nthe Tribe’s facility located on the Tribe’s reservation lands.”\nId. As the newly elected tribal council occupied the casino\nand tribal government offices, approximately thirty API\nagents forced their way into both buildings, which were\nlocated on tribal land. Id. The agents were armed with\nbatons, at least one carried a firearm, and they seized\n\f24 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\nconfidential information from both facilities related to the\ntribe’s gaming operations and finances. Id. In addition to\nthe wrongfully seized confidential information, the agents\ncaused approximately $7,000 in property damage and\ncommitted various intentional torts against tribal members.\nId. The tribe filed suit in tribal court for trespass to tribal\nland and chattels, misappropriation of trade secrets, and\nother claims. Id.\n\n API argued that tort claims do not in the ordinary course\nthreaten the political integrity, economic security, or the\nhealth or welfare of the tribe and thus the tribal court had no\njurisdiction over the tribe’s claims under Montana’s second\nexception. Id. at 937. Relying on Plains Commerce Bank,\nthe court in Attorney’s Process stated that courts “should not\nsimply consider the abstract elements of the tribal claim at\nissue, but must focus on the specific nonmember conduct\nalleged, taking a functional view of the regulatory effect of\nthe claim on the nonmember.” Id. at 938. The court\nconcluded that API’s raid on the casino and government\noffices, leading to the claims for trespass to land, trespass to\nchattels, and misappropriation of tribal trade secrets,\n“menace[d] the ‘political integrity, the economic security,\n[and] the health [and] welfare’ of the Tribe to such a degree\nthat it ‘imperil[ed] the subsistence’ of the tribal community”\nand that the tribe therefore retained the inherent power under\nthe second Montana exception to regulate that conduct. Id.\nat 939 (alterations in original) (quoting Plains Commerce\nBank, 554 U.S. at 341).\n\n While Knighton’s conduct constitutes a different type of\nviolation, it was of long duration and had a great impact upon\nthe Tribe, and so we conclude that the alleged harm to the\nTribe caused by her conduct “‘imperil[ed] the subsistence’\nof the tribal community.” Evans v. Shoshone-Bannock Land\n\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 25\n\nUse Policy Comm’n, 736 F.3d 1298, 1306 (9th Cir. 2013)\n(quoting Plains Commerce Bank, 554 U.S. at 341). Among\nthe tribe’s many claims are allegations that Knighton\ninvested the Tribe’s money without appropriate authority,\nconcealed investment documents and audit reports from the\nTribe, and attempted to enter financial agreements without\nthe appropriate authorization or waiver of tribal sovereign\nimmunity. The Tribe also alleges that Knighton made\nunreasonably risky investments that led to investment losses\nin excess of $1.2 million, excess transaction fees, and state\nand federal tax exposure, and that she breached her fiduciary\nduty and deceived the Tribe, causing it to pay $300,000,\n$150,000 above market value, for the RISE building\npurchase. Finally, the Tribe alleges that when she resigned\nher employment with the Tribe, Knighton took all files,\nincluding files belonging to the Tribe, room furnishings, and\na computer, representing to the Tribe that the property\nremoved belonged to RISE. We conclude that this conduct\nthreatened the Tribe’s very subsistence and that the Tribe\ntherefore retains the inherent power under the second\nMontana exception to regulate this conduct.\n\nII. Adjudicatory Jurisdiction\n\n Knighton also contends that the Tribe is seeking to\nexercise greater adjudicative authority over her than it was\ncapable of at the time of her employment. She argues that\nthe adjudicatory authority of the Tribe is limited to the\ndisciplinary procedures provided for in the Personnel\nManual. At the time of her employment, the disciplinary\nactions detailed in the Personnel Manual for an employee’s\nbreach of rules and standards of conduct in the course of\nemployment included a verbal warning, written reprimand,\nsuspension without pay, demotion, and involuntary\ntermination. The Personnel Manual provided that when the\n\f26 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\nTribal Administrator was the subject of disciplinary action,\nthe Community Council directly oversaw the disciplinary\nand grievance procedures.\n\n We hold that the Tribe has the power to regulate\nKnighton’s conduct incident to its sovereign powers to\nexclude nonmembers from tribal land, and also, in the\nalternative, under both Montana exceptions. “[W]here tribes\npossess authority to regulate the activities of nonmembers,\n‘[c]ivil jurisdiction over [disputes arising out of] such\nactivities presumptively lies in the tribal courts.’” Strate v.\nA-1 Contractors, 520 U.S. 438, 453 (1997) (second and third\nalterations in original) (quoting Iowa Mut. Ins. Co. v.\nLaPlante, 480 U.S. 9, 18 (1987)). However, a tribe’s\nadjudicative authority over nonmembers may not exceed its\nregulatory authority. Id.\n\n We conclude that under the facts of this case, the Tribe’s\nadjudicatory authority does not exceed the regulatory\nauthority it had over Knighton’s conduct during her\nemployment under Water Wheel’s right-to-exclude\nframework. As discussed above, the Personnel Manual\nregulated the conduct that forms the basis of the Tribe’s\nclaims against Knighton and conferred jurisdiction over her\nconduct as Tribal Administrator on the Community Council.\nThe fact that the Tribe now seeks to adjudicate these claims\nin the Tribal Court does not undermine its jurisdiction over\nthe Tribe’s claims.\n\n Likewise, examining the Tribe’s adjudicative authority\nover Knighton’s conduct under Montana, we return to the\nilluminating Eighth Circuit opinion in Attorney’s Process.\nSimilar to this case, in Attorney’s Process, the tribal court\nsystem was established after the tort claims against API\narose. 609 F.3d at 933. API argued that the tribe lacked\njurisdiction over its claims because there were no written\n\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 27\n\nregulations in place at the time which prohibited the tortious\nconduct that API was alleged to have committed. Id. at 938.\nThe court stated that “[i]f the Tribe retains the power under\nMontana to regulate such conduct, we fail to see how it\nmakes any difference whether it does so through precisely\ntailored regulations or through tort claims such as those at\nissue [in the case].” Id. The court concluded that because\nAPI’s intervention onto tribal land threatened the “‘political\nintegrity, the economic security, [and] the health [and]\nwelfare’ of the Tribe,” the tribe had the authority to regulate\nand adjudicate such conduct under Montana, as well as\nincident to its sovereign right to exclude nonmembers from\ntribal land. Id. at 940 (alterations in original) (quoting\nMontana, 450 U.S. at 566).\n\n As the court in Attorney’s Process recognized, our task\nis to outline the boundaries of the inherent sovereign power\nretained by the Indian tribes. “Those boundaries are\nestablished by federal law, a source of law external to the\ntribes.” Id. at 938 (citing Nat’l Farmers Union, 471 U.S.\nat 852). In contrast, “positive tribal law,” the court stated,\n“is internal to the tribes.” Id. “It is a manifestation of tribal\npower, and as such it does not contribute to the external\nlimitations which concern us here. Once it is determined that\ncertain conduct is within the scope of a tribe’s power as a\nmatter of federal law, our inquiry is at an end.” Id.\n\n In the present case, the Tribe’s authority to regulate\nKnighton’s conduct derived not only from its sovereign\npower to exclude nonmembers from tribal lands, but also\nfrom its inherent sovereign power to regulate consensual\nrelations with nonmembers “through taxation, licensing, or\nother means,” and to protect the “political integrity, the\neconomic security, [and] the health [and] welfare” of the\nTribe. Montana, 450 U.S. at 565–66.\n\f28 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\n\n Once the authority to regulate nonmember conduct\nexists, whether from Water Wheel or from Montana, then the\nobservation from the court in Attorney’s Process persuades\nus that it makes no difference whether the Tribe adjudicates\nKnighton’s conduct through the Personnel Manual or\nthrough tort law.\n\n CONCLUSION\n\n There is no general rule as to the extent of a tribe’s\nadjudicative jurisdiction over non-Indians on tribal land, but\n“it is clear that the general rule announced in Strate, and\nconfirmed in Hicks and Plains Commerce Bank, that\nadjudicative jurisdiction is confined by the bounds of a\ntribe’s regulatory jurisdiction” applies. Water Wheel,\n642 F.3d at 814. Given the existence of regulatory authority,\nthe sovereign interests at stake, and the congressional\ninterest in promoting tribal self-government, we conclude\nthat the Tribal Court has jurisdiction over the Tribe’s claims\nin this case.\n\n AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Duanna Knighton v. Cedarville Rancheria of Npi\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-13\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n DUANNA KNIGHTON, No. 17-15515\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:16-cv-02438-\\n WHO\\n CEDARVILLE RANCHERIA OF\\n NORTHERN PAIUTE INDIANS;\\n CEDARVILLE RANCHERIA TRIBAL OPINION\\n COURT; PATRICIA R. LENZI, in her\\n capacity as Chief Judge of the\\n Cedarville Rancheria Tribal Court,\\n Defendants-Appellees.\\n\\n Appeal from the United States District Court\\n for the Eastern District of California\\n William Horsley Orrick, District Judge, Presiding\\n\\n Argued and Submitted November 16, 2018\\n San Francisco, California\\n\\n Filed March 13, 2019\\n\\n Before: A. Wallace Tashima and Milan D. Smith, Jr.,\\n Circuit Judges, and Lawrence L. Piersol, * District Judge.\\n\\n Opinion by Judge Piersol\\n\\n *\\n The Honorable Lawrence L. Piersol, United States District Judge\\nfor the District of South Dakota, sitting by designation.\\n\\f2 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\n SUMMARY **\\n\\n\\n Tribal Jurisdiction\\n\\n The panel affirmed the district court’s dismissal of an\\naction challenging a tribal court’s subject matter jurisdiction\\nover tort claims brought by the tribe against a nonmember\\nemployee.\\n\\n The tort claims arose from conduct committed by the\\nnonmember on tribal lands during the scope of her\\nemployment. The panel held that a tribe’s regulatory power\\nover nonmembers on tribal land derives both from the tribe’s\\ninherent sovereign power to exclude nonmembers from\\ntribal land and from the tribe’s inherent sovereign power to\\nprotect self-government and control internal relations.\\n\\n The panel held that the tribe had authority to regulate the\\nnonmember employee’s conduct at issue pursuant to its\\nexclusionary power. Alternatively, the tribe had regulatory\\nauthority under both Montana exceptions, which allow a\\ntribe (1) to regulate the activities of nonmembers who enter\\nconsensual relationships with the tribe or its members and\\n(2) to exercise civil authority over the conduct of\\nnonmembers on fee lands within its reservation when that\\nconduct threatens or directly affects the political integrity,\\nthe economic security, or the health or welfare of the tribe.\\nGiven the existence of regulatory authority, the sovereign\\ninterests at stake, and the congressional interest in promoting\\n\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 3\\n\\nself-government, the tribal court had jurisdiction over the\\ntribe’s tort claims.\\n\\n\\n COUNSEL\\n\\nPatrick L. Deedon (argued), Maire & Deedon, Redding,\\nCalifornia, for Plaintiff-Appellant.\\n\\nJack Duran, Jr., Esq. (argued), Duran Law Office, Roseville,\\nCalifornia, for Defendants-Appellees.\\n\\n\\n OPINION\\n\\nPIERSOL, Senior District Judge:\\n\\n This case concerns the sources and scope of an Indian\\ntribe’s jurisdiction over tort claims brought by the tribe\\nagainst a nonmember employed by the tribe. The tort claims\\narose from conduct committed by the nonmember on tribal\\nlands during the scope of her employment. The question\\npresented is whether the tribal court has jurisdiction to\\nadjudicate tribal claims against its nonmember employee,\\nwhere the tribe’s personnel policies and procedures manual\\nregulated the nonmember’s conduct at issue and provided\\nthat the tribal council would address violations by the\\nnonmember during the course of her employment, and the\\ntribal court and tribal judicial code were established and\\nenacted after the nonmember left her employment with the\\ntribe.\\n\\n We previously held that a tribe’s inherent sovereign\\npower to exclude nonmembers from tribal land is an\\nindependent source of regulatory power over nonmember\\n\\f4 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nconduct on tribal land. See Water Wheel Camp Recreational\\nArea, Inc. v. LaRance, 642 F.3d 802, 814 (9th Cir. 2011) (per\\ncuriam) (stating that where the nonmember activity occurred\\non tribal land, and when there are no competing state\\ninterests at play, “the tribe’s status as landowner is enough\\nto support regulatory jurisdiction without considering\\nMontana [v. United States, 450 U.S. 544 (1981)]”). Today\\nwe also observe that a tribe’s regulatory power over\\nnonmembers on tribal land does not solely derive from an\\nIndian tribe’s exclusionary power, but also derives\\nseparately from its inherent sovereign power to protect self-\\ngovernment and control internal relations. See Montana,\\n450 U.S. at 564 (stating that Indian tribes retain their\\ninherent sovereign power to protect tribal self-government\\nand to control internal relations); see also Merrion v.\\nJicarilla Apache Tribe, 455 U.S. 130, 144–45 (1982)\\n(holding that the tribe’s authority to tax nonmember mining\\nand drilling on tribal land derived from its inherent power to\\ngovern and pay for the costs of self-government and stating\\nthat such regulations were also within the tribe’s inherent\\npower to condition the continued presence of nonmembers\\non tribal land).\\n\\n Accordingly, we now hold that under the circumstances\\npresented here, the tribe has authority to regulate the\\nnonmember employee’s conduct at issue pursuant to its\\ninherent power to exclude nonmembers from tribal lands.\\nWe also hold, in the alternative, that the tribe has regulatory\\nauthority over the nonmember employee’s conduct under\\nboth Montana exceptions. Given the existence of regulatory\\nauthority, the sovereign interests at stake, and the\\ncongressional interest in promoting tribal self-government,\\nwe conclude that the tribal court has jurisdiction over the\\ntribe’s claims in this case.\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 5\\n\\n FACTUAL AND PROCEDURAL BACKGROUND\\n\\nI. Factual Background\\n\\n A. The Cedarville Rancheria Tribe\\n\\n The Cedarville Rancheria of Northern Paiute Indians\\n(“the Tribe”) is a federally recognized Indian tribe that has\\napproximately twelve voting members and operates a\\n17-acre Rancheria in Cedarville, California (“the\\nRancheria”). The Rancheria is held in trust for the Tribe by\\nthe United States government. During the latter part of\\nevents at issue in this case, the Tribe’s administrative offices\\nwere relocated from the Rancheria to land held in fee 1 by the\\nTribe in Alturas, California.\\n\\n The Tribe’s governing body is the Community Council,\\nwhich is composed of all qualified voters of the Rancheria\\nwho are 18 years of age or older. Every three years, the\\nCommunity Council elects three of its members to serve on\\nthe Executive Committee—the Tribal Chairperson, Vice\\nChairperson, and Secretary. The Executive Committee\\nenforces the Community Council’s ordinances and other\\nenactments and represents the Tribe in negotiations with\\ntribal, federal, state, and local governments.\\n\\n B. Knighton’s Employment with the Tribe\\n\\n Duanna Knighton (“Knighton”) was employed by the\\nTribe from July 1996 until she resigned in March 2013.\\nKnighton is not a member of the Tribe and had never resided\\non or owned land within the Rancheria. At the time of her\\n\\n 1\\n Pending with the Bureau of Indian Affairs is a petition by the Tribe\\nto place the property on which the Tribe’s administrative offices are now\\nlocated in trust with the United States government.\\n\\f6 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nresignation, Knighton’s position was that of Tribal\\nAdministrator. As Tribal Administrator, she oversaw the\\nday-to-day management of the Rancheria, its personnel, and\\nmany aspects of its finances.\\n\\n During Knighton’s employment, the Tribe regulated its\\nemployees pursuant to the Cedarville Rancheria Personnel\\nPolicies and Procedures Manual (“the Personnel Manual”).\\nThe Personnel Manual regulated employee conduct\\nincluding, but not limited to: misfeasance and malfeasance\\nin the performance of duty, incompetency in the\\nperformance of job duties, theft, carelessness or negligence\\nwith the monies or property of the Rancheria, inducement of\\nan employee to act in violation of Rancheria regulations, and\\nviolation of personnel rules. Disciplinary actions for an\\nemployee’s breach of rules and standards of conduct in the\\ncourse of employment specified in the Personnel Manual\\nincluded a verbal warning, written reprimand, suspension\\nwithout pay, demotion, and involuntary termination.\\n\\n The Personnel Manual provided that where the Tribal\\nAdministrator was the subject of disciplinary action, the\\nCommunity Council directly oversaw the disciplinary\\nprocess.\\n\\n C. Knighton’s Employment with RISE\\n\\n From 2009 until at least 2016, in addition to her position\\nas Tribal Administrator, Knighton was also serving as an\\nemployee or officer of Resources for Student Education\\n(“RISE”), a California nonprofit, that provides education\\nservices and programs to Indian children. RISE is not a\\ntribally created or licensed business entity, and it receives\\nthe majority of its funding from state and federal grants and\\nprivate donations.\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 7\\n\\n D. The Tribe’s Purchase of RISE Property\\n\\n In mid-2009, Knighton, acting in her capacity as Tribal\\nAdministrator, negotiated the Tribe’s purchase from RISE of\\na building in Alturas, California, where the Tribe’s\\nadministrative offices are now located. During this time,\\nKnighton was also an employee or agent of RISE.\\n\\n Knighton initially recommended that the Tribe purchase\\nthe building for $350,000, allegedly representing that such a\\nprice was below market value even though she had not\\nreceived a professional appraisal of the property. The Tribe\\nlater discovered that the $350,000 purchase price\\nrecommended by Knighton was $200,000 above market\\nvalue. Knighton also represented to the Tribe that it could\\npay off its building loan within five years after the purchase\\nand that RISE would pay rent to the Tribe for its occupancy\\nuntil the note on the building was paid off.\\n\\n The Tribe asserts that at no time during the purchase\\nnegotiations did Knighton disclose she had a conflict of\\ninterest representing both RISE and the Tribe in the sale, that\\nRISE was close to insolvency, or that she had an agreement\\nwith RISE to split the proceeds of the building sale. The\\nparties settled on a purchase price of $300,000. Within\\ntwelve months of the sale, RISE moved its business\\noperations out of the building.\\n\\n E. Knighton’s Resignation\\n\\n Before Knighton resigned in March 2013 as Tribal\\nAdministrator, she allegedly cashed out $29,925 in vacation\\nand sick pay in violation of the Tribe’s policies and\\nprocedures. The Tribe issued a check in the amount of\\n$29,925, payable to RISE on Knighton’s behalf. The Tribal\\nVice Chairman approved Knighton’s request to cash out\\n\\f8 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nbased on her representation that her request had been\\napproved by the Tribal Chairperson, when in fact, the Tribal\\nChairperson had denied Knighton’s request.\\n\\n When Knighton resigned in March 2013, she took with\\nher all files, including files belonging to the Tribe, room\\nfurnishings, and a computer, representing to the Tribe that\\nthe property removed belonged to RISE.\\n\\n In late 2013, the Tribe wrote a letter to RISE demanding\\nthe return of the $29,925 and any and all tribal property,\\nincluding the computer. Both RISE and Knighton refused\\nthrough their counsel to return the funds or any of the\\nproperty.\\n\\n F. Creation of Constitution, Tribal Judicial Code,\\n and Tribal Court\\n\\n In February 2011, while Knighton was still employed by\\nthe Tribe, the Tribe’s voting membership adopted the\\nConstitution and Bylaws of the Cedarville Rancheria, which\\nwas approved by the Regional Director of the Bureau of\\nIndian Affairs. Article II of the Tribe’s Constitution\\nprovides that the “jurisdiction of [the Tribe] shall extend to\\nthe land now within the confines of the [] Rancheria and to\\nsuch other lands as may thereafter be added thereto.”\\n\\n In December 2013, nine months after Knighton’s\\nresignation, the Tribe enacted the Cedarville Rancheria\\nJudicial Code (“the Tribal Judicial Code”) and established\\nthe Cedarville Rancheria Tribal Court, which consists of a\\ntribal court (“the Tribal Court”) and a tribal court of appeals\\n(“the Tribal Court of Appeals”). All judges must be lawyers\\nexperienced in the practice of tribal and federal Indian law\\nand licensed to practice in the highest court of any state.\\nJudges cannot be the Tribal Administrator, Assistant Clerks,\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 9\\n\\nor members of the Executive Committee. The Tribal\\nJudicial Code provides that the Tribal Court and Tribal Court\\nof Appeals have jurisdiction over all civil causes of action\\nthat arise within the boundaries of the Rancheria. Pursuant\\nto the Tribal Judicial Code, the Tribal Court has the power\\nto issue orders and judgments and to award limited money\\ndamages.\\n\\n G. The Tribe’s Audit Findings\\n\\n In early 2014, after Knighton resigned, the Tribe\\nconducted a forensic accounting of the Tribe’s financial\\nposition. The Tribe alleges that the forensic accounting\\ncame about after the former Tribal Chairperson shot and\\nkilled four tribal members at an Executive Committee\\nmeeting on February 20, 2014. The Tribal Chairman was a\\nvocal critic of Knighton’s performance. He was among\\nthose killed by his sister. During this accounting, the Tribe\\nreviewed its annual audit reports dating back to 2005 and\\nfound that the reports detailed several material weakness\\nfindings by the auditor. The auditor’s findings noted major\\ndeficiencies in the accounting of the Tribe’s finances, which\\nKnighton oversaw, and noted that the Tribe had not adopted\\na policy regarding the investment of tribal funds. The Tribe\\nalso discovered that in 2008, an annual audit of the Tribe’s\\nfinances showed that $3.07 million of the Tribe’s money had\\nbeen invested by Knighton in high-risk investments, which\\nhad declined in value by more than $1.2 million by the end\\nof 2008. The Tribe also discovered that tribal funds\\nbelonging to the Tribe’s children had been co-mingled with\\nfunds invested on behalf of adults, resulting in improper\\ntaxation.\\n\\n The Tribe asserts that the annual audit reports, and the\\nmaterial weakness findings and investment losses detailed\\ntherein, had not been provided by Knighton to the Tribe and\\n\\f10 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nwere only discovered by the Tribe after Knighton’s\\nresignation.\\n\\nII. Procedural Background\\n\\n The Tribe filed a complaint in the Tribal Court against\\nKnighton, RISE, and Oppenheimer Funds, Inc. 2 The\\ncomplaint included claims for fraud and deceit, recovery of\\nunauthorized and excessive pension payments, breach of\\nfiduciary duty, aiding and abetting breach of fiduciary duty,\\nand unjust enrichment, common count-account stated, and\\ncommon count-money had and received. In support of its\\nclaims, the Tribe alleged that Knighton improperly\\nmanipulated the Tribe’s policies and procedures to provide\\nher salary and fringe benefits, including a pension in excess\\nof what would normally be paid to a Tribal Administrator for\\na like-sized tribe. The Tribe also alleged that Knighton\\ninvested its money in high-risk investments without the\\nappropriate authority, and attempted to enter financial\\nagreements without appropriate authorization or waivers of\\ntribal sovereign immunity.\\n\\n Knighton responded by filing a motion to dismiss,\\nclaiming, in relevant part, that the Tribal Court lacked\\nsubject matter jurisdiction under Montana v. United States.\\n\\n The Tribal Court denied Knighton’s motion to dismiss,\\nfinding that it had jurisdiction over the Tribe’s claims against\\nKnighton under both Montana exceptions because Knighton\\nentered into a consensual relationship with the Tribe, by\\nvirtue of her employment with the Tribe, and because\\nKnighton’s conduct threatened or had a direct effect on the\\n\\n 2\\n RISE and Oppenheimer Funds, Inc. are no longer parties in this\\nlawsuit.\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 11\\n\\npolitical integrity, economic security, and health and welfare\\nof the Tribe. The Tribal Court’s decision was affirmed by\\nthe Tribal Court of Appeals, but the case was remanded to\\nthe Tribal Court to determine whether RISE was an\\nindispensable party to the suit, following a finding that the\\nissue had not been raised in the Tribal Court.\\n\\n On remand, pursuant to a stipulation between the parties,\\nthe Tribal Court stayed the case to allow Knighton to contest\\nin federal district court the Tribal Court’s jurisdiction over\\nthe Tribe’s asserted claims. As a result of the stay, there is\\nno Tribal Court exhaustion issue in this case.\\n\\n Knighton filed a lawsuit in federal district court seeking,\\namong other things, a declaratory judgment that the Tribal\\nCourt lacks subject matter jurisdiction over the Tribe’s\\nclaims against Knighton under both Montana exceptions,\\nand a permanent injunction against further proceedings in\\nthe Tribal Court. The defendants moved to dismiss\\nKnighton’s complaint on the basis that the Tribal Court’s\\njurisdiction over the Tribe’s claims was proper under both\\nMontana exceptions.\\n\\n The district court ruled that the Tribal Court has subject\\nmatter jurisdiction over the Tribe’s claims, and granted\\ndefendants’ motion to dismiss. The district court declined to\\napply Montana in its jurisdictional analysis based on its\\nfinding that Knighton’s alleged conduct occurred either on\\ntribal land within the Rancheria’s borders or was closely\\nrelated to tribal land. The district court stated that under\\nWater Wheel, the Montana framework did not apply to\\njurisdictional issues involving nonmember conduct on tribal\\nland. The district court concluded that the Tribe had\\nauthority to regulate Knighton’s conduct because\\n“Knighton’s employment activities directly affected the\\nTribe’s inherent powers to protect the welfare of its members\\n\\f12 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nand preserve the integrity of its government” and because\\n“her conduct threatened the Tribe’s very economic\\nsurvival,” and held that the Tribal Court had jurisdiction to\\nadjudicate the Tribe’s claims.\\n\\n Knighton appealed.\\n\\n JURISDICTION AND STANDARD OF REVIEW\\n\\n We have jurisdiction under 28 U.S.C. § 1291. The\\nquestion of tribal court jurisdiction is a question of federal\\nlaw, which we review de novo, with factual findings\\nreviewed for clear error. Smith v. Salish Kootenai Coll.,\\n434 F.3d 1127, 1130 (9th Cir. 2006) (en banc).\\n\\n ANALYSIS\\n\\n “To exercise its inherent civil authority over a defendant,\\na tribal court must have both subject matter jurisdiction—\\nconsisting of regulatory and adjudicative jurisdiction—and\\npersonal jurisdiction.” Water Wheel, 642 F.3d at 809. At\\nissue in this case is whether the Tribal Court has subject\\nmatter jurisdiction over the Tribe’s claims against Knighton.\\n\\nI. Regulatory Jurisdiction\\n\\n A. Legal Precedent and This Case\\n\\n “The sovereignty that the Indian tribes retain is of a\\nunique and limited character. It exists only at the sufferance\\nof Congress and is subject to complete defeasance. But until\\nCongress acts, the tribes retain their existing sovereign\\npowers.” United States v. Wheeler, 435 U.S. 313, 323\\n(1978). “Indian tribes still possess those aspects of\\nsovereignty not withdrawn by treaty or statute, or by\\nimplication as a necessary result of their dependent status.”\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 13\\n\\nId. In National Farmers Union Insurance Cos. v. Crow\\nTribe of Indians, the Court recognized that “[t]he tribes also\\nretain some of the inherent powers of the self-governing\\npolitical communities that were formed long before\\nEuropeans first settled in North America.” 471 U.S. 845,\\n851 (1985) (citing White Mountain Apache Tribe v. Bracker,\\n448 U.S. 136, 142 (1980); Santa Clara Pueblo v. Martinez,\\n436 U.S. 49 55–56 (1978)). The Court went on to say that\\nunder 28 U.S.C. § 1331, a federal court may determine\\n“whether a tribal court has exceeded the lawful limits of its\\njurisdiction.” Id. at 853. Thus, the outer boundaries of tribal\\ncourt jurisdiction are a matter of federal common law.\\n\\n We have noted that the Court has long recognized that as\\npart of their residual sovereignty, tribes retain the inherent\\npower to exclude nonmembers from tribal lands. See Water\\nWheel, 642 F.3d at 808; see also New Mexico v. Mescalero\\nApache Tribe, 462 U.S. 324, 333 (1983) (“A tribe’s power\\nto exclude nonmembers entirely or to condition their\\npresence on [tribal land] is [] well established.”). “From a\\ntribe’s inherent sovereign powers flow lesser powers,\\nincluding the power to regulate [nonmembers] on tribal\\nland.” Water Wheel, 642 F.3d at 808–09 (citing South\\nDakota v. Bourland, 508 U.S. 679, 689 (1993)).\\n\\n The Court has made clear, however, “that once tribal\\nland is converted into fee simple [land], the tribe loses\\nplenary jurisdiction over it . . . . As a general rule, then ‘the\\ntribe has no authority itself, by way of tribal ordinance or\\nactions in the tribal courts, to regulate the use of fee land.’”\\nPlains Commerce Bank v. Long Family Land & Cattle Co.,\\n554 U.S. 316, 328–29 (2008) (quoting Brendale v.\\nConfederated Tribes & Bands of Yakima Indian Nation,\\n492 U.S. 408, 430 (1989) (plurality opinion)). In Montana\\nv. United States, the Court recognized two exceptions to this\\n\\f14 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\ngeneral rule. First, “[a] tribe may regulate, through taxation,\\nlicensing, or other means, the activities of nonmembers who\\nenter consensual relationships with the tribe or its members,\\nthrough commercial dealing, contracts, leases, or other\\narrangements.” Montana, 450 U.S. at 565. Second, a tribe\\nmay exercise civil authority over the conduct of\\nnonmembers on fee lands within its reservation when “that\\nconduct threatens or has some direct effect on the political\\nintegrity, the economic security, or the health or welfare of\\nthe tribe.” Id. at 566.\\n\\n “Since deciding Montana, the Supreme Court has\\napplied those exceptions almost exclusively to questions of\\njurisdiction arising on [non-tribal] land.” Water Wheel,\\n642 F.3d at 809. The exception is Nevada v. Hicks, 533 U.S.\\n353 (2001). Water Wheel, 642 F.3d at 809. In Hicks, the\\nCourt addressed a tribal court’s jurisdiction over claims\\nagainst state officers arising from the execution of a search\\nwarrant on tribal land for alleged violations of state poaching\\nlaws—specifically, the killing of bighorn sheep off the\\nreservation. 533 U.S. at 356–57. Both the state court and\\nthen the tribal court issued search warrants. Id. at 356. The\\nCourt stated that although ownership status of the land “may\\nsometimes be a dispositive factor” in determining a tribe’s\\nauthority to regulate nonmember activity on tribal land, the\\ntribe’s power to exclude nonmembers from tribal land was\\n“not alone enough to support” the tribe’s regulatory\\njurisdiction over the state officers’ activities when the state\\nhad a competing interest in executing a warrant for an off-\\nreservation crime. Id. at 360. The Court applied Montana\\nand concluded that “tribal authority to regulate state officers\\nin executing process related to the violation, off reservation,\\nof state laws is not essential to tribal self-government or\\ninternal relations” while “[t]he State’s interest in execution\\nof process is considerable.” Id. at 364.\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 15\\n\\n Although some jurisdictions have interpreted Hicks as\\neliminating the right-to-exclude framework as an\\nindependent source of regulatory power over nonmember\\nconduct on tribal land, we have declined to do so. In Water\\nWheel, we observed that Hicks “expressly limited its holding\\nto ‘the question of tribal-court jurisdiction over state officers\\nenforcing state law.’” Water Wheel, 642 F.3d at 813\\n(quoting Hicks, 533 U.S. at 358 n.2). Indeed, the Hicks\\nCourt specifically “le[ft] open the question of tribal-court\\njurisdiction over nonmember defendants in general.” Hicks,\\n533 U.S. at 358 n.2. In Water Wheel, we held that a “tribe’s\\nstatus as landowner is enough to support regulatory\\njurisdiction” except “when the specific concerns at issue [in\\nHicks] exist.” 642 F.3d at 813. “Doing otherwise would\\nimpermissibly broaden Montana’s scope beyond what any\\nprecedent requires and restrain tribal sovereign authority\\ndespite Congress’s clearly stated federal interest in\\npromoting tribal self-government.” Id. at 813.\\n\\n In Hicks, the defendants were state officers enforcing a\\nstate-court-issued search warrant, so there was a significant\\nstate interest at stake. By contrast, the present case involves\\na private, consensual employment relationship between\\nKnighton and the Tribe, which occurred primarily on tribal\\nland. There are no significant competing state interests, as\\nin Hicks. Accordingly, our Water Wheel precedent compels\\nthe conclusion that the Tribe possesses regulatory\\njurisdiction over its claims against Knighton.\\n\\n Since Hicks’s limited holding, the Court in Plains\\nCommerce Bank held that a tribal court did not have\\njurisdiction to adjudicate a discrimination claim concerning\\na non-Indian defendant’s sale of fee land. Plains Commerce\\nBank, 554 U.S. at 323, 340–41. The land in question was\\nsold as part of the 1908 Allotment Act and was owned by a\\n\\f16 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nnon-Indian party for at least 50 years. Id. at 331, 341. The\\nCourt found that the discrimination law that the plaintiffs\\nwere attempting to enforce operated as a restraint on\\nalienation and had the effect of regulating the substantive\\nterms on which the non-Indian bank was able to offer its fee\\nland for sale. Id. at 331. The Court stated that while\\n“Montana and its progeny permit tribal regulation of\\nnonmember conduct inside the reservation that implicates\\nthe tribe’s sovereign interests,” that case “does not permit\\nIndian tribes to regulate the sale of non-Indian fee land,” as\\nneither of the Montana exceptions applies. Id. at 332. By\\ncontrast, in the present case the nonmember defendant while\\non tribal land allegedly used her position as Tribal\\nAdministrator to violate the terms of her employment in a\\nwide variety of ways that were significantly detrimental to\\nthe management and financial security of the Tribe.\\n\\n B. Appellant’s Arguments\\n\\n Knighton argues that treating ownership status of the\\nland as a dispositive factor in upholding a tribe’s power to\\nregulate nonmember conduct on tribal land (unless, as in\\nHicks, there are significant state interests present) is contrary\\nto our prior rulings in McDonald v. Means, 309 F.3d 530\\n(9th Cir. 2002), and Smith v. Salish Kootenai College. We\\ndisagree. In McDonald, we specifically recognized that a\\ntribe’s jurisdiction over civil claims against nonmembers\\narising on tribal land is limited under Hicks only in cases\\nwhere significant state interests are present. See 309 F.3d\\nat 540. And in Window Rock Unified School District v.\\nReeves, 861 F.3d 894, 902 n.9 (9th Cir. 2017), we concluded\\nthat Smith did not limit a tribe’s jurisdiction over civil claims\\nagainst nonmembers bearing a direct connection to tribal\\nland. We concluded that Smith was distinguishable because\\nit involved a nonmember plaintiff, as opposed to a\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 17\\n\\nnonmember defendant, who had entered into a consensual\\nrelationship with the tribe by filing his action in tribal court.\\nId.\\n\\n Knighton’s argument that a tribe’s regulatory power over\\nnonmember conduct on tribal land is limited to conduct that\\ndirectly interferes with a tribe’s inherent powers to exclude\\nand manage its own lands is also unavailing. In Window\\nRock, we concluded that the tribal court’s jurisdiction over\\nemployment-related claims that did not involve access to\\ntribal land was plausible; accordingly, we held that the\\nnonmember defendants were required to exhaust their tribal\\ncourt remedies before proceeding in federal court. Id. at 896,\\n906. Moreover, limiting a tribe’s regulatory power over\\nnonmember conduct to that which directly interferes with a\\ntribe’s inherent powers to exclude and manage its own lands,\\nas Knighton suggests, would restrict tribal sovereignty\\nabsent explicit authorization from Congress—an approach\\nwe specifically rejected in Water Wheel. See 642 F.3d at 812\\n(stating that the tribe’s right to exclude nonmembers from\\ntribal land includes the power to regulate them “unless\\nCongress has said otherwise, or unless the Supreme Court\\nhas recognized that such power conflicts with federal\\ninterests promoting tribal self government”).\\n\\n Knighton also argues that under the facts of this case,\\nWater Wheel’s right-to-exclude framework is inapplicable\\nbecause some of her alleged misconduct occurred off tribal\\nland, after the tribal administrative offices were relocated to\\nfee land owned by the Tribe. Although the Tribe’s\\ncomplaint does not allege precisely where the conduct at\\nissue occurred, most of the claims alleged against Knighton\\ninvolve conduct that took place on tribal land, before the\\nTribe’s administrative offices were moved in mid-2009 to\\nthe RISE building in Alturas, California. Moreover, the facts\\n\\f18 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nof this case are unique in that any claims that may have\\narisen outside tribal land are based on alleged misconduct\\nand misrepresentations made by Knighton on tribal land.\\nSee Smith, 434 F.3d at 1135 (stating that jurisdictional\\ninquiry is not limited to deciding precisely when and where\\nthe claim arose, but whether it bears some direct connection\\nto tribal lands). For example, the $29,925 overpayment for\\nunused vacation and sick leave that the Tribe seeks to\\nrecover stems from misrepresentations that Knighton\\nallegedly made throughout the course of her employment,\\nbefore the Tribe’s administrative offices relocated. In\\naddition, the relocation of the Tribe’s administrative offices\\nfrom tribal land to the RISE building on tribal fee land was\\nallegedly due to misrepresentations by Knighton.\\n\\n Knighton further argues that even if the Tribe had the\\npower to regulate her conduct on tribal land during the\\ncourse of her employment under Water Wheel’s right-to-\\nexclude framework, the Tribe’s authority is limited to the\\nregulations that were in place during her employment—\\nwhich is to say, those provided for in the Personnel Manual.\\nKnighton contends that the Tribe is attempting to impose\\nnew regulations on her through tort law after she left her\\nemployment with the Tribe.\\n\\n A tribe’s power to exclude nonmembers from tribal lands\\npermits a tribe to condition a nonmember’s entry or\\ncontinued presence on tribal land, see Merrion, 455 U.S. at\\n144–45, but this inherent power does not permit the Tribe to\\nimpose new regulations upon Knighton’s conduct\\nretroactively when she is no longer present on tribal land.\\nHowever, we agree with the district court that Knighton’s\\nalleged conduct violated the Tribe’s regulations that were in\\nplace at the time of her employment. The Personnel Manual\\nregulated employee conduct including, but not limited to,\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 19\\n\\nmisfeasance or malfeasance in the performance of duty,\\nincompetency in the performance of job duties, theft,\\ncarelessness or negligence with the monies or property of the\\nRancheria, inducement of an employee to act in violation of\\nRancheria regulations, and violation of personnel rules—all\\nconduct that forms the basis of the Tribe’s claims against\\nKnighton.\\n\\n C. Sources of Authority\\n\\n In Water Wheel, we concluded that a tribe’s inherent\\nsovereign power to exclude nonmembers from tribal land\\nprovides an independent basis upon which a tribe may\\nregulate the conduct of nonmembers on tribal land. But, a\\ntribe’s power to exclude is not the only source of its\\nregulatory authority over nonmembers on tribal land. See\\nBrendale, 492 U.S. at 425 (“An Indian tribe’s [] power to\\nexclude nonmembers of the tribe from its lands is not the\\nonly source of Indian regulatory authority.”). “[T]ribes have\\ninherent sovereignty independent of that authority arising\\nfrom their power to exclude.” Id. (citing Merrion, 455 U.S.\\nat 141); see also Babbitt Ford, Inc. v. Navajo Indian Tribe,\\n710 F.2d 587, 592 (9th Cir. 1983) (“The power to exercise\\ntribal civil authority over [nonmembers] derives not only\\nfrom the tribe’s inherent powers necessary to self-\\ngovernment and territorial management, but also from the\\npower to exclude nonmembers from tribal land.” (citing\\nMerrion, 455 U.S. at 141–44)).\\n\\n In addition to the power to exclude, we have the\\nMontana Court’s acknowledgment that Indian tribes retain\\ntheir inherent sovereign power to protect tribal self-\\ngovernment and to control internal relations. 450 U.S.\\nat 564. “[I]n accordance with that right tribes ‘may regulate\\nnonmember behavior that implicates [these sovereign\\ninterests].’” Attorney’s Process & Investigation Servs., Inc.\\n\\f20 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nv. Sac & Fox Tribe of Miss. in Iowa, 609 F.3d 927, 936\\n(8th Cir. 2010) (quoting Plains Commerce Bank, 554 U.S. at\\n335).\\n\\n Subsequent to Montana, in Merrion, the Court affirmed\\nthat Indian tribes have inherent sovereign power to regulate\\nnonmember conduct on tribal land independent of that\\nauthority arising from their power to exclude. Merrion,\\n455 U.S. at 144. The Court in Merrion concluded that a\\ntribe’s power to tax nonmember mining and drilling on tribal\\nland derived from its inherent “power to govern and to pay\\nfor the costs of self-government,” and concluded that such\\nregulatory authority was also within the tribe’s inherent\\npower to condition the continued presence of nonmembers\\non tribal land. Id. at 144–45. These varied sources of tribal\\nregulatory power over nonmember conduct on the\\nreservation were affirmed by the Court in Plains Commerce\\nBank. 554 U.S. at 337 (“[T]he regulation must stem from\\nthe tribe’s inherent sovereign authority to set conditions on\\nentry, preserve tribal self-government, or control internal\\nrelations.”).\\n\\n While the district court believed that our caselaw\\nprohibited the application of the Montana framework to\\ntribal jurisdictional issues involving nonmember conduct on\\ntribal land, it also recognized that a tribe’s regulatory power\\nover nonmembers on tribal land does not solely derive from\\nits power to set conditions on entry or continued presence.\\nAccordingly, it concluded that the Tribe had regulatory\\njurisdiction over Knighton’s conduct because “Knighton’s\\nemployment activities directly affected the Tribe’s inherent\\npowers to protect the welfare of its members and preserve\\nthe integrity of its government,” and because “her conduct\\nthreatened the Tribe’s very economic survival.”\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 21\\n\\n We now clarify Water Wheel and our subsequent cases\\ninvolving tribal jurisdictional issues on tribal land do not\\nexclude Montana as a source of tribal regulatory authority\\nover nonmember conduct on tribal land. Rather, our caselaw\\nstates that an Indian tribe has power to regulate nonmember\\nconduct on tribal land incident to its sovereign power to\\nexclude nonmembers from tribal land, regardless of whether\\neither of the Montana exceptions is satisfied. See Water\\nWheel, 642 F.3d at 814 (“[T]he tribe’s status as landowner is\\nenough to support regulatory jurisdiction without\\nconsidering Montana.” (emphasis added)); Grand Canyon\\nSkywalk Dev., LLC v. ‘Sa’ Nyu Wa Inc., 715 F.3d 1196, 1204\\n(9th Cir. 2013) (“[A] tribe’s inherent authority over tribal\\nland may provide for regulatory authority over\\n[nonmembers] on that land without the need to consider\\nMontana.” (emphasis added)); Window Rock, 861 F.3d at\\n902 (“[I]n civil cases involving nonmember conduct on\\ntribal land, we have held that tribal courts have jurisdiction\\nunless a treaty or federal statute provides otherwise—\\nregardless of whether the Montana exceptions would be\\nsatisfied.” (emphasis added)). Certainly, as our caselaw has\\ndiscussed at length, without evidence of a contrary intent by\\nCongress, a tribe’s power to regulate nonmember conduct on\\ntribal land flows from its inherent power to exclude and is\\ncircumscribed only to the limited extent that the\\ncircumstances in Hicks—significant state interests—are\\npresent. See Water Wheel, 642 F.3d at 813; Grand Canyon,\\n715 F.3d at 1205; Window Rock, 861 F.3d at 902. However,\\nthe Court has made clear that a tribe also has sovereign\\nauthority to regulate nonmember conduct on tribal lands\\nindependent of its authority to exclude if that conduct\\nintrudes on a tribe’s inherent sovereign power to preserve\\nself-government or control internal relations. The Montana\\nexceptions are “rooted” in the tribes’ inherent power to\\nregulate nonmember behavior that implicates these\\n\\f22 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nsovereign interests. Attorney’s Process, 609 F.3d at 936\\n(citing Plains Commerce Bank, 554 U.S. at 335).\\n\\n Accordingly, although we conclude that the Tribe had\\nauthority to regulate Knighton’s conduct on tribal land\\npursuant to its sovereign exclusionary powers, a separate\\nquestion remains as to whether the Tribe also had regulatory\\nauthority over Knighton’s conduct pursuant to Montana.\\n\\n i. First Montana Exception\\n\\n Montana’s consensual relationship exception recognizes\\nthat tribes have jurisdiction to regulate consensual relations\\n“through taxation, licensing, or other means.” 450 U.S.\\nat 565. Courts have recognized that tort law, under which\\nthe Tribe’s claims against Knighton arise, constitutes a form\\nof regulation. See Attorney’s Process, 609 F.3d at 938\\n(stating that if a tribe retains the power under Montana to\\nregulate nonmember conduct, it does not make any\\ndifference whether it does so through precisely tailored\\nregulations or through tort claims). However, Montana’s\\nconsensual relationship exception requires that “the\\nregulation imposed by the Indian tribe have a nexus to the\\nconsensual relationship itself.” Atkinson Trading Co. v.\\nShirley, 532 U.S. 645, 656 (2001). “A nonmember’s\\nconsensual relationship in one area thus does not trigger\\ntribal civil authority in another.” Id.\\n\\n Examining the facts of this case, we conclude that the\\nTribe has regulatory authority over Knighton’s conduct in\\nthis case under Montana’s consensual relationship\\nexception. The conduct that the Tribe seeks to regulate\\nthrough tort law arises directly out of the consensual\\nemployment relationship between the Tribe and Knighton.\\nMoreover, given the circumstances, Knighton should have\\nreasonably anticipated that her conduct might “trigger” tribal\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 23\\n\\nauthority. Water Wheel, 642 F.3d at 818 (quoting Plains\\nCommerce Bank, 554 U.S. at 338). Knighton is no stranger\\nto the Tribe’s governance and laws. She had been an\\nemployee of the Tribe for approximately sixteen years and,\\nas Tribal Administrator, was responsible for the overall\\nsupervision and management of tribal operations and\\ncarrying out tribal projects consistent with the Tribal\\nConstitution. The Tribal Constitution, adopted\\napproximately two years before Knighton resigned as Tribal\\nAdministrator, specifically provided that the “jurisdiction of\\n[the Tribe] shall extend to land now within the confines of\\nthe [Rancheria] and to such other lands as may thereafter be\\nadded thereto.” We conclude that given these\\ncircumstances, Knighton should reasonably have anticipated\\nthat her conduct on tribal land would fall within the Tribe’s\\nregulatory jurisdiction.\\n\\n ii. Second Montana Exception\\n\\n In determining whether Knighton’s conduct threatens or\\nhas some direct effect on the political integrity, the economic\\nsecurity, or the health or welfare of the tribe—the second\\nMontana exception, 450 U.S. at 566—we find instructive the\\nEighth Circuit’s decision in Attorney’s Process. In that case,\\nAPI, a nonmember corporation, was hired by a tribal\\ngovernment leader who refused to step down from\\nleadership after he lost in a special tribal election. 609 F.3d\\nat 932. Under their contract, API agreed to perform services\\nrelating to “the investigation of a takeover by dissidents at\\nthe Tribe’s facility located on the Tribe’s reservation lands.”\\nId. As the newly elected tribal council occupied the casino\\nand tribal government offices, approximately thirty API\\nagents forced their way into both buildings, which were\\nlocated on tribal land. Id. The agents were armed with\\nbatons, at least one carried a firearm, and they seized\\n\\f24 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nconfidential information from both facilities related to the\\ntribe’s gaming operations and finances. Id. In addition to\\nthe wrongfully seized confidential information, the agents\\ncaused approximately $7,000 in property damage and\\ncommitted various intentional torts against tribal members.\\nId. The tribe filed suit in tribal court for trespass to tribal\\nland and chattels, misappropriation of trade secrets, and\\nother claims. Id.\\n\\n API argued that tort claims do not in the ordinary course\\nthreaten the political integrity, economic security, or the\\nhealth or welfare of the tribe and thus the tribal court had no\\njurisdiction over the tribe’s claims under Montana’s second\\nexception. Id. at 937. Relying on Plains Commerce Bank,\\nthe court in Attorney’s Process stated that courts “should not\\nsimply consider the abstract elements of the tribal claim at\\nissue, but must focus on the specific nonmember conduct\\nalleged, taking a functional view of the regulatory effect of\\nthe claim on the nonmember.” Id. at 938. The court\\nconcluded that API’s raid on the casino and government\\noffices, leading to the claims for trespass to land, trespass to\\nchattels, and misappropriation of tribal trade secrets,\\n“menace[d] the ‘political integrity, the economic security,\\n[and] the health [and] welfare’ of the Tribe to such a degree\\nthat it ‘imperil[ed] the subsistence’ of the tribal community”\\nand that the tribe therefore retained the inherent power under\\nthe second Montana exception to regulate that conduct. Id.\\nat 939 (alterations in original) (quoting Plains Commerce\\nBank, 554 U.S. at 341).\\n\\n While Knighton’s conduct constitutes a different type of\\nviolation, it was of long duration and had a great impact upon\\nthe Tribe, and so we conclude that the alleged harm to the\\nTribe caused by her conduct “‘imperil[ed] the subsistence’\\nof the tribal community.” Evans v. Shoshone-Bannock Land\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 25\\n\\nUse Policy Comm’n, 736 F.3d 1298, 1306 (9th Cir. 2013)\\n(quoting Plains Commerce Bank, 554 U.S. at 341). Among\\nthe tribe’s many claims are allegations that Knighton\\ninvested the Tribe’s money without appropriate authority,\\nconcealed investment documents and audit reports from the\\nTribe, and attempted to enter financial agreements without\\nthe appropriate authorization or waiver of tribal sovereign\\nimmunity. The Tribe also alleges that Knighton made\\nunreasonably risky investments that led to investment losses\\nin excess of $1.2 million, excess transaction fees, and state\\nand federal tax exposure, and that she breached her fiduciary\\nduty and deceived the Tribe, causing it to pay $300,000,\\n$150,000 above market value, for the RISE building\\npurchase. Finally, the Tribe alleges that when she resigned\\nher employment with the Tribe, Knighton took all files,\\nincluding files belonging to the Tribe, room furnishings, and\\na computer, representing to the Tribe that the property\\nremoved belonged to RISE. We conclude that this conduct\\nthreatened the Tribe’s very subsistence and that the Tribe\\ntherefore retains the inherent power under the second\\nMontana exception to regulate this conduct.\\n\\nII. Adjudicatory Jurisdiction\\n\\n Knighton also contends that the Tribe is seeking to\\nexercise greater adjudicative authority over her than it was\\ncapable of at the time of her employment. She argues that\\nthe adjudicatory authority of the Tribe is limited to the\\ndisciplinary procedures provided for in the Personnel\\nManual. At the time of her employment, the disciplinary\\nactions detailed in the Personnel Manual for an employee’s\\nbreach of rules and standards of conduct in the course of\\nemployment included a verbal warning, written reprimand,\\nsuspension without pay, demotion, and involuntary\\ntermination. The Personnel Manual provided that when the\\n\\f26 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\nTribal Administrator was the subject of disciplinary action,\\nthe Community Council directly oversaw the disciplinary\\nand grievance procedures.\\n\\n We hold that the Tribe has the power to regulate\\nKnighton’s conduct incident to its sovereign powers to\\nexclude nonmembers from tribal land, and also, in the\\nalternative, under both Montana exceptions. “[W]here tribes\\npossess authority to regulate the activities of nonmembers,\\n‘[c]ivil jurisdiction over [disputes arising out of] such\\nactivities presumptively lies in the tribal courts.’” Strate v.\\nA-1 Contractors, 520 U.S. 438, 453 (1997) (second and third\\nalterations in original) (quoting Iowa Mut. Ins. Co. v.\\nLaPlante, 480 U.S. 9, 18 (1987)). However, a tribe’s\\nadjudicative authority over nonmembers may not exceed its\\nregulatory authority. Id.\\n\\n We conclude that under the facts of this case, the Tribe’s\\nadjudicatory authority does not exceed the regulatory\\nauthority it had over Knighton’s conduct during her\\nemployment under Water Wheel’s right-to-exclude\\nframework. As discussed above, the Personnel Manual\\nregulated the conduct that forms the basis of the Tribe’s\\nclaims against Knighton and conferred jurisdiction over her\\nconduct as Tribal Administrator on the Community Council.\\nThe fact that the Tribe now seeks to adjudicate these claims\\nin the Tribal Court does not undermine its jurisdiction over\\nthe Tribe’s claims.\\n\\n Likewise, examining the Tribe’s adjudicative authority\\nover Knighton’s conduct under Montana, we return to the\\nilluminating Eighth Circuit opinion in Attorney’s Process.\\nSimilar to this case, in Attorney’s Process, the tribal court\\nsystem was established after the tort claims against API\\narose. 609 F.3d at 933. API argued that the tribe lacked\\njurisdiction over its claims because there were no written\\n\\f KNIGHTON V. CEDARVILLE RANCHERIA OF NPI 27\\n\\nregulations in place at the time which prohibited the tortious\\nconduct that API was alleged to have committed. Id. at 938.\\nThe court stated that “[i]f the Tribe retains the power under\\nMontana to regulate such conduct, we fail to see how it\\nmakes any difference whether it does so through precisely\\ntailored regulations or through tort claims such as those at\\nissue [in the case].” Id. The court concluded that because\\nAPI’s intervention onto tribal land threatened the “‘political\\nintegrity, the economic security, [and] the health [and]\\nwelfare’ of the Tribe,” the tribe had the authority to regulate\\nand adjudicate such conduct under Montana, as well as\\nincident to its sovereign right to exclude nonmembers from\\ntribal land. Id. at 940 (alterations in original) (quoting\\nMontana, 450 U.S. at 566).\\n\\n As the court in Attorney’s Process recognized, our task\\nis to outline the boundaries of the inherent sovereign power\\nretained by the Indian tribes. “Those boundaries are\\nestablished by federal law, a source of law external to the\\ntribes.” Id. at 938 (citing Nat’l Farmers Union, 471 U.S.\\nat 852). In contrast, “positive tribal law,” the court stated,\\n“is internal to the tribes.” Id. “It is a manifestation of tribal\\npower, and as such it does not contribute to the external\\nlimitations which concern us here. Once it is determined that\\ncertain conduct is within the scope of a tribe’s power as a\\nmatter of federal law, our inquiry is at an end.” Id.\\n\\n In the present case, the Tribe’s authority to regulate\\nKnighton’s conduct derived not only from its sovereign\\npower to exclude nonmembers from tribal lands, but also\\nfrom its inherent sovereign power to regulate consensual\\nrelations with nonmembers “through taxation, licensing, or\\nother means,” and to protect the “political integrity, the\\neconomic security, [and] the health [and] welfare” of the\\nTribe. Montana, 450 U.S. at 565–66.\\n\\f28 KNIGHTON V. CEDARVILLE RANCHERIA OF NPI\\n\\n Once the authority to regulate nonmember conduct\\nexists, whether from Water Wheel or from Montana, then the\\nobservation from the court in Attorney’s Process persuades\\nus that it makes no difference whether the Tribe adjudicates\\nKnighton’s conduct through the Personnel Manual or\\nthrough tort law.\\n\\n CONCLUSION\\n\\n There is no general rule as to the extent of a tribe’s\\nadjudicative jurisdiction over non-Indians on tribal land, but\\n“it is clear that the general rule announced in Strate, and\\nconfirmed in Hicks and Plains Commerce Bank, that\\nadjudicative jurisdiction is confined by the bounds of a\\ntribe’s regulatory jurisdiction” applies. Water Wheel,\\n642 F.3d at 814. Given the existence of regulatory authority,\\nthe sovereign interests at stake, and the congressional\\ninterest in promoting tribal self-government, we conclude\\nthat the Tribal Court has jurisdiction over the Tribe’s claims\\nin this case.\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1286,"cells":{"id":{"kind":"number","value":4599152,"string":"4,599,152"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Advanced Building & Fabricatio v. Curtis Ayers"},"decision_date":{"kind":"string","value":"2019-03-13"},"docket_number":{"kind":"string","value":"17-16618"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Milan D. Smith, Jr. and Jacqueline H. Nguyen, Circuit Judges, and Jane A. Restani, * Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"NGUYEN, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nADVANCED BUILDING & No. 17-16618\\nFABRICATION, INC., a California\\ncorporation; ROBERT HONAN, D.C. No.\\n Plaintiffs-Appellees, 2:13-cv-02380-\\n MCE-CKD\\n v.\\n\\nCALIFORNIA HIGHWAY PATROL; OPINION\\nJOHN WILSON,\\n Defendants,\\n\\n and\\n\\nCURTIS AYERS,\\n Defendant-Appellant.\\n\\n\\n\\n Appeal from the United States District Court\\n for the Eastern District of California\\n Morrison C. England, Jr., District Judge, Presiding\\n\\n Argued and Submitted December 19, 2018\\n San Francisco, California\\n\\n Filed March 13, 2019\\n\\f2 ADVANCED BLDG. & FABRICATION V. AYERS\\n\\n Before: Milan D. Smith, Jr. and Jacqueline H. Nguyen,\\n Circuit Judges, and Jane A. Restani, * Judge.\\n\\n Opinion by Judge Nguyen\\n\\n\\n SUMMARY **\\n\\n\\n Civil Rights\\n\\n The panel affirmed the district court’s denial, on\\nsummary judgment, of qualified immunity to Curtis Ayers,\\na former employee of the California State Board of\\nEqualization, in an action brought against him by plaintiff\\nRobert Honan and his business alleging Ayers violated\\nclearly established law by participating in a search of\\nplaintiff’s business following an altercation between Ayers\\nand plaintiff.\\n\\n Ayers, who had mistaken plaintiff’s business for another\\nbusiness, arrived at plaintiff’s premises without an\\nappointment, and identified himself as a Board of\\nEqualization employee, but did not present his ID badge. A\\nverbal altercation ensued and Ayers alleged that as he was\\nleaving, plaintiff tackled him and pushed him through the\\ndoor, resulting in Ayers dropping and damaging his laptop.\\nAyers reported the incident and the California Highway\\nPatrol subsequently obtained a search warrant to search the\\n\\n *\\n The Honorable Jane A. Restani, Judge for the United States Court\\nof International Trade, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f ADVANCED BLDG. & FABRICATION V. AYERS 3\\n\\npremises of plaintiff’s business on the grounds that plaintiff\\ncommitted felony threats and vandalism against Ayers.\\nCalifornia Highway Patrol executed the warrant and Ayers\\nparticipated in the search, during which he allegedly\\nsearched through plaintiff’s personal property.\\n\\n Plaintiff sued Ayers and other defendants, including\\nCalifornia Highway Patrol officers, alleging claims for state-\\nlaw conversion and other tort claims, as well as a federal\\nclaim under 42 U.S.C. § 1983. The panel affirmed the denial\\nof qualified immunity to Ayers and resolved the remaining\\nclaims against the other defendants in concurrently filed\\nmemorandum dispositions.\\n\\n The panel rejected Ayers’ contention that there was no\\nFourth Amendment violation because California law permits\\nState Board of Equalization employees to request inspection\\nof sales records from any business, and that therefore his\\nparticipation in the search would qualify under the\\n“administrative search” exception to the warrant\\nrequirement. The panel held that none of the state provisions\\ncited by Ayers authorized forcible entry or searches.\\nMoreover, even assuming that state law permitted\\nwarrantless inspections of business records, the intrusive\\nsearch here would not withstand scrutiny under the Fourth\\nAmendment. The panel further held that Ayers’s presence\\nwas not necessary to aid in the officers’ execution of the\\nwarrant, which had the purpose of seizing evidence relating\\nto allegations of criminal threats and vandalism. Because\\nAyers’s presence was not related to the objective of the\\nsearch warrant, his conduct violated plaintiff’s Fourth\\nAmendment rights. The panel remanded for further\\nproceedings consistent with the panel’s opinion.\\n\\f4 ADVANCED BLDG. & FABRICATION V. AYERS\\n\\n COUNSEL\\n\\nMichael von Loewenfeldt (argued) and Daniel J. Veroff,\\nKerr & Wagstaffe LLP, San Francisco, California, for\\nDefendant-Appellant.\\n\\nPolly J. Estes (argued), Estes Law Group, San Rafael,\\nCalifornia, for Plaintiffs-Appellees.\\n\\n\\n OPINION\\n\\nNGUYEN, Circuit Judge:\\n\\n Curtis Ayers, a former employee of the California State\\nBoard of Equalization (“BOE”), appeals the district court’s\\ndenial of qualified immunity in a suit brought against him by\\nPlaintiffs Robert Honan and his business, Advanced\\nBuilding & Fabrication, Inc. (“Advanced Building”). The\\nallegations in the complaint stem from an altercation\\nbetween Ayers and Honan, which led to the execution of a\\nsearch warrant at Advanced Building by officers of the\\nCalifornia Highway Patrol (“CHP”). Honan contends that\\nAyers violated clearly established law by participating in the\\nsearch. We agree. Because the administrative search\\nexception does not apply and Ayers’s presence was not\\nnecessary to aid in the officers’ execution of the warrant, we\\naffirm the district court’s denial of qualified immunity. 1\\n\\n\\n 1\\n Honan sued numerous other defendants in addition to Ayers,\\nincluding CHP Officer John Wilson and the CHP itself. Citing numerous\\nfactual disputes, the district court denied summary judgment and\\nimmunity to all the defendants. We address only Ayers’s appeal of the\\ndenial of qualified immunity in this opinion, and resolve the remaining\\n\\f ADVANCED BLDG. & FABRICATION V. AYERS 5\\n\\n I.\\n\\n BACKGROUND\\n\\n Robert Honan owns and operates Advanced Building, a\\nmetal fabrication business that makes metal machines and\\nspecialized metal parts. On May 7, 2012, Curtis Ayers, who\\nat that time was employed by the BOE, arrived at Advanced\\nBuilding without a prior appointment. Ayers identified\\nhimself as a BOE employee, but he did not present his ID\\nbadge and had mistaken Advanced Building for another\\nbusiness. Honan suspected Ayers of a “fraudulent scam”\\nand, after a brief exchange of words, Honan told Ayers to\\n“Get the F* out.”\\n\\n The parties dispute what happened next, but, as Ayers\\nwas leaving (or being thrown out of) the building, his laptop\\ndropped and sustained damaged. According to Honan,\\nAyers must have heard Honan approaching and “became\\nstartled and stumbled out the exit door, falling to the\\nground.” Ayers, on the other hand, claimed that Honan\\ntackled him and pushed him through the door as he was\\nleaving. In the process, Ayers alleges, the laptop flew out of\\nhis hands, the monitor broke, and the laptop was rendered\\ninoperable.\\n\\n Ayers contacted his supervisor, Dwayne Sims to report\\nthe incident. Sims recommended that Ayers contact the\\npolice, and they ultimately reported the incident to the CHP.\\nCHP Officer John Wilson responded and interviewed Ayers.\\nOfficer Wilson subsequently detained Honan along the\\n\\nclaims of state law immunity raised by Ayers on appeal in a\\nmemorandum disposition filed concurrently with this opinion.\\nSimilarly, we resolve Officer Wilson’s and the CHP’s separate appeal,\\nNo. 17-16669, in a separately filed memorandum disposition.\\n\\f6 ADVANCED BLDG. & FABRICATION V. AYERS\\n\\nhighway, searched his truck, and interviewed him. Honan\\noffered to discuss the incident and invited Officer Wilson\\nand his partner back to Advanced Building. Honan gave\\nthem a tour of Advanced Building and “showed them some\\nof the things Advanced was working on, including [an]\\nammunition reloader,” and “a couple gold nuggets from [his]\\ncollection.” The parties also dispute whether, during the\\ntour, Honan claimed to Officer Wilson that he had\\nsurveillance video that would corroborate his version of the\\nincident with Ayers.\\n\\n Officer Wilson subsequently obtained a search warrant\\nauthorizing “any sheriff, marshal or police officer in the\\ncounty of Butte” to lawfully search the premises of\\nAdvanced Building. To obtain the warrant, Officer Wilson\\nrecounted in an affidavit the events as described to him by\\nboth Ayers and Honan, and indicated that there was probable\\ncause to believe that Honan committed felony threats and\\nvandalism against Ayers. The warrant issued for officers to\\nsearch for and seize (1) items “capable of storing . . . video\\nand audio media,” (2) “indicia proving [Honan’s] ownership\\nand control” of the business, and (3) “evidence of\\ncorrespondence” that would corroborate the alleged felonies.\\nOfficer Wilson indicated to Sims that “he wanted us, me and\\nCurtis [Ayers] to . . . complete the inspection” during the\\nexecution of the search warrant because it was “important\\nthat [the BOE] complete [their] inspection and not allow this\\nbusiness or owner to get away with anything.” Sims\\naccepted the invitation, and told Ayers that he would like the\\ntwo of them to go during the search.\\n\\n On May 30, 2012, the CHP executed the search warrant\\nat Advanced Building. Sims and Ayers waited in the car\\nuntil the CHP confirmed they could enter. They “did not\\nassist the CHP in their search efforts nor were [they] asked\\n\\f ADVANCED BLDG. & FABRICATION V. AYERS 7\\n\\nto do so,” though Ayers did at one point identify Honan and\\nshow Officer Wilson where the assault occurred. Honan\\nasserts that he saw Ayers searching through a file cabinet of\\npersonal records. Ayers acknowledged that he examined\\nany “sales permits or records [they] saw,” but denied taking\\nor damaging any property. Honan alleges that a substantial\\namount of gold and silver was taken that day, as well as a\\nnumber of additional items and personal effects.\\n\\n Honan and Advanced Building collectively filed a\\ncomplaint against the CHP, Officer Wilson, Ayers, and other\\nunnamed individuals, alleging state-law claims for\\nconversion and other tort claims, as well as a federal claim\\nunder 42 U.S.C. § 1983. Ayers moved for summary\\njudgment, including on the grounds of federal- and state-law\\nimmunity. The district court denied summary judgment due\\nto “overwhelming” material factual disputes, and Ayers and\\nthe other defendants timely appealed.\\n\\n II.\\n\\n JURISDICTION AND STANDARD OF REVIEW\\n\\n We have jurisdiction under 28 U.S.C. § 1291 to review\\nan order denying summary judgment on the grounds of\\nqualified immunity. Plumhoff v. Rickard, 572 U.S. 765,\\n771–73 (2014). However, “the scope of our review over the\\nappeal [in this context] is circumscribed.” George v. Morris,\\n736 F.3d 829, 834 (9th Cir. 2013). We may not consider “a\\nfact-related dispute about the pretrial record,” that is,\\n“whether or not the evidence in the pretrial record was\\nsufficient to show a genuine issue of fact for trial.” Johnson\\nv. Jones, 515 U.S. 304, 307 (1995). Accordingly, we have\\njurisdiction only where “the issue appealed concerned, not\\nwhich facts the parties might be able to prove, but, rather,\\nwhether or not certain given facts showed” that immunity\\n\\f8 ADVANCED BLDG. & FABRICATION V. AYERS\\n\\nwould apply. See id. at 311 (quoting Mitchell v. Forsyth,\\n472 U.S. 511, 528 (1985)).\\n\\n Here, although Plaintiffs’ claim under § 1983 alleges a\\nnumber of constitutional violations by Defendants, we\\nconsider just one issue: whether Ayers’s participation in the\\nMay 2012 search at Advanced Building violated clearly\\nestablished law. Ayers’s claim of qualified immunity does\\nnot turn on the resolution of any factual dispute, and thus we\\nhave jurisdiction to review the district court’s denial of\\nqualified immunity. 2\\n\\n III.\\n\\n DISCUSSION\\n\\n To survive a claim of qualified immunity, the plaintiff\\nmust show: “(1) that the right was violated; and (2) that the\\nright was clearly established at the time of the alleged\\nmisconduct.” Isayeva v. Sacramento Sheriff’s Dep’t,\\n872 F.3d 938, 946 (9th Cir. 2017).\\n\\n A.\\n\\n Ayers argues that he did not violate Plaintiffs’ Fourth\\nAmendment rights because he claims California law permits\\na BOE employee to request inspection of sales records from\\nany business, and his participation in the search would thus\\nqualify under the “administrative search” exception to the\\n\\n 2\\n Ayers suggested at oral argument that “it’s not in the record here,\\n[but] one of the reasons he was asked to come along [was] to show where\\nthe altercation occurred.” Indeed, the record is silent as to this purported\\nrationale, but instead confirms that the purpose of inviting Ayers and\\nSims was for them to “complete the inspection” of Advanced Building’s\\nbusiness records.\\n\\f ADVANCED BLDG. & FABRICATION V. AYERS 9\\n\\nwarrant requirement. See, e.g., Cal. Rev. & Tax Code\\n§§ 7053–54; 18 Cal. Code Regs. § 1698(b)(1)). Ayers’s\\nreliance on these state provisions is a nonstarter because\\nnone of the state statutes or regulations on which he relies\\nauthorizes forcible entry or searches. For example, 18 Cal.\\nCode Regs. section 1698(b)(1) merely mandates a taxpayer\\nto “make available for examination . . . all records necessary\\nto determine the correct tax liability.” As Honan correctly\\npoints out, section 1698.5 (which immediately follows the\\nsection cited by Ayers) details the audit procedures and sets\\nforth a process for conferring and providing requested\\ndocumentation. Id. § 1698.5. The regulation even goes so\\nfar as to impose a duty on BOE staff to “not directly access\\nthe taxpayer’s computer system if the taxpayer objects to\\nsuch access, except in the case of a search warrant.” Id.\\n§ 1698.5(b)(4)(D). Nowhere in the regulation is there broad\\nauthorization to conduct a warrantless search and seizure of\\nbusiness records without first requesting them from the\\nbusiness owner. Ayers acknowledges that, consistent with\\nthese regulations, BOE protocol if an inspector is asked to\\nleave would be to “see if we can call back at a better time.”\\nTherefore, we conclude that Ayers’s conduct is not\\nauthorized by any state statute or regulation.\\n\\n Even assuming that state law permits warrantless\\ninspection of business records, the intrusive search here\\nwould not withstand scrutiny under the Fourth Amendment.\\nAyers argues that “[a]dministrative searches of a\\ncommercial location pursuant to statute can present an\\nexception to the general warrant requirement.” However, his\\nreading of precedent is too broad to support application of\\nsuch an exception to the warrant requirement here. For\\nexample, in New York v. Burger, the Supreme Court\\nconsidered the diminished privacy expectations for\\n“pervasively regulated business[es],” upholding warrantless\\n\\f10 ADVANCED BLDG. & FABRICATION V. AYERS\\n\\ninspections only where “necessary to further [the] regulatory\\nscheme.” 482 U.S. 691, 702 (1987) (alteration in original)\\n(quoting Donovan v. Dewey, 452 U.S. 594, 600 (1981)).\\nSimilarly, in United States v. Biswell, the Court upheld a\\nwarrantless inspection of a federally licensed weapons\\ndealer because “close scrutiny of [firearm] traffic is\\nundeniably of central importance to federal efforts to prevent\\nviolent crime.” 406 U.S. 311, 315 (1972).\\n\\n Here, no analogously pervasive regulation or special\\ngovernmental interest justifies a diminished expectation of\\nprivacy. Indeed, should Ayers be correct, it would seem that\\nevery business in California “selling tangible personal\\nproperty” would be subject to a warrantless search at any\\ntime. See Cal. Rev. & Tax Code § 7054. We conclude that\\nthe “administrative search” exception does not apply to\\nAyers’s presence at the search.\\n\\n B.\\n\\n We next consider whether Ayers could nevertheless\\nparticipate in the search at the invitation of the CHP. In\\nWilson v. Layne, the Supreme Court held that “it is a\\nviolation of the Fourth Amendment for police to bring . . .\\nthird parties into a home during the execution of a warrant\\nwhen the presence of the third parties in the home was not in\\naid of the execution of the warrant.” 526 U.S. 603, 614\\n(1999). There, police invited reporters on a “media ride-\\nalong” to execute arrest warrants at a private residence. Id.\\nat 605, 607. The Court held that the reporters’ presence\\nconstituted a violation of the plaintiff’s Fourth Amendment\\nrights because their presence “inside the home was not\\nrelated to the objectives of the authorized intrusion.” Id.\\nat 611. The Court distinguished situations where the third\\nparties “directly aided in the execution of the warrant,” such\\nas identifying stolen property that was the subject of the\\n\\f ADVANCED BLDG. & FABRICATION V. AYERS 11\\n\\nwarrant, conduct that “has long been approved by [the] Court\\nand our common-law tradition.” Id. at 611–12. By contrast,\\nthe Court rejected the presence of third parties serving only\\nto “further the law enforcement objectives of the police in a\\ngeneral sense,” rather than the specific “purposes of the\\nsearch.” Id. at 612 (emphasis added). The Court clarified\\nthat, in those situations, even if the police are lawfully\\npresent, “the violation of the Fourth Amendment is the\\npresence of the [third party].” Id. at 614 n.2.\\n\\n Under Wilson, Ayers’s participation in the search\\nviolated Plaintiffs’ Fourth Amendment rights. The warrant\\nissued for officers to search specifically for (1) items\\n“capable of storing . . . video and audio media,” (2) “indicia\\nproving [Honan’s] ownership and control” of the business,\\nand (3) “evidence of correspondence” that would\\ncorroborate the alleged felonies against Ayers. Ayers\\nacknowledged that he “attend[ed] the search to complete the\\ninspection” he had begun of Advanced Building’s business\\nlicensing and tax records. Such an inspection was obviously\\nnot “in aid of the execution of the warrant” to seize the above\\nevidence relating to the allegations of criminal threats and\\nvandalism. See Wilson, 526 U.S. at 614. Thus, because\\nAyers’s presence was not “related to the objectives of the\\nauthorized intrusion,” see id. at 611, his conduct violated\\nPlaintiffs’ Fourth Amendment rights.\\n\\n Moreover, this right was “clearly established” following\\nthe Court’s opinion in Wilson in 1999. While the Supreme\\nCourt has cautioned against “defin[ing] clearly established\\nlaw at a high level of generality,” see Ashcroft v. al-Kidd,\\n563 U.S. 731, 742 (2011), Wilson held explicitly that officers\\nmay not simply “bring members of the media or other third\\nparties . . . during the execution of a warrant” unless it was\\n“in aid of the warrant’s execution.” 526 U.S. at 614\\n\\f12 ADVANCED BLDG. & FABRICATION V. AYERS\\n\\n(emphasis added). Ayers argues that Wilson is not\\nsufficiently specific because that case involved reporters, not\\nother government agents. He contends that his position as a\\ngovernment employee—one charged with inspecting\\nbusiness records—distinguishes him from the journalists at\\nissue in Wilson.\\n\\n In fact, his liability under § 1983 is premised on this very\\ndistinction: private citizens (like reporters) are not ordinarily\\nliable under § 1983 for their presence during the execution\\nof a search warrant because they are not state actors. See\\nBrunette v. Humane Soc’y of Ventura Cty., 294 F.3d 1205,\\n1211 (9th Cir. 2002). Wilson did not rest constitutionality\\non the third party’s employer, but rather drew a bright line at\\nwhether their presence assisted execution of the warrant.\\nThe “contours of the right” here were “sufficiently clear that\\na reasonable official would understand that what he is doing\\nviolates that right.” See Wilson, 526 U.S. at 615 (quoting\\nAnderson v. Creighton, 483 U.S. 635, 640 (1987)). And,\\nalthough Ayers claims that he was “acting at the direction of\\nhis supervisor,” he did not cite to any binding precedent\\nholding that a supervisor’s instruction would somehow\\nobviate a clear constitutional violation. Accordingly, we\\nfind that Ayers’s conduct violated Plaintiffs’ clearly\\nestablished rights under Wilson.\\n\\n IV.\\n\\n CONCLUSION\\n\\n Because the administrative exception to the warrant\\nrequirement does not apply, and Ayers’s presence was not in\\naid of the execution of the search warrant, his presence and\\nsearch violated Plaintiffs’ clearly established constitutional\\nrights. We affirm the district court’s denial of qualified\\n\\f ADVANCED BLDG. & FABRICATION V. AYERS 13\\n\\nimmunity and remand for proceedings consistent with this\\nopinion.\\n\\n AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4376405/\", \"author_raw\": \"NGUYEN, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"MILAN D SMITH JR"},"judge2":{"kind":"string","value":"JACQUELINE H NGUYEN"},"judge3":{"kind":"string","value":"JANE A RESTANI"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4599152/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Advanced Building & Fabricatio v. Curtis Ayers\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-13\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nADVANCED BUILDING & No. 17-16618\nFABRICATION, INC., a California\ncorporation; ROBERT HONAN, D.C. No.\n Plaintiffs-Appellees, 2:13-cv-02380-\n MCE-CKD\n v.\n\nCALIFORNIA HIGHWAY PATROL; OPINION\nJOHN WILSON,\n Defendants,\n\n and\n\nCURTIS AYERS,\n Defendant-Appellant.\n\n\n\n Appeal from the United States District Court\n for the Eastern District of California\n Morrison C. England, Jr., District Judge, Presiding\n\n Argued and Submitted December 19, 2018\n San Francisco, California\n\n Filed March 13, 2019\n\f2 ADVANCED BLDG. & FABRICATION V. AYERS\n\n Before: Milan D. Smith, Jr. and Jacqueline H. Nguyen,\n Circuit Judges, and Jane A. Restani, * Judge.\n\n Opinion by Judge Nguyen\n\n\n SUMMARY **\n\n\n Civil Rights\n\n The panel affirmed the district court’s denial, on\nsummary judgment, of qualified immunity to Curtis Ayers,\na former employee of the California State Board of\nEqualization, in an action brought against him by plaintiff\nRobert Honan and his business alleging Ayers violated\nclearly established law by participating in a search of\nplaintiff’s business following an altercation between Ayers\nand plaintiff.\n\n Ayers, who had mistaken plaintiff’s business for another\nbusiness, arrived at plaintiff’s premises without an\nappointment, and identified himself as a Board of\nEqualization employee, but did not present his ID badge. A\nverbal altercation ensued and Ayers alleged that as he was\nleaving, plaintiff tackled him and pushed him through the\ndoor, resulting in Ayers dropping and damaging his laptop.\nAyers reported the incident and the California Highway\nPatrol subsequently obtained a search warrant to search the\n\n *\n The Honorable Jane A. Restani, Judge for the United States Court\nof International Trade, sitting by designation.\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f ADVANCED BLDG. & FABRICATION V. AYERS 3\n\npremises of plaintiff’s business on the grounds that plaintiff\ncommitted felony threats and vandalism against Ayers.\nCalifornia Highway Patrol executed the warrant and Ayers\nparticipated in the search, during which he allegedly\nsearched through plaintiff’s personal property.\n\n Plaintiff sued Ayers and other defendants, including\nCalifornia Highway Patrol officers, alleging claims for state-\nlaw conversion and other tort claims, as well as a federal\nclaim under 42 U.S.C. § 1983. The panel affirmed the denial\nof qualified immunity to Ayers and resolved the remaining\nclaims against the other defendants in concurrently filed\nmemorandum dispositions.\n\n The panel rejected Ayers’ contention that there was no\nFourth Amendment violation because California law permits\nState Board of Equalization employees to request inspection\nof sales records from any business, and that therefore his\nparticipation in the search would qualify under the\n“administrative search” exception to the warrant\nrequirement. The panel held that none of the state provisions\ncited by Ayers authorized forcible entry or searches.\nMoreover, even assuming that state law permitted\nwarrantless inspections of business records, the intrusive\nsearch here would not withstand scrutiny under the Fourth\nAmendment. The panel further held that Ayers’s presence\nwas not necessary to aid in the officers’ execution of the\nwarrant, which had the purpose of seizing evidence relating\nto allegations of criminal threats and vandalism. Because\nAyers’s presence was not related to the objective of the\nsearch warrant, his conduct violated plaintiff’s Fourth\nAmendment rights. The panel remanded for further\nproceedings consistent with the panel’s opinion.\n\f4 ADVANCED BLDG. & FABRICATION V. AYERS\n\n COUNSEL\n\nMichael von Loewenfeldt (argued) and Daniel J. Veroff,\nKerr & Wagstaffe LLP, San Francisco, California, for\nDefendant-Appellant.\n\nPolly J. Estes (argued), Estes Law Group, San Rafael,\nCalifornia, for Plaintiffs-Appellees.\n\n\n OPINION\n\nNGUYEN, Circuit Judge:\n\n Curtis Ayers, a former employee of the California State\nBoard of Equalization (“BOE”), appeals the district court’s\ndenial of qualified immunity in a suit brought against him by\nPlaintiffs Robert Honan and his business, Advanced\nBuilding & Fabrication, Inc. (“Advanced Building”). The\nallegations in the complaint stem from an altercation\nbetween Ayers and Honan, which led to the execution of a\nsearch warrant at Advanced Building by officers of the\nCalifornia Highway Patrol (“CHP”). Honan contends that\nAyers violated clearly established law by participating in the\nsearch. We agree. Because the administrative search\nexception does not apply and Ayers’s presence was not\nnecessary to aid in the officers’ execution of the warrant, we\naffirm the district court’s denial of qualified immunity. 1\n\n\n 1\n Honan sued numerous other defendants in addition to Ayers,\nincluding CHP Officer John Wilson and the CHP itself. Citing numerous\nfactual disputes, the district court denied summary judgment and\nimmunity to all the defendants. We address only Ayers’s appeal of the\ndenial of qualified immunity in this opinion, and resolve the remaining\n\f ADVANCED BLDG. & FABRICATION V. AYERS 5\n\n I.\n\n BACKGROUND\n\n Robert Honan owns and operates Advanced Building, a\nmetal fabrication business that makes metal machines and\nspecialized metal parts. On May 7, 2012, Curtis Ayers, who\nat that time was employed by the BOE, arrived at Advanced\nBuilding without a prior appointment. Ayers identified\nhimself as a BOE employee, but he did not present his ID\nbadge and had mistaken Advanced Building for another\nbusiness. Honan suspected Ayers of a “fraudulent scam”\nand, after a brief exchange of words, Honan told Ayers to\n“Get the F* out.”\n\n The parties dispute what happened next, but, as Ayers\nwas leaving (or being thrown out of) the building, his laptop\ndropped and sustained damaged. According to Honan,\nAyers must have heard Honan approaching and “became\nstartled and stumbled out the exit door, falling to the\nground.” Ayers, on the other hand, claimed that Honan\ntackled him and pushed him through the door as he was\nleaving. In the process, Ayers alleges, the laptop flew out of\nhis hands, the monitor broke, and the laptop was rendered\ninoperable.\n\n Ayers contacted his supervisor, Dwayne Sims to report\nthe incident. Sims recommended that Ayers contact the\npolice, and they ultimately reported the incident to the CHP.\nCHP Officer John Wilson responded and interviewed Ayers.\nOfficer Wilson subsequently detained Honan along the\n\nclaims of state law immunity raised by Ayers on appeal in a\nmemorandum disposition filed concurrently with this opinion.\nSimilarly, we resolve Officer Wilson’s and the CHP’s separate appeal,\nNo. 17-16669, in a separately filed memorandum disposition.\n\f6 ADVANCED BLDG. & FABRICATION V. AYERS\n\nhighway, searched his truck, and interviewed him. Honan\noffered to discuss the incident and invited Officer Wilson\nand his partner back to Advanced Building. Honan gave\nthem a tour of Advanced Building and “showed them some\nof the things Advanced was working on, including [an]\nammunition reloader,” and “a couple gold nuggets from [his]\ncollection.” The parties also dispute whether, during the\ntour, Honan claimed to Officer Wilson that he had\nsurveillance video that would corroborate his version of the\nincident with Ayers.\n\n Officer Wilson subsequently obtained a search warrant\nauthorizing “any sheriff, marshal or police officer in the\ncounty of Butte” to lawfully search the premises of\nAdvanced Building. To obtain the warrant, Officer Wilson\nrecounted in an affidavit the events as described to him by\nboth Ayers and Honan, and indicated that there was probable\ncause to believe that Honan committed felony threats and\nvandalism against Ayers. The warrant issued for officers to\nsearch for and seize (1) items “capable of storing . . . video\nand audio media,” (2) “indicia proving [Honan’s] ownership\nand control” of the business, and (3) “evidence of\ncorrespondence” that would corroborate the alleged felonies.\nOfficer Wilson indicated to Sims that “he wanted us, me and\nCurtis [Ayers] to . . . complete the inspection” during the\nexecution of the search warrant because it was “important\nthat [the BOE] complete [their] inspection and not allow this\nbusiness or owner to get away with anything.” Sims\naccepted the invitation, and told Ayers that he would like the\ntwo of them to go during the search.\n\n On May 30, 2012, the CHP executed the search warrant\nat Advanced Building. Sims and Ayers waited in the car\nuntil the CHP confirmed they could enter. They “did not\nassist the CHP in their search efforts nor were [they] asked\n\f ADVANCED BLDG. & FABRICATION V. AYERS 7\n\nto do so,” though Ayers did at one point identify Honan and\nshow Officer Wilson where the assault occurred. Honan\nasserts that he saw Ayers searching through a file cabinet of\npersonal records. Ayers acknowledged that he examined\nany “sales permits or records [they] saw,” but denied taking\nor damaging any property. Honan alleges that a substantial\namount of gold and silver was taken that day, as well as a\nnumber of additional items and personal effects.\n\n Honan and Advanced Building collectively filed a\ncomplaint against the CHP, Officer Wilson, Ayers, and other\nunnamed individuals, alleging state-law claims for\nconversion and other tort claims, as well as a federal claim\nunder 42 U.S.C. § 1983. Ayers moved for summary\njudgment, including on the grounds of federal- and state-law\nimmunity. The district court denied summary judgment due\nto “overwhelming” material factual disputes, and Ayers and\nthe other defendants timely appealed.\n\n II.\n\n JURISDICTION AND STANDARD OF REVIEW\n\n We have jurisdiction under 28 U.S.C. § 1291 to review\nan order denying summary judgment on the grounds of\nqualified immunity. Plumhoff v. Rickard, 572 U.S. 765,\n771–73 (2014). However, “the scope of our review over the\nappeal [in this context] is circumscribed.” George v. Morris,\n736 F.3d 829, 834 (9th Cir. 2013). We may not consider “a\nfact-related dispute about the pretrial record,” that is,\n“whether or not the evidence in the pretrial record was\nsufficient to show a genuine issue of fact for trial.” Johnson\nv. Jones, 515 U.S. 304, 307 (1995). Accordingly, we have\njurisdiction only where “the issue appealed concerned, not\nwhich facts the parties might be able to prove, but, rather,\nwhether or not certain given facts showed” that immunity\n\f8 ADVANCED BLDG. & FABRICATION V. AYERS\n\nwould apply. See id. at 311 (quoting Mitchell v. Forsyth,\n472 U.S. 511, 528 (1985)).\n\n Here, although Plaintiffs’ claim under § 1983 alleges a\nnumber of constitutional violations by Defendants, we\nconsider just one issue: whether Ayers’s participation in the\nMay 2012 search at Advanced Building violated clearly\nestablished law. Ayers’s claim of qualified immunity does\nnot turn on the resolution of any factual dispute, and thus we\nhave jurisdiction to review the district court’s denial of\nqualified immunity. 2\n\n III.\n\n DISCUSSION\n\n To survive a claim of qualified immunity, the plaintiff\nmust show: “(1) that the right was violated; and (2) that the\nright was clearly established at the time of the alleged\nmisconduct.” Isayeva v. Sacramento Sheriff’s Dep’t,\n872 F.3d 938, 946 (9th Cir. 2017).\n\n A.\n\n Ayers argues that he did not violate Plaintiffs’ Fourth\nAmendment rights because he claims California law permits\na BOE employee to request inspection of sales records from\nany business, and his participation in the search would thus\nqualify under the “administrative search” exception to the\n\n 2\n Ayers suggested at oral argument that “it’s not in the record here,\n[but] one of the reasons he was asked to come along [was] to show where\nthe altercation occurred.” Indeed, the record is silent as to this purported\nrationale, but instead confirms that the purpose of inviting Ayers and\nSims was for them to “complete the inspection” of Advanced Building’s\nbusiness records.\n\f ADVANCED BLDG. & FABRICATION V. AYERS 9\n\nwarrant requirement. See, e.g., Cal. Rev. & Tax Code\n§§ 7053–54; 18 Cal. Code Regs. § 1698(b)(1)). Ayers’s\nreliance on these state provisions is a nonstarter because\nnone of the state statutes or regulations on which he relies\nauthorizes forcible entry or searches. For example, 18 Cal.\nCode Regs. section 1698(b)(1) merely mandates a taxpayer\nto “make available for examination . . . all records necessary\nto determine the correct tax liability.” As Honan correctly\npoints out, section 1698.5 (which immediately follows the\nsection cited by Ayers) details the audit procedures and sets\nforth a process for conferring and providing requested\ndocumentation. Id. § 1698.5. The regulation even goes so\nfar as to impose a duty on BOE staff to “not directly access\nthe taxpayer’s computer system if the taxpayer objects to\nsuch access, except in the case of a search warrant.” Id.\n§ 1698.5(b)(4)(D). Nowhere in the regulation is there broad\nauthorization to conduct a warrantless search and seizure of\nbusiness records without first requesting them from the\nbusiness owner. Ayers acknowledges that, consistent with\nthese regulations, BOE protocol if an inspector is asked to\nleave would be to “see if we can call back at a better time.”\nTherefore, we conclude that Ayers’s conduct is not\nauthorized by any state statute or regulation.\n\n Even assuming that state law permits warrantless\ninspection of business records, the intrusive search here\nwould not withstand scrutiny under the Fourth Amendment.\nAyers argues that “[a]dministrative searches of a\ncommercial location pursuant to statute can present an\nexception to the general warrant requirement.” However, his\nreading of precedent is too broad to support application of\nsuch an exception to the warrant requirement here. For\nexample, in New York v. Burger, the Supreme Court\nconsidered the diminished privacy expectations for\n“pervasively regulated business[es],” upholding warrantless\n\f10 ADVANCED BLDG. & FABRICATION V. AYERS\n\ninspections only where “necessary to further [the] regulatory\nscheme.” 482 U.S. 691, 702 (1987) (alteration in original)\n(quoting Donovan v. Dewey, 452 U.S. 594, 600 (1981)).\nSimilarly, in United States v. Biswell, the Court upheld a\nwarrantless inspection of a federally licensed weapons\ndealer because “close scrutiny of [firearm] traffic is\nundeniably of central importance to federal efforts to prevent\nviolent crime.” 406 U.S. 311, 315 (1972).\n\n Here, no analogously pervasive regulation or special\ngovernmental interest justifies a diminished expectation of\nprivacy. Indeed, should Ayers be correct, it would seem that\nevery business in California “selling tangible personal\nproperty” would be subject to a warrantless search at any\ntime. See Cal. Rev. & Tax Code § 7054. We conclude that\nthe “administrative search” exception does not apply to\nAyers’s presence at the search.\n\n B.\n\n We next consider whether Ayers could nevertheless\nparticipate in the search at the invitation of the CHP. In\nWilson v. Layne, the Supreme Court held that “it is a\nviolation of the Fourth Amendment for police to bring . . .\nthird parties into a home during the execution of a warrant\nwhen the presence of the third parties in the home was not in\naid of the execution of the warrant.” 526 U.S. 603, 614\n(1999). There, police invited reporters on a “media ride-\nalong” to execute arrest warrants at a private residence. Id.\nat 605, 607. The Court held that the reporters’ presence\nconstituted a violation of the plaintiff’s Fourth Amendment\nrights because their presence “inside the home was not\nrelated to the objectives of the authorized intrusion.” Id.\nat 611. The Court distinguished situations where the third\nparties “directly aided in the execution of the warrant,” such\nas identifying stolen property that was the subject of the\n\f ADVANCED BLDG. & FABRICATION V. AYERS 11\n\nwarrant, conduct that “has long been approved by [the] Court\nand our common-law tradition.” Id. at 611–12. By contrast,\nthe Court rejected the presence of third parties serving only\nto “further the law enforcement objectives of the police in a\ngeneral sense,” rather than the specific “purposes of the\nsearch.” Id. at 612 (emphasis added). The Court clarified\nthat, in those situations, even if the police are lawfully\npresent, “the violation of the Fourth Amendment is the\npresence of the [third party].” Id. at 614 n.2.\n\n Under Wilson, Ayers’s participation in the search\nviolated Plaintiffs’ Fourth Amendment rights. The warrant\nissued for officers to search specifically for (1) items\n“capable of storing . . . video and audio media,” (2) “indicia\nproving [Honan’s] ownership and control” of the business,\nand (3) “evidence of correspondence” that would\ncorroborate the alleged felonies against Ayers. Ayers\nacknowledged that he “attend[ed] the search to complete the\ninspection” he had begun of Advanced Building’s business\nlicensing and tax records. Such an inspection was obviously\nnot “in aid of the execution of the warrant” to seize the above\nevidence relating to the allegations of criminal threats and\nvandalism. See Wilson, 526 U.S. at 614. Thus, because\nAyers’s presence was not “related to the objectives of the\nauthorized intrusion,” see id. at 611, his conduct violated\nPlaintiffs’ Fourth Amendment rights.\n\n Moreover, this right was “clearly established” following\nthe Court’s opinion in Wilson in 1999. While the Supreme\nCourt has cautioned against “defin[ing] clearly established\nlaw at a high level of generality,” see Ashcroft v. al-Kidd,\n563 U.S. 731, 742 (2011), Wilson held explicitly that officers\nmay not simply “bring members of the media or other third\nparties . . . during the execution of a warrant” unless it was\n“in aid of the warrant’s execution.” 526 U.S. at 614\n\f12 ADVANCED BLDG. & FABRICATION V. AYERS\n\n(emphasis added). Ayers argues that Wilson is not\nsufficiently specific because that case involved reporters, not\nother government agents. He contends that his position as a\ngovernment employee—one charged with inspecting\nbusiness records—distinguishes him from the journalists at\nissue in Wilson.\n\n In fact, his liability under § 1983 is premised on this very\ndistinction: private citizens (like reporters) are not ordinarily\nliable under § 1983 for their presence during the execution\nof a search warrant because they are not state actors. See\nBrunette v. Humane Soc’y of Ventura Cty., 294 F.3d 1205,\n1211 (9th Cir. 2002). Wilson did not rest constitutionality\non the third party’s employer, but rather drew a bright line at\nwhether their presence assisted execution of the warrant.\nThe “contours of the right” here were “sufficiently clear that\na reasonable official would understand that what he is doing\nviolates that right.” See Wilson, 526 U.S. at 615 (quoting\nAnderson v. Creighton, 483 U.S. 635, 640 (1987)). And,\nalthough Ayers claims that he was “acting at the direction of\nhis supervisor,” he did not cite to any binding precedent\nholding that a supervisor’s instruction would somehow\nobviate a clear constitutional violation. Accordingly, we\nfind that Ayers’s conduct violated Plaintiffs’ clearly\nestablished rights under Wilson.\n\n IV.\n\n CONCLUSION\n\n Because the administrative exception to the warrant\nrequirement does not apply, and Ayers’s presence was not in\naid of the execution of the search warrant, his presence and\nsearch violated Plaintiffs’ clearly established constitutional\nrights. We affirm the district court’s denial of qualified\n\f ADVANCED BLDG. & FABRICATION V. AYERS 13\n\nimmunity and remand for proceedings consistent with this\nopinion.\n\n AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Advanced Building & Fabricatio v. Curtis Ayers\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-13\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nADVANCED BUILDING & No. 17-16618\\nFABRICATION, INC., a California\\ncorporation; ROBERT HONAN, D.C. No.\\n Plaintiffs-Appellees, 2:13-cv-02380-\\n MCE-CKD\\n v.\\n\\nCALIFORNIA HIGHWAY PATROL; OPINION\\nJOHN WILSON,\\n Defendants,\\n\\n and\\n\\nCURTIS AYERS,\\n Defendant-Appellant.\\n\\n\\n\\n Appeal from the United States District Court\\n for the Eastern District of California\\n Morrison C. England, Jr., District Judge, Presiding\\n\\n Argued and Submitted December 19, 2018\\n San Francisco, California\\n\\n Filed March 13, 2019\\n\\f2 ADVANCED BLDG. & FABRICATION V. AYERS\\n\\n Before: Milan D. Smith, Jr. and Jacqueline H. Nguyen,\\n Circuit Judges, and Jane A. Restani, * Judge.\\n\\n Opinion by Judge Nguyen\\n\\n\\n SUMMARY **\\n\\n\\n Civil Rights\\n\\n The panel affirmed the district court’s denial, on\\nsummary judgment, of qualified immunity to Curtis Ayers,\\na former employee of the California State Board of\\nEqualization, in an action brought against him by plaintiff\\nRobert Honan and his business alleging Ayers violated\\nclearly established law by participating in a search of\\nplaintiff’s business following an altercation between Ayers\\nand plaintiff.\\n\\n Ayers, who had mistaken plaintiff’s business for another\\nbusiness, arrived at plaintiff’s premises without an\\nappointment, and identified himself as a Board of\\nEqualization employee, but did not present his ID badge. A\\nverbal altercation ensued and Ayers alleged that as he was\\nleaving, plaintiff tackled him and pushed him through the\\ndoor, resulting in Ayers dropping and damaging his laptop.\\nAyers reported the incident and the California Highway\\nPatrol subsequently obtained a search warrant to search the\\n\\n *\\n The Honorable Jane A. Restani, Judge for the United States Court\\nof International Trade, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f ADVANCED BLDG. & FABRICATION V. AYERS 3\\n\\npremises of plaintiff’s business on the grounds that plaintiff\\ncommitted felony threats and vandalism against Ayers.\\nCalifornia Highway Patrol executed the warrant and Ayers\\nparticipated in the search, during which he allegedly\\nsearched through plaintiff’s personal property.\\n\\n Plaintiff sued Ayers and other defendants, including\\nCalifornia Highway Patrol officers, alleging claims for state-\\nlaw conversion and other tort claims, as well as a federal\\nclaim under 42 U.S.C. § 1983. The panel affirmed the denial\\nof qualified immunity to Ayers and resolved the remaining\\nclaims against the other defendants in concurrently filed\\nmemorandum dispositions.\\n\\n The panel rejected Ayers’ contention that there was no\\nFourth Amendment violation because California law permits\\nState Board of Equalization employees to request inspection\\nof sales records from any business, and that therefore his\\nparticipation in the search would qualify under the\\n“administrative search” exception to the warrant\\nrequirement. The panel held that none of the state provisions\\ncited by Ayers authorized forcible entry or searches.\\nMoreover, even assuming that state law permitted\\nwarrantless inspections of business records, the intrusive\\nsearch here would not withstand scrutiny under the Fourth\\nAmendment. The panel further held that Ayers’s presence\\nwas not necessary to aid in the officers’ execution of the\\nwarrant, which had the purpose of seizing evidence relating\\nto allegations of criminal threats and vandalism. Because\\nAyers’s presence was not related to the objective of the\\nsearch warrant, his conduct violated plaintiff’s Fourth\\nAmendment rights. The panel remanded for further\\nproceedings consistent with the panel’s opinion.\\n\\f4 ADVANCED BLDG. & FABRICATION V. AYERS\\n\\n COUNSEL\\n\\nMichael von Loewenfeldt (argued) and Daniel J. Veroff,\\nKerr & Wagstaffe LLP, San Francisco, California, for\\nDefendant-Appellant.\\n\\nPolly J. Estes (argued), Estes Law Group, San Rafael,\\nCalifornia, for Plaintiffs-Appellees.\\n\\n\\n OPINION\\n\\nNGUYEN, Circuit Judge:\\n\\n Curtis Ayers, a former employee of the California State\\nBoard of Equalization (“BOE”), appeals the district court’s\\ndenial of qualified immunity in a suit brought against him by\\nPlaintiffs Robert Honan and his business, Advanced\\nBuilding & Fabrication, Inc. (“Advanced Building”). The\\nallegations in the complaint stem from an altercation\\nbetween Ayers and Honan, which led to the execution of a\\nsearch warrant at Advanced Building by officers of the\\nCalifornia Highway Patrol (“CHP”). Honan contends that\\nAyers violated clearly established law by participating in the\\nsearch. We agree. Because the administrative search\\nexception does not apply and Ayers’s presence was not\\nnecessary to aid in the officers’ execution of the warrant, we\\naffirm the district court’s denial of qualified immunity. 1\\n\\n\\n 1\\n Honan sued numerous other defendants in addition to Ayers,\\nincluding CHP Officer John Wilson and the CHP itself. Citing numerous\\nfactual disputes, the district court denied summary judgment and\\nimmunity to all the defendants. We address only Ayers’s appeal of the\\ndenial of qualified immunity in this opinion, and resolve the remaining\\n\\f ADVANCED BLDG. & FABRICATION V. AYERS 5\\n\\n I.\\n\\n BACKGROUND\\n\\n Robert Honan owns and operates Advanced Building, a\\nmetal fabrication business that makes metal machines and\\nspecialized metal parts. On May 7, 2012, Curtis Ayers, who\\nat that time was employed by the BOE, arrived at Advanced\\nBuilding without a prior appointment. Ayers identified\\nhimself as a BOE employee, but he did not present his ID\\nbadge and had mistaken Advanced Building for another\\nbusiness. Honan suspected Ayers of a “fraudulent scam”\\nand, after a brief exchange of words, Honan told Ayers to\\n“Get the F* out.”\\n\\n The parties dispute what happened next, but, as Ayers\\nwas leaving (or being thrown out of) the building, his laptop\\ndropped and sustained damaged. According to Honan,\\nAyers must have heard Honan approaching and “became\\nstartled and stumbled out the exit door, falling to the\\nground.” Ayers, on the other hand, claimed that Honan\\ntackled him and pushed him through the door as he was\\nleaving. In the process, Ayers alleges, the laptop flew out of\\nhis hands, the monitor broke, and the laptop was rendered\\ninoperable.\\n\\n Ayers contacted his supervisor, Dwayne Sims to report\\nthe incident. Sims recommended that Ayers contact the\\npolice, and they ultimately reported the incident to the CHP.\\nCHP Officer John Wilson responded and interviewed Ayers.\\nOfficer Wilson subsequently detained Honan along the\\n\\nclaims of state law immunity raised by Ayers on appeal in a\\nmemorandum disposition filed concurrently with this opinion.\\nSimilarly, we resolve Officer Wilson’s and the CHP’s separate appeal,\\nNo. 17-16669, in a separately filed memorandum disposition.\\n\\f6 ADVANCED BLDG. & FABRICATION V. AYERS\\n\\nhighway, searched his truck, and interviewed him. Honan\\noffered to discuss the incident and invited Officer Wilson\\nand his partner back to Advanced Building. Honan gave\\nthem a tour of Advanced Building and “showed them some\\nof the things Advanced was working on, including [an]\\nammunition reloader,” and “a couple gold nuggets from [his]\\ncollection.” The parties also dispute whether, during the\\ntour, Honan claimed to Officer Wilson that he had\\nsurveillance video that would corroborate his version of the\\nincident with Ayers.\\n\\n Officer Wilson subsequently obtained a search warrant\\nauthorizing “any sheriff, marshal or police officer in the\\ncounty of Butte” to lawfully search the premises of\\nAdvanced Building. To obtain the warrant, Officer Wilson\\nrecounted in an affidavit the events as described to him by\\nboth Ayers and Honan, and indicated that there was probable\\ncause to believe that Honan committed felony threats and\\nvandalism against Ayers. The warrant issued for officers to\\nsearch for and seize (1) items “capable of storing . . . video\\nand audio media,” (2) “indicia proving [Honan’s] ownership\\nand control” of the business, and (3) “evidence of\\ncorrespondence” that would corroborate the alleged felonies.\\nOfficer Wilson indicated to Sims that “he wanted us, me and\\nCurtis [Ayers] to . . . complete the inspection” during the\\nexecution of the search warrant because it was “important\\nthat [the BOE] complete [their] inspection and not allow this\\nbusiness or owner to get away with anything.” Sims\\naccepted the invitation, and told Ayers that he would like the\\ntwo of them to go during the search.\\n\\n On May 30, 2012, the CHP executed the search warrant\\nat Advanced Building. Sims and Ayers waited in the car\\nuntil the CHP confirmed they could enter. They “did not\\nassist the CHP in their search efforts nor were [they] asked\\n\\f ADVANCED BLDG. & FABRICATION V. AYERS 7\\n\\nto do so,” though Ayers did at one point identify Honan and\\nshow Officer Wilson where the assault occurred. Honan\\nasserts that he saw Ayers searching through a file cabinet of\\npersonal records. Ayers acknowledged that he examined\\nany “sales permits or records [they] saw,” but denied taking\\nor damaging any property. Honan alleges that a substantial\\namount of gold and silver was taken that day, as well as a\\nnumber of additional items and personal effects.\\n\\n Honan and Advanced Building collectively filed a\\ncomplaint against the CHP, Officer Wilson, Ayers, and other\\nunnamed individuals, alleging state-law claims for\\nconversion and other tort claims, as well as a federal claim\\nunder 42 U.S.C. § 1983. Ayers moved for summary\\njudgment, including on the grounds of federal- and state-law\\nimmunity. The district court denied summary judgment due\\nto “overwhelming” material factual disputes, and Ayers and\\nthe other defendants timely appealed.\\n\\n II.\\n\\n JURISDICTION AND STANDARD OF REVIEW\\n\\n We have jurisdiction under 28 U.S.C. § 1291 to review\\nan order denying summary judgment on the grounds of\\nqualified immunity. Plumhoff v. Rickard, 572 U.S. 765,\\n771–73 (2014). However, “the scope of our review over the\\nappeal [in this context] is circumscribed.” George v. Morris,\\n736 F.3d 829, 834 (9th Cir. 2013). We may not consider “a\\nfact-related dispute about the pretrial record,” that is,\\n“whether or not the evidence in the pretrial record was\\nsufficient to show a genuine issue of fact for trial.” Johnson\\nv. Jones, 515 U.S. 304, 307 (1995). Accordingly, we have\\njurisdiction only where “the issue appealed concerned, not\\nwhich facts the parties might be able to prove, but, rather,\\nwhether or not certain given facts showed” that immunity\\n\\f8 ADVANCED BLDG. & FABRICATION V. AYERS\\n\\nwould apply. See id. at 311 (quoting Mitchell v. Forsyth,\\n472 U.S. 511, 528 (1985)).\\n\\n Here, although Plaintiffs’ claim under § 1983 alleges a\\nnumber of constitutional violations by Defendants, we\\nconsider just one issue: whether Ayers’s participation in the\\nMay 2012 search at Advanced Building violated clearly\\nestablished law. Ayers’s claim of qualified immunity does\\nnot turn on the resolution of any factual dispute, and thus we\\nhave jurisdiction to review the district court’s denial of\\nqualified immunity. 2\\n\\n III.\\n\\n DISCUSSION\\n\\n To survive a claim of qualified immunity, the plaintiff\\nmust show: “(1) that the right was violated; and (2) that the\\nright was clearly established at the time of the alleged\\nmisconduct.” Isayeva v. Sacramento Sheriff’s Dep’t,\\n872 F.3d 938, 946 (9th Cir. 2017).\\n\\n A.\\n\\n Ayers argues that he did not violate Plaintiffs’ Fourth\\nAmendment rights because he claims California law permits\\na BOE employee to request inspection of sales records from\\nany business, and his participation in the search would thus\\nqualify under the “administrative search” exception to the\\n\\n 2\\n Ayers suggested at oral argument that “it’s not in the record here,\\n[but] one of the reasons he was asked to come along [was] to show where\\nthe altercation occurred.” Indeed, the record is silent as to this purported\\nrationale, but instead confirms that the purpose of inviting Ayers and\\nSims was for them to “complete the inspection” of Advanced Building’s\\nbusiness records.\\n\\f ADVANCED BLDG. & FABRICATION V. AYERS 9\\n\\nwarrant requirement. See, e.g., Cal. Rev. & Tax Code\\n§§ 7053–54; 18 Cal. Code Regs. § 1698(b)(1)). Ayers’s\\nreliance on these state provisions is a nonstarter because\\nnone of the state statutes or regulations on which he relies\\nauthorizes forcible entry or searches. For example, 18 Cal.\\nCode Regs. section 1698(b)(1) merely mandates a taxpayer\\nto “make available for examination . . . all records necessary\\nto determine the correct tax liability.” As Honan correctly\\npoints out, section 1698.5 (which immediately follows the\\nsection cited by Ayers) details the audit procedures and sets\\nforth a process for conferring and providing requested\\ndocumentation. Id. § 1698.5. The regulation even goes so\\nfar as to impose a duty on BOE staff to “not directly access\\nthe taxpayer’s computer system if the taxpayer objects to\\nsuch access, except in the case of a search warrant.” Id.\\n§ 1698.5(b)(4)(D). Nowhere in the regulation is there broad\\nauthorization to conduct a warrantless search and seizure of\\nbusiness records without first requesting them from the\\nbusiness owner. Ayers acknowledges that, consistent with\\nthese regulations, BOE protocol if an inspector is asked to\\nleave would be to “see if we can call back at a better time.”\\nTherefore, we conclude that Ayers’s conduct is not\\nauthorized by any state statute or regulation.\\n\\n Even assuming that state law permits warrantless\\ninspection of business records, the intrusive search here\\nwould not withstand scrutiny under the Fourth Amendment.\\nAyers argues that “[a]dministrative searches of a\\ncommercial location pursuant to statute can present an\\nexception to the general warrant requirement.” However, his\\nreading of precedent is too broad to support application of\\nsuch an exception to the warrant requirement here. For\\nexample, in New York v. Burger, the Supreme Court\\nconsidered the diminished privacy expectations for\\n“pervasively regulated business[es],” upholding warrantless\\n\\f10 ADVANCED BLDG. & FABRICATION V. AYERS\\n\\ninspections only where “necessary to further [the] regulatory\\nscheme.” 482 U.S. 691, 702 (1987) (alteration in original)\\n(quoting Donovan v. Dewey, 452 U.S. 594, 600 (1981)).\\nSimilarly, in United States v. Biswell, the Court upheld a\\nwarrantless inspection of a federally licensed weapons\\ndealer because “close scrutiny of [firearm] traffic is\\nundeniably of central importance to federal efforts to prevent\\nviolent crime.” 406 U.S. 311, 315 (1972).\\n\\n Here, no analogously pervasive regulation or special\\ngovernmental interest justifies a diminished expectation of\\nprivacy. Indeed, should Ayers be correct, it would seem that\\nevery business in California “selling tangible personal\\nproperty” would be subject to a warrantless search at any\\ntime. See Cal. Rev. & Tax Code § 7054. We conclude that\\nthe “administrative search” exception does not apply to\\nAyers’s presence at the search.\\n\\n B.\\n\\n We next consider whether Ayers could nevertheless\\nparticipate in the search at the invitation of the CHP. In\\nWilson v. Layne, the Supreme Court held that “it is a\\nviolation of the Fourth Amendment for police to bring . . .\\nthird parties into a home during the execution of a warrant\\nwhen the presence of the third parties in the home was not in\\naid of the execution of the warrant.” 526 U.S. 603, 614\\n(1999). There, police invited reporters on a “media ride-\\nalong” to execute arrest warrants at a private residence. Id.\\nat 605, 607. The Court held that the reporters’ presence\\nconstituted a violation of the plaintiff’s Fourth Amendment\\nrights because their presence “inside the home was not\\nrelated to the objectives of the authorized intrusion.” Id.\\nat 611. The Court distinguished situations where the third\\nparties “directly aided in the execution of the warrant,” such\\nas identifying stolen property that was the subject of the\\n\\f ADVANCED BLDG. & FABRICATION V. AYERS 11\\n\\nwarrant, conduct that “has long been approved by [the] Court\\nand our common-law tradition.” Id. at 611–12. By contrast,\\nthe Court rejected the presence of third parties serving only\\nto “further the law enforcement objectives of the police in a\\ngeneral sense,” rather than the specific “purposes of the\\nsearch.” Id. at 612 (emphasis added). The Court clarified\\nthat, in those situations, even if the police are lawfully\\npresent, “the violation of the Fourth Amendment is the\\npresence of the [third party].” Id. at 614 n.2.\\n\\n Under Wilson, Ayers’s participation in the search\\nviolated Plaintiffs’ Fourth Amendment rights. The warrant\\nissued for officers to search specifically for (1) items\\n“capable of storing . . . video and audio media,” (2) “indicia\\nproving [Honan’s] ownership and control” of the business,\\nand (3) “evidence of correspondence” that would\\ncorroborate the alleged felonies against Ayers. Ayers\\nacknowledged that he “attend[ed] the search to complete the\\ninspection” he had begun of Advanced Building’s business\\nlicensing and tax records. Such an inspection was obviously\\nnot “in aid of the execution of the warrant” to seize the above\\nevidence relating to the allegations of criminal threats and\\nvandalism. See Wilson, 526 U.S. at 614. Thus, because\\nAyers’s presence was not “related to the objectives of the\\nauthorized intrusion,” see id. at 611, his conduct violated\\nPlaintiffs’ Fourth Amendment rights.\\n\\n Moreover, this right was “clearly established” following\\nthe Court’s opinion in Wilson in 1999. While the Supreme\\nCourt has cautioned against “defin[ing] clearly established\\nlaw at a high level of generality,” see Ashcroft v. al-Kidd,\\n563 U.S. 731, 742 (2011), Wilson held explicitly that officers\\nmay not simply “bring members of the media or other third\\nparties . . . during the execution of a warrant” unless it was\\n“in aid of the warrant’s execution.” 526 U.S. at 614\\n\\f12 ADVANCED BLDG. & FABRICATION V. AYERS\\n\\n(emphasis added). Ayers argues that Wilson is not\\nsufficiently specific because that case involved reporters, not\\nother government agents. He contends that his position as a\\ngovernment employee—one charged with inspecting\\nbusiness records—distinguishes him from the journalists at\\nissue in Wilson.\\n\\n In fact, his liability under § 1983 is premised on this very\\ndistinction: private citizens (like reporters) are not ordinarily\\nliable under § 1983 for their presence during the execution\\nof a search warrant because they are not state actors. See\\nBrunette v. Humane Soc’y of Ventura Cty., 294 F.3d 1205,\\n1211 (9th Cir. 2002). Wilson did not rest constitutionality\\non the third party’s employer, but rather drew a bright line at\\nwhether their presence assisted execution of the warrant.\\nThe “contours of the right” here were “sufficiently clear that\\na reasonable official would understand that what he is doing\\nviolates that right.” See Wilson, 526 U.S. at 615 (quoting\\nAnderson v. Creighton, 483 U.S. 635, 640 (1987)). And,\\nalthough Ayers claims that he was “acting at the direction of\\nhis supervisor,” he did not cite to any binding precedent\\nholding that a supervisor’s instruction would somehow\\nobviate a clear constitutional violation. Accordingly, we\\nfind that Ayers’s conduct violated Plaintiffs’ clearly\\nestablished rights under Wilson.\\n\\n IV.\\n\\n CONCLUSION\\n\\n Because the administrative exception to the warrant\\nrequirement does not apply, and Ayers’s presence was not in\\naid of the execution of the search warrant, his presence and\\nsearch violated Plaintiffs’ clearly established constitutional\\nrights. We affirm the district court’s denial of qualified\\n\\f ADVANCED BLDG. & FABRICATION V. AYERS 13\\n\\nimmunity and remand for proceedings consistent with this\\nopinion.\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1287,"cells":{"id":{"kind":"number","value":4603420,"string":"4,603,420"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"homeaway.com, Inc. v. City of Santa Monica"},"decision_date":{"kind":"string","value":"2019-03-13"},"docket_number":{"kind":"string","value":"18-55367"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Mary M. Schroeder and Jacqueline H. Nguyen, Circuit Judges, and Michael H. Simon, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"NGUYEN, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nHOMEAWAY.COM, INC.; AIRBNB No. 18-55367\\nINC.,\\n Plaintiffs-Appellants, D.C. Nos.\\n 2:16-cv-06641-\\n v. ODW-AFM\\n 2:16-cv-06645-\\nCITY OF SANTA MONICA, ODW-AFM\\n Defendant-Appellee.\\n\\n\\n\\nHOMEAWAY.COM, INC., No. 18-55805\\n Plaintiff-Appellant,\\n D.C. Nos.\\n and 2:16-cv-06641-\\n ODW-AFM\\nAIRBNB INC., 2:16-cv-06645-\\n Plaintiff, ODW-AFM\\n\\n v.\\n\\nCITY OF SANTA MONICA,\\n Defendant-Appellee.\\n\\f2 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\n\\n AIRBNB INC., No. 18-55806\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:16-cv-06645-\\n ODW-AFM\\n CITY OF SANTA MONICA,\\n Defendant-Appellee. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Otis D. Wright II, District Judge, Presiding\\n\\n Argued and Submitted October 12, 2018\\n Pasadena, California\\n\\n Filed March 13, 2019\\n\\n Before: Mary M. Schroeder and Jacqueline H. Nguyen,\\n Circuit Judges, and Michael H. Simon, * District Judge.\\n\\n Opinion by Judge Nguyen\\n\\n\\n\\n\\n *\\n The Honorable Michael H. Simon, United States District Judge for\\nthe District of Oregon, sitting by designation.\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 3\\n\\n SUMMARY **\\n\\n\\n Civil Rights\\n\\n The panel affirmed the district court’s dismissal, for\\nfailure to state a claim, of a complaint brought by\\nHomeAway.com and Airbnb Inc. challenging the City of\\nSanta Monica’s Ordinance 2535, which imposes various\\nobligations on companies that host online platforms for\\nshort-term vacation rentals.\\n\\n The Ordinance, as amended in 2017, imposes four\\nobligations on hosting platforms: (1) collecting and\\nremitting Transient Occupancy Taxes; (2) regularly\\ndisclosing listings and booking information to the City; (3)\\nrefraining from booking properties not licensed and listed on\\nthe City’s registry; (4) and refraining from collecting a fee\\nfor ancillary services.\\n\\n The panel rejected plaintiffs’ assertion that the\\nOrdinance violated the Communications Decency Act of\\n1996, 47 U.S.C. § 230 because it required them to monitor\\nand remove third-party content, and therefore interfered with\\nfederal policy protecting internet companies from liability\\nfor posting third-party content. The panel stated that the\\nOrdinance prohibits processing transactions for unregistered\\nproperties. It does not require the Platforms to review the\\ncontent provided by the hosts of listings on their websites.\\nRather, the panel noted that the only monitoring that\\nappeared necessary in order to comply with the Ordinance\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f4 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nrelated to incoming requests to complete a booking\\ntransaction—content that, while resulting from the third-\\nparty listings, was distinct, internal, and nonpublic. The\\npanel concluded that the Ordinance was not inconsistent\\nwith the Communications Decency Act, and therefore was\\nnot expressly preempted by its terms. The panel further\\nconcluded that the Ordinance would not pose an obstacle to\\nCongress’s aim to encourage self-monitoring of third-party\\ncontent, and therefore obstacle preemption did not preclude\\nSanta Monica from enforcing the Ordinance.\\n\\n The panel held that the Ordinance did not implicate\\nspeech protected by the First Amendment, concluding that\\nthe Ordinance’s prohibitions regulate nonexpressive\\nconduct, specifically booking transactions, and do not single\\nout those engaged in expressive activity.\\n\\n\\n COUNSEL\\n\\nDonald B. Verrilli, Jr. (argued) and Chad Golder, Munger,\\nTolles & Olson LLP, Washington, D.C.; Joseph W. Cotchett\\nand Alexandra P. Summer, Cotchett, Pitre & McCarthy,\\nLLP, Santa Monica, California; Jonathan H. Blavin and\\nJoshua Patashnik, Munger, Tolles & Olson LLP, San\\nFrancisco, California; John B. Major, Munger, Tolles &\\nOlson, Los Angeles, California; for Plaintiff-Appellant\\nAirbnb, Inc.\\n\\nStephen M. Rummage and Ambika K. Doran, Davis Wright\\nTremaine LLP, Seattle, Washington, for Plaintiff-Appellant\\nHomeAway.com, Inc.\\n\\nGeorge S. Cardona (argued), Deputy City Attorney; Lane\\nDilg, City Attorney; Yibin Shen, Chief Deputy City\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 5\\n\\nAttorney; Heidi Von Tongeln and Michael R. Cobden,\\nDeputy City Attorneys; Santa Monica City Attorney’s\\nOffice, Santa Monica, California, for Defendant-Appellee.\\n\\nDavid Salmons, Bryan Killian, and James Nelson, Morgan,\\nLewis & Bockius LLP, Washington, D.C., for Amici Curiae\\nChris Cox and NetChoice.\\n\\nCatherine R. Gellis, Esq., Sausalito, California, for Amicus\\nCuriae Floor64, Inc. d/b/a The Copia Institute.\\n\\nIan C. Ballon and Lori Chang, Greenberg Traurig, LLP, Los\\nAngeles, California, for Amici Curiae Ebay Inc., Glassdoor,\\nInc., Lyft, Inc., Offerup, Inc., Taskrabbit, Inc., Thumbtack,\\nInc., Uber Technologies, Inc., and Upwork, Inc.\\n\\nChristi Hogin, Best Best & Krieger LLP, Manhattan Beach,\\nCalifornia, for Amici Curiae League of California Cities,\\nInternational Municipal Lawyers Association, and\\nCalifornia State Association of Counties.\\n\\nJeremy B. Rosen, Eric S. Boorstin, and Ryan C. Chapman,\\nHorvitz & Levy LLP, Burbank, California; Michael T.\\nWilliams and Allison R. Mclaughlin, Wheeler Trigg\\nO’Donnell LLP, Denver, Colorado; David C. Frederick,\\nBrendan J. Crimmins, and Rachel P. May, Kellogg, Hansen,\\nTodd, Figel & Frederick, P.L.L.C., Washington, D.C., for\\nAmicus Curiae Apartment Investment and Management\\nCompany.\\n\\nHeidi Palutke, Sacramento, California, for Amicus Curiae\\nCalifornia Apartment Association.\\n\\nEdward M. Schulman, Ballston Tower, Arlington, Virginia,\\nfor Amicus Curiae AvalonBay Communities, Inc.\\n\\f6 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nGary S. Kessler, Kulik Gottesman Siegel & Ware LLP,\\nSherman Oaks, California, for Amicus Curiae Community\\nAssociations Institute.\\n\\nAbbey R. Stemler and Matthew C. Turk, Indiana University\\nKelley School of Business, Bloomington, Indiana; Jahan C.\\nSagafi and Relic Sun, Outten & Golden LLP, San Francisco,\\nCalifornia; Peter Romer-Friedman, Outten & Golden LLP,\\nWashington, D.C., for Amici Curiae Internet, Business, and\\nLocal Government Law Professors.\\n\\nRichard G. McCracken and Paul L. More, McCracken,\\nStemerman & Holsberry, LLP, San Francisco, California,\\nfor Amicus Curiae UNITE HERE International Union.\\n\\nDennis J. Herrera, San Francisco City Attorney; Christine\\nVan Aken, Chief of Appellate Litigation; Yvonne R. Meré,\\nChief of Complex and Affirmative Litigation; Sara J.\\nEisenberg, Deputy City Attorney; San Francisco, California\\nfor Amicus Curiae City and County of San Francisco.\\n\\nKarl A. Racine, Attorney General for the District of\\nColumbia, Washington, D.C., for Amicus Curiae District of\\nColumbia.\\n\\nAndre M. Davis, City Solicitor, City of Baltimore,\\nBaltimore, Maryland, for Amicus Curiae Mayor and City\\nCouncil of Baltimore.\\n\\nZach Klein, Columbus City Attorney, Columbus, Ohio, for\\nAmicus Curiae City of Columbus.\\n\\nBarbara J. Parker, Oakland City Attorney; Maria Bee,\\nSpecial Counsel; Erin Bernstein, Supervising Deputy City\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 7\\n\\nAttorney; Oakland, California; for Amicus Curiae City of\\nOakland.\\n\\nPeter S. Holmes, Seattle City Attorney, Seattle, Washington,\\nfor Amicus Curiae City of Seattle.\\n\\nJill E. Habig, Founder & President; Joanna Pearl, Legal\\nDirector; Oakland, California; for Amicus Curiae Public\\nRights Project, A Project of Tides Center.\\n\\n\\n OPINION\\n\\nNGUYEN, Circuit Judge:\\n\\n Located on the coast of Southern California, the city of\\nSanta Monica consists of only about eight square miles but\\nserves 90,000 residents and as many as 500,000 visitors on\\nweekends and holidays. Similar to other popular tourist\\ndestinations, Santa Monica is struggling to manage the\\ndisruptions brought about by the rise of short-term rentals\\nfacilitated by innovative startups such as Appellants\\nHomeAway.com, Inc. and Airbnb Inc. (the “Platforms”).\\nWebsites like those operated by the Platforms are essentially\\nonline marketplaces that allow “guests” seeking\\naccommodations and “hosts” offering accommodations to\\nconnect and enter into rental agreements with one another. 1\\nAs of February 2018, Airbnb had approximately 1,400\\nlistings in Santa Monica, of which about 30 percent are in\\n\\n 1\\n The Platforms do not own, lease, or manage any of the properties\\nlisted on their websites, nor are they parties to the rental agreements.\\nInstead, the content provided alongside the listings—such as description,\\nprice, and availability—are provided by the hosts. For their services, the\\nPlatforms collect a fee from each successful booking.\\n\\f8 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nthe “coastal zone” covered by the California Coastal Act,\\nwhile HomeAway.com had approximately 300 live listings\\nin Santa Monica, of which approximately 40 percent are in\\nthe coastal zone.\\n\\n Santa Monica’s council reported that the proliferation of\\nshort-term rentals had negatively impacted the quality and\\ncharacter of its neighborhoods by “bringing commercial\\nactivity and removing residential housing stock from the\\nmarket” at a time when California is already suffering from\\nsevere housing shortages. In response, the city passed an\\nordinance regulating the short-term vacation rental market\\nby authorizing licensed “home-sharing” (rentals where\\nresidents remain on-site with guests) but prohibiting all other\\nshort-term home rentals of 30 consecutive days or less.\\n\\n The Platforms filed suit, alleging that the city ordinance\\nis preempted by the Communications Decency Act and\\nimpermissibly infringes upon their First Amendment rights.\\nThe district court denied preliminary injunctive relief, and\\ndismissed the Platforms’ complaints for failure to state a\\nclaim under the Communications Decency Act and the First\\nAmendment. We affirm.\\n\\n BACKGROUND\\n\\n In May 2015, Santa Monica passed its initial ordinance\\nregulating the short-term vacation rental market by\\nauthorizing licensed “home-sharing” (rentals where\\nresidents remain on-site with guests) but prohibiting all other\\nforms of short-term rentals for 30 consecutive days or less.\\nSanta Monica Ordinance 2484 (May 12, 2015), codified as\\namended, Santa Monica Mun. Code §§ 6.20.010–6.20.100.\\nThe ordinance reflected the city’s housing goals of\\n“preserving its housing stock and preserving the quality and\\ncharacter of its existing single and multi-family residential\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 9\\n\\nneighborhoods.” Id. As originally enacted, the ordinance\\nprohibited hosting platforms from acting to “undertake,\\nmaintain, authorize, aid, facilitate or advertise any Home-\\nSharing activity” that was not authorized by the city.\\nHosting platforms also were required to collect and remit\\ntaxes, and to regularly disclose listings and booking\\ninformation to the city.\\n\\n The Platforms each filed a complaint in the Central\\nDistrict of California challenging the initial ordinance, and\\nthe district court consolidated the cases for discovery and\\npretrial matters. On September 21, 2016, the parties\\nstipulated to stay the case while the city considered\\namendments to the local ordinance. During the stay period,\\nthe district court for the Northern District of California\\ndenied a preliminary injunction requested by the plaintiffs in\\na separate case challenging a similar ordinance in San\\nFrancisco. See Airbnb Inc. v. City & County of San\\nFrancisco, 217 F. Supp. 3d 1066 (N.D. Cal. 2016). That\\ncase ended in a settlement in which the Platforms agreed to\\ncomply with an amended version of San Francisco’s\\nordinance that prohibited booking unlawful transactions but\\nprovided a safe harbor wherein any platform that complies\\nwith the responsibilities set out in the Ordinance will be\\npresumed to be in compliance with the law.\\n\\n In January 2017, Santa Monica likewise amended its\\nown ordinance. The version challenged here, Ordinance\\n2535 (the “Ordinance”), retains its prohibitions on most\\ntypes of short-term rentals, with the exception of licensed\\nhome-shares. In addition, the Ordinance imposes four\\nobligations on hosting platforms directly: (1) collecting and\\nremitting “Transient Occupancy Taxes,” (2) disclosing\\ncertain listing and booking information regularly,\\n(3) refraining from completing any booking transaction for\\n\\f10 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nproperties not licensed and listed on the City’s registry, and\\n(4) refraining from collecting or receiving a fee for\\n“facilitating or providing services ancillary to a vacation\\nrental or unregistered home-share.” If a housing platform\\noperates in compliance with these obligations, the Ordinance\\nprovides a safe harbor by presuming the platform to be in\\ncompliance with the law. Otherwise, violations are\\npunishable by a fine of up to $500 and/or imprisonment for\\nup to six months.\\n\\n After the district court lifted the stay, the Platforms\\namended their complaint to challenge the revised ordinance\\nand moved for a preliminary injunction. Santa Monica\\nmoved to dismiss the amended complaint. The court denied\\nthe Platforms’ motion for preliminary injunctive relief and\\nsubsequently granted Santa Monica’s motion to dismiss on\\nthe ground that the Platforms failed to state a claim under\\nfederal law, including the Communications Decency Act of\\n1996 and the First Amendment. The district court also\\ndeclined to exercise supplemental jurisdiction over their\\nremaining state-law claims. 2 The Platforms timely appealed\\nthese decisions, and we consolidated the appeals.\\n\\n\\n\\n\\n 2\\n The Platforms do not appeal the district court’s dismissal of other\\nfederal claims brought under the Fourth Amendment and the Stored\\nCommunications Act. Similarly, they do not challenge the court’s\\ndecision not to exercise supplemental jurisdiction over the state-law\\nclaims under the California Coastal Act if we affirm the dismissal of their\\nfederal claims. Because we affirm the district court’s dismissal, we need\\nnot consider the state-law claims. We deny Santa Monica’s motion for\\njudicial notice of its prior enforcement actions because the dispute as to\\nits prior actions relates only to the state-law claims.\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 11\\n\\n JURISDICTION AND STANDARD OF REVIEW\\n\\n We have jurisdiction under 28 U.S.C. § 1291. We\\nreview the district court’s order of dismissal de novo,\\n“accepting all factual allegations in the complaint as true and\\nconstruing them in the light most favorable to the\\nnonmoving party.” Yagman v. Garcetti, 852 F.3d 859, 863\\n(2017) (quoting Ebner v. Fresh, Inc., 838 F.3d 958, 962 (9th\\nCir. 2016)).\\n\\n DISCUSSION\\n\\n I. Communications Decency Act\\n\\n The Communications Decency Act of 1996 (“CDA” or\\nthe “Act”), 47 U.S.C. § 230, provides internet companies\\nwith immunity from certain claims in furtherance of its\\nstated policy “to promote the continued development of the\\nInternet and other interactive computer services.” Id.\\n§ 230(b)(1). Construing this immunity broadly, the\\nPlatforms argue that the Ordinance requires them to monitor\\nand remove third-party content, and therefore violates the\\nCDA by interfering with federal policy protecting internet\\ncompanies from liability for posting third-party content.\\nSanta Monica, on the other hand, argues that the Ordinance\\ndoes not implicate the CDA because it imposes no obligation\\non the Platforms to monitor or edit any listings provided by\\nhosts. Santa Monica contends that the Ordinance is simply\\nan exercise of its right to enact regulations to preserve\\nhousing by curtailing “incentives for landlords to evade rent\\ncontrol laws, evict tenants, and convert residential units into\\nde facto hotels.”\\n\\n We begin our analysis with the text of the CDA. See BP\\nAmerica Prod. Co. v. Burton, 549 U.S. 84, 91 (2006).\\nSection 230(c)(1) states that “[n]o provider or user of an\\n\\f12 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\ninteractive computer service shall be treated as the publisher\\nor speaker of any information provided by another\\ninformation content provider.” Id. § 230(c)(1). The CDA\\nexplicitly preempts inconsistent state laws: “Nothing in this\\nsection shall be construed to prevent any State from\\nenforcing any State law that is consistent with this section.\\nNo cause of action may be brought and no liability may be\\nimposed under any State or local law that is inconsistent with\\nthis section.” Id. § 230(e)(3).\\n\\n We have construed these provisions to extend immunity\\nto “(1) a provider or user of an interactive computer service\\n(2) whom a plaintiff seeks to treat, under a state law cause of\\naction, as a publisher or speaker (3) of information provided\\nby another information content provider.” Barnes v. Yahoo!,\\nInc., 570 F.3d 1096, 1100–01 (9th Cir. 2009). Only the\\nsecond element is at issue here: whether the Ordinance treats\\nthe Platforms as a “publisher or speaker” in a manner that is\\nbarred by the CDA. Although the CDA does not define\\n“publisher,” we have defined “publication” in this context to\\n“involve[] reviewing, editing, and deciding whether to\\npublish or to withdraw from publication third-party\\ncontent.” Id. at 1102 (citing Fair Hous. Council v.\\nRoommates.com, LLC, 521 F.3d 1157, 1170–71 (9th Cir.\\n2008) (en banc)).\\n\\n The Platforms offer two different theories as to how the\\nOrdinance in fact reaches “publication” activities. First, the\\nPlatforms claim that the Ordinance is expressly preempted\\nby the CDA because, as they argue, it implicitly requires\\nthem “to monitor the content of a third-party listing and\\ncompare it against the City’s short-term rental registry\\nbefore allowing any booking to proceed.” Relying on Doe\\nv. Internet Brands, 824 F.3d 846, 851 (9th Cir. 2016), the\\nPlatforms take the view that CDA immunity follows\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 13\\n\\nwhenever a legal duty “affects” how an internet company\\n“monitors” a website.\\n\\n However, the Platforms read Internet Brands too\\nbroadly. In that case, two individuals used the defendant’s\\nwebsite to message and lure the plaintiff to sham auditions\\nwhere she was drugged and raped. Id. at 848. We held that,\\nwhere the website provider was alleged to have known\\nindependently of the ongoing scheme beforehand, the CDA\\ndid not bar an action under state law for failure to warn. Id.\\nat 854. We observed that a duty to warn would not\\n“otherwise affect how [the defendant] publishes or\\nmonitors” user content. Id. at 851. Though the defendant\\ndid, in its business, act as a publisher of third-party content,\\nthe underlying legal duty at issue did not seek to hold the\\ndefendant liable as a “publisher or speaker” of third-party\\ncontent. Id. at 853; see 47 U.S.C. § 230(c)(1). We therefore\\ndeclined to extend CDA immunity to the defendant for the\\nplaintiff’s failure-to-warn claim. Internet Brands, 824 F.3d\\nat 854.\\n\\n We do not read Internet Brands to suggest that CDA\\nimmunity attaches any time a legal duty might lead a\\ncompany to respond with monitoring or other publication\\nactivities. It is not enough that third-party content is\\ninvolved; Internet Brands rejected use of a “but-for” test that\\nwould provide immunity under the CDA solely because a\\ncause of action would not otherwise have accrued but for the\\nthird-party content. Id. at 853. We look instead to what the\\nduty at issue actually requires: specifically, whether the duty\\nwould necessarily require an internet company to monitor\\nthird-party content. See id. at 851, 853.\\n\\n Here, the Ordinance does not require the Platforms to\\nmonitor third-party content and thus falls outside of the\\nCDA’s immunity. The Ordinance prohibits processing\\n\\f14 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\ntransactions for unregistered properties. It does not require\\nthe Platforms to review the content provided by the hosts of\\nlistings on their websites. Rather, the only monitoring that\\nappears necessary in order to comply with the Ordinance\\nrelates to incoming requests to complete a booking\\ntransaction—content that, while resulting from the third-\\nparty listings, is distinct, internal, and nonpublic. As in\\nInternet Brands, it is not enough that the third-party listings\\nare a “but-for” cause of such internal monitoring. See\\n824 F.3d at 853. The text of the CDA is “clear that neither\\nthis subsection nor any other declares a general immunity\\nfrom liability deriving from third-party content.” Barnes,\\n570 F.3d at 1100. To provide broad immunity “every time a\\nwebsite uses data initially obtained from third parties would\\neviscerate [the CDA].” Barnes, 570 F.3d at 1100 (quoting\\nRoommates.com, 521 F.3d at 1171 (9th Cir. 2008) (en\\nbanc)). That is not the result that Congress intended.\\n\\n Nor could a duty to cross-reference bookings against\\nSanta Monica’s property registry give rise to CDA\\nimmunity. While keeping track of the city’s registry is\\n“monitoring” third-party content in the most basic sense,\\nsuch conduct cannot be fairly classified as “publication” of\\nthird-party content. The Platforms have no editorial control\\nover the registry whatsoever. As with tax regulations or\\ncriminal statutes, the Ordinance can fairly charge parties\\nwith keeping abreast of the law without running afoul of the\\nCDA.\\n\\n Second, the Platforms argue that the Ordinance “in\\noperation and effect . . . forces [them] to remove third-party\\ncontent.” Although it is clear that the Ordinance does not\\nexpressly mandate that they do so, the Platforms claim that\\n“common sense explains” that they cannot “leave in place a\\nwebsite chock-full of un-bookable listings.” For purposes of\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 15\\n\\nour review, we accept at face value the Platforms’ assertion\\nthat they will choose to remove noncompliant third-party\\nlistings on their website as a consequence of the Ordinance. 3\\nNonetheless, their choice to remove listings is insufficient to\\nimplicate the CDA.\\n\\n On its face, the Ordinance does not proscribe, mandate,\\nor even discuss the content of the listings that the Platforms\\ndisplay on their websites. See Santa Monica Mun. Code\\n§§ 6.20.010–6.20.100. It requires only that transactions\\ninvolve licensed properties. We acknowledge that, as the\\nPlatforms explain in Airbnb’s complaint and in the briefing\\non appeal, removal of these listings would be the best option\\n“from a business standpoint.” But, as in Internet Brands, the\\nunderlying duty “could have been satisfied without changes\\nto content posted by the website’s users.” See 824 F.3d\\nat 851. Even assuming that removing certain listings may be\\nthe Platforms’ most practical compliance option, allowing\\ninternet companies to claim CDA immunity under these\\ncircumstances would risk exempting them from most local\\nregulations and would, as this court feared in\\nRoommates.com, 521 F.3d at 1164, “create a lawless no-\\nman’s-land on the Internet.” We hold that the Ordinance is\\nnot “inconsistent” with the CDA, and is therefore not\\nexpressly preempted by its terms. See 47 U.S.C. § 230(e)(3).\\n\\n Finally, the Platforms argue that, even if the Ordinance\\nis not expressly preempted by the CDA, the Ordinance\\nimposes “an obstacle to the accomplishment and execution\\n 3\\n The Platforms argued below that the district court must accept as\\ntrue their allegation that they would “have to” monitor and screen\\nlistings. As a matter of law, the Ordinance does not require them to do\\nso. Courts are “not bound to accept as true a legal conclusion couched\\nas a factual allegation.” Bell Atlantic Corp. v. Twombly, 550 U.S. 544,\\n555 (2007) (quoting Papasan v. Allain, 478 U.S. 265, 286 (1986)).\\n\\f16 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nof the full purposes and objectives of Congress.” See Crosby\\nv. Nat’l Foreign Trade Council, 530 U.S. 363, 372–73\\n(2000). Reading the CDA expansively, they argue that the\\nOrdinance conflicts with the CDA’s goal “to preserve the\\nvibrant and competitive free market that presently exists for\\nthe Internet . . . unfettered by Federal or State regulation.”\\nSee § 230(b)(2). We have consistently eschewed an\\nexpansive reading of the statute that would render unlawful\\nconduct “magically . . . lawful when [conducted] online,”\\nand therefore “giv[ing] online businesses an unfair\\nadvantage over their real-world counterparts.” See\\nRoommates.com, 521 F.3d at 1164, 1164–65 n.15. For the\\nsame reasons, while we acknowledge the Platforms’\\nconcerns about the difficulties of complying with numerous\\nstate and local regulations, the CDA does not provide\\ninternet companies with a one-size-fits-all body of law. Like\\ntheir brick-and-mortar counterparts, internet companies\\nmust also comply with any number of local regulations\\nconcerning, for example, employment, tax, or zoning.\\nBecause the Ordinance would not pose an obstacle to\\nCongress’s aim to encourage self-monitoring of third-party\\ncontent, we hold that obstacle preemption does not preclude\\nSanta Monica from enforcing the Ordinance.\\n\\n Fundamentally, the parties dispute how broadly to\\nconstrue the CDA so as to continue serving the purposes\\nCongress envisioned while allowing state and local\\ngovernments breathing room to address the pressing issues\\nfaced by their communities. We have previously\\nacknowledged that the CDA’s immunity reaches beyond the\\ninitial state court decision that sparked its enactment. See\\nFair Hous. Council v. Roommates.com, LLC, 521 F.3d 1157,\\n1163 (9th Cir. 2008) (en banc) (discussing Stratton\\nOakmont, Inc. v. Prodigy Servs. Co., which held an internet\\ncompany liable for defamation when it removed some, but\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 17\\n\\nnot all, harmful content from its public message boards,\\n1995 WL 323710 (N.Y. Sup. Ct. May 24, 1995)\\n(unpublished)). As the Platforms correctly note, the Act’s\\npolicy statements broadly promote “the vibrant and\\ncompetitive free market that presently exists for the Internet\\n. . . unfettered by Federal or State regulation.” See 47 U.S.C.\\n§ 230(b)(2). “[A] law’s scope often differs from its genesis,”\\nand we have repeatedly held the scope of immunity to reach\\nbeyond defamation cases. Barnes, 570 F.3d at 1101 (quoting\\nChicago Lawyers’ Comm. for Civil Rights Under Law, Inc.\\nv. Craigslist, Inc., 519 F.3d 666, 671 (7th Cir. 2008), as\\namended (May 2, 2008)) (citing cases applying immunity for\\ncauses of action including discrimination, fraud, and\\nnegligence).\\n\\n At the same time, our cases have hewn closely to the\\nstatutory language of the CDA and have limited the\\nexpansion of its immunity beyond the protection Congress\\nenvisioned. As we have observed, “the [relevant] section is\\ntitled ‘Protection for “good Samaritan” blocking and\\nscreening of offensive material.’” Roommates.com,\\n521 F.3d at 1163–64 (quoting 47 U.S.C. § 230(c)); see also\\nInternet Brands, 824 F.3d at 852. Congress intended to\\n“spare interactive computer services [the] grim choice”\\nbetween voluntarily filtering content and being subject to\\nliability on the one hand, and “ignoring all problematic posts\\naltogether [to] escape liability.” Roommates.com, 521 F.3d\\nat 1163–64. In contrast, the Platforms face no liability for\\nthe content of the bookings; rather, any liability arises only\\nfrom unlicensed bookings. We do not discount the\\nPlatforms’ concerns about the administrative burdens of\\nstate and local regulations, but we nonetheless disagree that\\n§ 230(c)(1) of the CDA may be read as broadly as they\\nadvocate, or that we may ourselves expand its provisions\\nbeyond what Congress initially intended.\\n\\f18 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\n In sum, neither express preemption nor obstacle\\npreemption apply to the Ordinance. We therefore affirm the\\ndistrict court’s dismissal for failure to state a claim under the\\nCDA.\\n\\n II. First Amendment\\n\\n The Platforms also contend that the district court erred in\\ndismissing their First Amendment claims. They argue that,\\neven if the plain language of the Ordinance only reaches\\n“conduct,” i.e., booking unlicensed properties, the law\\neffectively imposes a “content-based financial burden” on\\ncommercial speech and is thus subject to First Amendment\\nscrutiny. The district court concluded that the Ordinance\\n“regulates conduct, not speech, and that the conduct banned\\n. . . does not have such a ‘significant expressive element’ as\\nto draw First Amendment protection.” We agree.\\n\\n That the Ordinance regulates “conduct” is not alone\\ndispositive. The Supreme Court has previously applied First\\nAmendment scrutiny when “‘speech’ and ‘nonspeech’\\nelements are combined in the same course of conduct.” See\\nUnited States v. O’Brien, 391 U.S. 367, 376 (1968). But\\n“restrictions on protected expression are distinct from\\nrestrictions on economic activity or, more generally, on\\nnonexpressive conduct.” Sorrell v. IMS Health Inc.,\\n564 U.S. 552, 567 (2011). While the former is entitled to\\nprotection, “the First Amendment does not prevent\\nrestrictions directed at commerce or conduct from imposing\\nincidental burdens on speech.” Id.\\n\\n To determine whether the First Amendment applies, we\\nmust first ask the “threshold question [of] whether conduct\\nwith a ‘significant expressive element’ drew the legal\\nremedy or the ordinance has the inevitable effect of ‘singling\\nout those engaged in expressive activity.’” Int’l Franchise\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 19\\n\\nAss’n v. City of Seattle, 803 F.3d 389, 408 (9th Cir. 2015)\\n(quoting Arcara v. Cloud Books, Inc., 478 U.S. 697, 706–07\\n(1986)). A court may consider the “inevitable effect of a\\nstatute on its face,” as well as a statute’s “stated purpose.”\\nSorrell, 564 U.S. at 565. However, absent narrow\\ncircumstances, a court may not conduct an inquiry into\\nlegislative purpose or motive beyond what is stated within\\nthe statute itself. See O’Brien, 391 U.S. at 383 n.30.\\nBecause the conduct at issue—completing booking\\ntransactions for unlawful rentals—consists only of\\nnonspeech, nonexpressive conduct, we hold that the\\nOrdinance does not implicate the First Amendment.\\n\\n First, the prohibitions here did not target conduct with “a\\nsignificant expressive element.” See Arcara, 478 U.S.\\nat 706. Our decision in International Franchise Ass’n is\\nanalogous. There, the plaintiff challenged a minimum wage\\nordinance that would have accelerated the raising of the\\nminimum wage to $15 per hour for franchise owners and\\nother large employers. 803 F.3d at 389. In denying a\\npreliminary injunction, the district court held that the\\nplaintiffs were not likely to succeed on their First\\nAmendment argument that the ordinance treated them\\ndifferently based on their “speech and association” decisions\\nto operate within a franchise relationship framework. Id. at\\n408–09. We agreed, concluding that the “business\\nagreement or business dealings” were not conduct with a\\n“significant expressive element.” Id. at 408. Instead,\\n“Seattle’s minimum wage ordinance [was] plainly an\\neconomic regulation that [did] not target speech or\\nexpressive conduct.” Id.\\n\\n Similarly, here, the Ordinance is plainly a housing and\\nrental regulation. The “inevitable effect of the [Ordinance]\\non its face” is to regulate nonexpressive conduct—namely,\\n\\f20 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nbooking transactions—not speech. See Sorrell, 564 U.S.\\nat 565. As in International Franchise Ass’n, the “business\\nagreement or business dealings” associated with processing\\na booking is not conduct with a “significant expressive\\nelement.” See 803 F.3d at 408 (citation and quotation marks\\nomitted). Contrary to the Platforms’ claim, the Ordinance\\ndoes not “require” that they monitor or screen\\nadvertisements. It instead leaves them to decide how best to\\ncomply with the prohibition on booking unlawful\\ntransactions.\\n\\n Nor can the Platforms rely on the Ordinance’s “stated\\npurpose” to argue that it intends to regulate speech. The\\nOrdinance itself makes clear that the City’s “central and\\nsignificant goal . . . is preservation of its housing stock and\\npreserving the quality and nature of residential\\nneighborhoods.” As such, with respect to the Platforms, the\\nonly inevitable effect, and the stated purpose, of the\\nOrdinance is to prohibit them from completing booking\\ntransactions for unlawful rentals.\\n\\n As for the second prong of our inquiry, whether the\\nOrdinance has the effect of “singling out those engaged in\\nexpressive activity,” Arcara, 478 U.S. at 706–07, we\\nconclude that it does not. As the Platforms point out,\\nwebsites like Craigslist “advertise the very same properties,”\\nbut do not process transactions. Unlike the Platforms, those\\nwebsites would not be subject to the Ordinance,\\nunderscoring that the Ordinance does not target websites that\\npost listings, but rather companies that engage in unlawful\\nbooking transactions.\\n\\n Moreover, the incidental impacts on speech cited by the\\nPlatforms raise minimal concerns. The Platforms argue that\\nthe Ordinance chills commercial speech, namely,\\nadvertisements for third-party rentals. But even accepting\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 21\\n\\nthat the Platforms will need to engage in efforts to validate\\ntransactions before completing them, incidental burdens like\\nthese are not always sufficient to trigger First Amendment\\nscrutiny. See Int’l Franchise Ass’n, 803 F.3d at 408\\n(“[S]ubjecting every incidental impact on speech to First\\nAmendment scrutiny ‘would lead to the absurd result that\\nany government action that had some conceivable speech\\ninhibiting consequences . . . would require analysis under\\nthe First Amendment.’” (quoting Arcara, 478 U.S. at 708\\n(O’Connor, J., concurring))). Furthermore, to the extent that\\nthe speech chilled advertises unlawful rentals, “[a]ny First\\nAmendment interest . . . is altogether absent when the\\ncommercial activity itself is illegal and the restriction on\\nadvertising is incidental to a valid limitation on economic\\nactivity.” See Pittsburgh Press Co. v. Pittsburgh Comm’n\\non Human Relations, 413 U.S. 376, 389 (1973).\\n\\n Finally, because the Ordinance does not implicate speech\\nprotected by the First Amendment, we similarly reject the\\nPlatforms’ argument that the Ordinance is unconstitutional\\nwithout a scienter requirement. In most cases, there is no\\n“closed definition” on when a criminal statute must contain\\na scienter requirement. See Morissette v. United States,\\n342 U.S. 246, 260 (1952). However, the Supreme Court has\\ndrawn a bright line in certain contexts, such as holding that\\nthe First Amendment requires statutes imposing criminal\\nliability for obscenity or child pornography to contain a\\nscienter requirement. See New York v. Ferber, 458 U.S. 747,\\n765 (1982). Such a requirement prevents “a severe\\nlimitation on the public’s access to constitutionally protected\\nmatter” as would result from inflexible laws criminalizing\\n“bookshops and periodical stands.” Smith v. California,\\n361 U.S. 147, 153 (1959).\\n\\f22 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\n Here, even assuming that the Ordinance would lead the\\nPlatforms to voluntarily remove some advertisements for\\nlawful rentals, there would not be a “severe limitation on the\\npublic’s access” to lawful advertisements, especially\\nconsidering the existence of alternative channels like\\nCraigslist. Id. Such an incidental burden is far from “a\\nsubstantial restriction on the freedom of speech” that would\\nnecessitate a scienter requirement. Id. at 150. Otherwise,\\n“[t]here is no specific constitutional inhibition against\\nmaking the distributors of good[s] the strictest censors of\\ntheir merchandise.” Id. at 152.\\n\\n III. Remaining Claims\\n\\n On appeal, the Platforms do not challenge dismissal of\\ntheir other federal law claims “in light of the district court’s\\ninterpretation of the Ordinance as only requiring disclosure\\nof information pursuant to requests that comply with the\\nFourth Amendment and Stored Communications Act.”\\nSimilarly, the parties specified that they would “not\\nchallenge the district court’s decision to decline\\nsupplemental jurisdiction if all the Platforms’ federal claims\\nwere properly dismissed.” Accordingly, we need not\\nconsider the remaining claims.\\n\\n * * *\\n\\n Because the district court properly dismissed the\\nPlatforms’ complaints for failure to state a claim, we dismiss\\nas moot the appeals from the denial of preliminary injunctive\\nrelief.\\n\\n AFFIRMED in part, DISMISSED in part.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4380673/\", \"author_raw\": \"NGUYEN, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"MARY M SCHROEDER"},"judge2":{"kind":"string","value":"JACQUELINE H NGUYEN"},"judge3":{"kind":"string","value":"MICHAEL H SIMON"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4603420/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: homeaway.com, Inc. v. City of Santa Monica\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-13\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nHOMEAWAY.COM, INC.; AIRBNB No. 18-55367\nINC.,\n Plaintiffs-Appellants, D.C. Nos.\n 2:16-cv-06641-\n v. ODW-AFM\n 2:16-cv-06645-\nCITY OF SANTA MONICA, ODW-AFM\n Defendant-Appellee.\n\n\n\nHOMEAWAY.COM, INC., No. 18-55805\n Plaintiff-Appellant,\n D.C. Nos.\n and 2:16-cv-06641-\n ODW-AFM\nAIRBNB INC., 2:16-cv-06645-\n Plaintiff, ODW-AFM\n\n v.\n\nCITY OF SANTA MONICA,\n Defendant-Appellee.\n\f2 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\n\n AIRBNB INC., No. 18-55806\n Plaintiff-Appellant,\n D.C. No.\n v. 2:16-cv-06645-\n ODW-AFM\n CITY OF SANTA MONICA,\n Defendant-Appellee. OPINION\n\n\n\n Appeal from the United States District Court\n for the Central District of California\n Otis D. Wright II, District Judge, Presiding\n\n Argued and Submitted October 12, 2018\n Pasadena, California\n\n Filed March 13, 2019\n\n Before: Mary M. Schroeder and Jacqueline H. Nguyen,\n Circuit Judges, and Michael H. Simon, * District Judge.\n\n Opinion by Judge Nguyen\n\n\n\n\n *\n The Honorable Michael H. Simon, United States District Judge for\nthe District of Oregon, sitting by designation.\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 3\n\n SUMMARY **\n\n\n Civil Rights\n\n The panel affirmed the district court’s dismissal, for\nfailure to state a claim, of a complaint brought by\nHomeAway.com and Airbnb Inc. challenging the City of\nSanta Monica’s Ordinance 2535, which imposes various\nobligations on companies that host online platforms for\nshort-term vacation rentals.\n\n The Ordinance, as amended in 2017, imposes four\nobligations on hosting platforms: (1) collecting and\nremitting Transient Occupancy Taxes; (2) regularly\ndisclosing listings and booking information to the City; (3)\nrefraining from booking properties not licensed and listed on\nthe City’s registry; (4) and refraining from collecting a fee\nfor ancillary services.\n\n The panel rejected plaintiffs’ assertion that the\nOrdinance violated the Communications Decency Act of\n1996, 47 U.S.C. § 230 because it required them to monitor\nand remove third-party content, and therefore interfered with\nfederal policy protecting internet companies from liability\nfor posting third-party content. The panel stated that the\nOrdinance prohibits processing transactions for unregistered\nproperties. It does not require the Platforms to review the\ncontent provided by the hosts of listings on their websites.\nRather, the panel noted that the only monitoring that\nappeared necessary in order to comply with the Ordinance\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f4 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\nrelated to incoming requests to complete a booking\ntransaction—content that, while resulting from the third-\nparty listings, was distinct, internal, and nonpublic. The\npanel concluded that the Ordinance was not inconsistent\nwith the Communications Decency Act, and therefore was\nnot expressly preempted by its terms. The panel further\nconcluded that the Ordinance would not pose an obstacle to\nCongress’s aim to encourage self-monitoring of third-party\ncontent, and therefore obstacle preemption did not preclude\nSanta Monica from enforcing the Ordinance.\n\n The panel held that the Ordinance did not implicate\nspeech protected by the First Amendment, concluding that\nthe Ordinance’s prohibitions regulate nonexpressive\nconduct, specifically booking transactions, and do not single\nout those engaged in expressive activity.\n\n\n COUNSEL\n\nDonald B. Verrilli, Jr. (argued) and Chad Golder, Munger,\nTolles & Olson LLP, Washington, D.C.; Joseph W. Cotchett\nand Alexandra P. Summer, Cotchett, Pitre & McCarthy,\nLLP, Santa Monica, California; Jonathan H. Blavin and\nJoshua Patashnik, Munger, Tolles & Olson LLP, San\nFrancisco, California; John B. Major, Munger, Tolles &\nOlson, Los Angeles, California; for Plaintiff-Appellant\nAirbnb, Inc.\n\nStephen M. Rummage and Ambika K. Doran, Davis Wright\nTremaine LLP, Seattle, Washington, for Plaintiff-Appellant\nHomeAway.com, Inc.\n\nGeorge S. Cardona (argued), Deputy City Attorney; Lane\nDilg, City Attorney; Yibin Shen, Chief Deputy City\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 5\n\nAttorney; Heidi Von Tongeln and Michael R. Cobden,\nDeputy City Attorneys; Santa Monica City Attorney’s\nOffice, Santa Monica, California, for Defendant-Appellee.\n\nDavid Salmons, Bryan Killian, and James Nelson, Morgan,\nLewis & Bockius LLP, Washington, D.C., for Amici Curiae\nChris Cox and NetChoice.\n\nCatherine R. Gellis, Esq., Sausalito, California, for Amicus\nCuriae Floor64, Inc. d/b/a The Copia Institute.\n\nIan C. Ballon and Lori Chang, Greenberg Traurig, LLP, Los\nAngeles, California, for Amici Curiae Ebay Inc., Glassdoor,\nInc., Lyft, Inc., Offerup, Inc., Taskrabbit, Inc., Thumbtack,\nInc., Uber Technologies, Inc., and Upwork, Inc.\n\nChristi Hogin, Best Best & Krieger LLP, Manhattan Beach,\nCalifornia, for Amici Curiae League of California Cities,\nInternational Municipal Lawyers Association, and\nCalifornia State Association of Counties.\n\nJeremy B. Rosen, Eric S. Boorstin, and Ryan C. Chapman,\nHorvitz & Levy LLP, Burbank, California; Michael T.\nWilliams and Allison R. Mclaughlin, Wheeler Trigg\nO’Donnell LLP, Denver, Colorado; David C. Frederick,\nBrendan J. Crimmins, and Rachel P. May, Kellogg, Hansen,\nTodd, Figel & Frederick, P.L.L.C., Washington, D.C., for\nAmicus Curiae Apartment Investment and Management\nCompany.\n\nHeidi Palutke, Sacramento, California, for Amicus Curiae\nCalifornia Apartment Association.\n\nEdward M. Schulman, Ballston Tower, Arlington, Virginia,\nfor Amicus Curiae AvalonBay Communities, Inc.\n\f6 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\nGary S. Kessler, Kulik Gottesman Siegel & Ware LLP,\nSherman Oaks, California, for Amicus Curiae Community\nAssociations Institute.\n\nAbbey R. Stemler and Matthew C. Turk, Indiana University\nKelley School of Business, Bloomington, Indiana; Jahan C.\nSagafi and Relic Sun, Outten & Golden LLP, San Francisco,\nCalifornia; Peter Romer-Friedman, Outten & Golden LLP,\nWashington, D.C., for Amici Curiae Internet, Business, and\nLocal Government Law Professors.\n\nRichard G. McCracken and Paul L. More, McCracken,\nStemerman & Holsberry, LLP, San Francisco, California,\nfor Amicus Curiae UNITE HERE International Union.\n\nDennis J. Herrera, San Francisco City Attorney; Christine\nVan Aken, Chief of Appellate Litigation; Yvonne R. Meré,\nChief of Complex and Affirmative Litigation; Sara J.\nEisenberg, Deputy City Attorney; San Francisco, California\nfor Amicus Curiae City and County of San Francisco.\n\nKarl A. Racine, Attorney General for the District of\nColumbia, Washington, D.C., for Amicus Curiae District of\nColumbia.\n\nAndre M. Davis, City Solicitor, City of Baltimore,\nBaltimore, Maryland, for Amicus Curiae Mayor and City\nCouncil of Baltimore.\n\nZach Klein, Columbus City Attorney, Columbus, Ohio, for\nAmicus Curiae City of Columbus.\n\nBarbara J. Parker, Oakland City Attorney; Maria Bee,\nSpecial Counsel; Erin Bernstein, Supervising Deputy City\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 7\n\nAttorney; Oakland, California; for Amicus Curiae City of\nOakland.\n\nPeter S. Holmes, Seattle City Attorney, Seattle, Washington,\nfor Amicus Curiae City of Seattle.\n\nJill E. Habig, Founder & President; Joanna Pearl, Legal\nDirector; Oakland, California; for Amicus Curiae Public\nRights Project, A Project of Tides Center.\n\n\n OPINION\n\nNGUYEN, Circuit Judge:\n\n Located on the coast of Southern California, the city of\nSanta Monica consists of only about eight square miles but\nserves 90,000 residents and as many as 500,000 visitors on\nweekends and holidays. Similar to other popular tourist\ndestinations, Santa Monica is struggling to manage the\ndisruptions brought about by the rise of short-term rentals\nfacilitated by innovative startups such as Appellants\nHomeAway.com, Inc. and Airbnb Inc. (the “Platforms”).\nWebsites like those operated by the Platforms are essentially\nonline marketplaces that allow “guests” seeking\naccommodations and “hosts” offering accommodations to\nconnect and enter into rental agreements with one another. 1\nAs of February 2018, Airbnb had approximately 1,400\nlistings in Santa Monica, of which about 30 percent are in\n\n 1\n The Platforms do not own, lease, or manage any of the properties\nlisted on their websites, nor are they parties to the rental agreements.\nInstead, the content provided alongside the listings—such as description,\nprice, and availability—are provided by the hosts. For their services, the\nPlatforms collect a fee from each successful booking.\n\f8 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\nthe “coastal zone” covered by the California Coastal Act,\nwhile HomeAway.com had approximately 300 live listings\nin Santa Monica, of which approximately 40 percent are in\nthe coastal zone.\n\n Santa Monica’s council reported that the proliferation of\nshort-term rentals had negatively impacted the quality and\ncharacter of its neighborhoods by “bringing commercial\nactivity and removing residential housing stock from the\nmarket” at a time when California is already suffering from\nsevere housing shortages. In response, the city passed an\nordinance regulating the short-term vacation rental market\nby authorizing licensed “home-sharing” (rentals where\nresidents remain on-site with guests) but prohibiting all other\nshort-term home rentals of 30 consecutive days or less.\n\n The Platforms filed suit, alleging that the city ordinance\nis preempted by the Communications Decency Act and\nimpermissibly infringes upon their First Amendment rights.\nThe district court denied preliminary injunctive relief, and\ndismissed the Platforms’ complaints for failure to state a\nclaim under the Communications Decency Act and the First\nAmendment. We affirm.\n\n BACKGROUND\n\n In May 2015, Santa Monica passed its initial ordinance\nregulating the short-term vacation rental market by\nauthorizing licensed “home-sharing” (rentals where\nresidents remain on-site with guests) but prohibiting all other\nforms of short-term rentals for 30 consecutive days or less.\nSanta Monica Ordinance 2484 (May 12, 2015), codified as\namended, Santa Monica Mun. Code §§ 6.20.010–6.20.100.\nThe ordinance reflected the city’s housing goals of\n“preserving its housing stock and preserving the quality and\ncharacter of its existing single and multi-family residential\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 9\n\nneighborhoods.” Id. As originally enacted, the ordinance\nprohibited hosting platforms from acting to “undertake,\nmaintain, authorize, aid, facilitate or advertise any Home-\nSharing activity” that was not authorized by the city.\nHosting platforms also were required to collect and remit\ntaxes, and to regularly disclose listings and booking\ninformation to the city.\n\n The Platforms each filed a complaint in the Central\nDistrict of California challenging the initial ordinance, and\nthe district court consolidated the cases for discovery and\npretrial matters. On September 21, 2016, the parties\nstipulated to stay the case while the city considered\namendments to the local ordinance. During the stay period,\nthe district court for the Northern District of California\ndenied a preliminary injunction requested by the plaintiffs in\na separate case challenging a similar ordinance in San\nFrancisco. See Airbnb Inc. v. City & County of San\nFrancisco, 217 F. Supp. 3d 1066 (N.D. Cal. 2016). That\ncase ended in a settlement in which the Platforms agreed to\ncomply with an amended version of San Francisco’s\nordinance that prohibited booking unlawful transactions but\nprovided a safe harbor wherein any platform that complies\nwith the responsibilities set out in the Ordinance will be\npresumed to be in compliance with the law.\n\n In January 2017, Santa Monica likewise amended its\nown ordinance. The version challenged here, Ordinance\n2535 (the “Ordinance”), retains its prohibitions on most\ntypes of short-term rentals, with the exception of licensed\nhome-shares. In addition, the Ordinance imposes four\nobligations on hosting platforms directly: (1) collecting and\nremitting “Transient Occupancy Taxes,” (2) disclosing\ncertain listing and booking information regularly,\n(3) refraining from completing any booking transaction for\n\f10 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\nproperties not licensed and listed on the City’s registry, and\n(4) refraining from collecting or receiving a fee for\n“facilitating or providing services ancillary to a vacation\nrental or unregistered home-share.” If a housing platform\noperates in compliance with these obligations, the Ordinance\nprovides a safe harbor by presuming the platform to be in\ncompliance with the law. Otherwise, violations are\npunishable by a fine of up to $500 and/or imprisonment for\nup to six months.\n\n After the district court lifted the stay, the Platforms\namended their complaint to challenge the revised ordinance\nand moved for a preliminary injunction. Santa Monica\nmoved to dismiss the amended complaint. The court denied\nthe Platforms’ motion for preliminary injunctive relief and\nsubsequently granted Santa Monica’s motion to dismiss on\nthe ground that the Platforms failed to state a claim under\nfederal law, including the Communications Decency Act of\n1996 and the First Amendment. The district court also\ndeclined to exercise supplemental jurisdiction over their\nremaining state-law claims. 2 The Platforms timely appealed\nthese decisions, and we consolidated the appeals.\n\n\n\n\n 2\n The Platforms do not appeal the district court’s dismissal of other\nfederal claims brought under the Fourth Amendment and the Stored\nCommunications Act. Similarly, they do not challenge the court’s\ndecision not to exercise supplemental jurisdiction over the state-law\nclaims under the California Coastal Act if we affirm the dismissal of their\nfederal claims. Because we affirm the district court’s dismissal, we need\nnot consider the state-law claims. We deny Santa Monica’s motion for\njudicial notice of its prior enforcement actions because the dispute as to\nits prior actions relates only to the state-law claims.\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 11\n\n JURISDICTION AND STANDARD OF REVIEW\n\n We have jurisdiction under 28 U.S.C. § 1291. We\nreview the district court’s order of dismissal de novo,\n“accepting all factual allegations in the complaint as true and\nconstruing them in the light most favorable to the\nnonmoving party.” Yagman v. Garcetti, 852 F.3d 859, 863\n(2017) (quoting Ebner v. Fresh, Inc., 838 F.3d 958, 962 (9th\nCir. 2016)).\n\n DISCUSSION\n\n I. Communications Decency Act\n\n The Communications Decency Act of 1996 (“CDA” or\nthe “Act”), 47 U.S.C. § 230, provides internet companies\nwith immunity from certain claims in furtherance of its\nstated policy “to promote the continued development of the\nInternet and other interactive computer services.” Id.\n§ 230(b)(1). Construing this immunity broadly, the\nPlatforms argue that the Ordinance requires them to monitor\nand remove third-party content, and therefore violates the\nCDA by interfering with federal policy protecting internet\ncompanies from liability for posting third-party content.\nSanta Monica, on the other hand, argues that the Ordinance\ndoes not implicate the CDA because it imposes no obligation\non the Platforms to monitor or edit any listings provided by\nhosts. Santa Monica contends that the Ordinance is simply\nan exercise of its right to enact regulations to preserve\nhousing by curtailing “incentives for landlords to evade rent\ncontrol laws, evict tenants, and convert residential units into\nde facto hotels.”\n\n We begin our analysis with the text of the CDA. See BP\nAmerica Prod. Co. v. Burton, 549 U.S. 84, 91 (2006).\nSection 230(c)(1) states that “[n]o provider or user of an\n\f12 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\ninteractive computer service shall be treated as the publisher\nor speaker of any information provided by another\ninformation content provider.” Id. § 230(c)(1). The CDA\nexplicitly preempts inconsistent state laws: “Nothing in this\nsection shall be construed to prevent any State from\nenforcing any State law that is consistent with this section.\nNo cause of action may be brought and no liability may be\nimposed under any State or local law that is inconsistent with\nthis section.” Id. § 230(e)(3).\n\n We have construed these provisions to extend immunity\nto “(1) a provider or user of an interactive computer service\n(2) whom a plaintiff seeks to treat, under a state law cause of\naction, as a publisher or speaker (3) of information provided\nby another information content provider.” Barnes v. Yahoo!,\nInc., 570 F.3d 1096, 1100–01 (9th Cir. 2009). Only the\nsecond element is at issue here: whether the Ordinance treats\nthe Platforms as a “publisher or speaker” in a manner that is\nbarred by the CDA. Although the CDA does not define\n“publisher,” we have defined “publication” in this context to\n“involve[] reviewing, editing, and deciding whether to\npublish or to withdraw from publication third-party\ncontent.” Id. at 1102 (citing Fair Hous. Council v.\nRoommates.com, LLC, 521 F.3d 1157, 1170–71 (9th Cir.\n2008) (en banc)).\n\n The Platforms offer two different theories as to how the\nOrdinance in fact reaches “publication” activities. First, the\nPlatforms claim that the Ordinance is expressly preempted\nby the CDA because, as they argue, it implicitly requires\nthem “to monitor the content of a third-party listing and\ncompare it against the City’s short-term rental registry\nbefore allowing any booking to proceed.” Relying on Doe\nv. Internet Brands, 824 F.3d 846, 851 (9th Cir. 2016), the\nPlatforms take the view that CDA immunity follows\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 13\n\nwhenever a legal duty “affects” how an internet company\n“monitors” a website.\n\n However, the Platforms read Internet Brands too\nbroadly. In that case, two individuals used the defendant’s\nwebsite to message and lure the plaintiff to sham auditions\nwhere she was drugged and raped. Id. at 848. We held that,\nwhere the website provider was alleged to have known\nindependently of the ongoing scheme beforehand, the CDA\ndid not bar an action under state law for failure to warn. Id.\nat 854. We observed that a duty to warn would not\n“otherwise affect how [the defendant] publishes or\nmonitors” user content. Id. at 851. Though the defendant\ndid, in its business, act as a publisher of third-party content,\nthe underlying legal duty at issue did not seek to hold the\ndefendant liable as a “publisher or speaker” of third-party\ncontent. Id. at 853; see 47 U.S.C. § 230(c)(1). We therefore\ndeclined to extend CDA immunity to the defendant for the\nplaintiff’s failure-to-warn claim. Internet Brands, 824 F.3d\nat 854.\n\n We do not read Internet Brands to suggest that CDA\nimmunity attaches any time a legal duty might lead a\ncompany to respond with monitoring or other publication\nactivities. It is not enough that third-party content is\ninvolved; Internet Brands rejected use of a “but-for” test that\nwould provide immunity under the CDA solely because a\ncause of action would not otherwise have accrued but for the\nthird-party content. Id. at 853. We look instead to what the\nduty at issue actually requires: specifically, whether the duty\nwould necessarily require an internet company to monitor\nthird-party content. See id. at 851, 853.\n\n Here, the Ordinance does not require the Platforms to\nmonitor third-party content and thus falls outside of the\nCDA’s immunity. The Ordinance prohibits processing\n\f14 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\ntransactions for unregistered properties. It does not require\nthe Platforms to review the content provided by the hosts of\nlistings on their websites. Rather, the only monitoring that\nappears necessary in order to comply with the Ordinance\nrelates to incoming requests to complete a booking\ntransaction—content that, while resulting from the third-\nparty listings, is distinct, internal, and nonpublic. As in\nInternet Brands, it is not enough that the third-party listings\nare a “but-for” cause of such internal monitoring. See\n824 F.3d at 853. The text of the CDA is “clear that neither\nthis subsection nor any other declares a general immunity\nfrom liability deriving from third-party content.” Barnes,\n570 F.3d at 1100. To provide broad immunity “every time a\nwebsite uses data initially obtained from third parties would\neviscerate [the CDA].” Barnes, 570 F.3d at 1100 (quoting\nRoommates.com, 521 F.3d at 1171 (9th Cir. 2008) (en\nbanc)). That is not the result that Congress intended.\n\n Nor could a duty to cross-reference bookings against\nSanta Monica’s property registry give rise to CDA\nimmunity. While keeping track of the city’s registry is\n“monitoring” third-party content in the most basic sense,\nsuch conduct cannot be fairly classified as “publication” of\nthird-party content. The Platforms have no editorial control\nover the registry whatsoever. As with tax regulations or\ncriminal statutes, the Ordinance can fairly charge parties\nwith keeping abreast of the law without running afoul of the\nCDA.\n\n Second, the Platforms argue that the Ordinance “in\noperation and effect . . . forces [them] to remove third-party\ncontent.” Although it is clear that the Ordinance does not\nexpressly mandate that they do so, the Platforms claim that\n“common sense explains” that they cannot “leave in place a\nwebsite chock-full of un-bookable listings.” For purposes of\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 15\n\nour review, we accept at face value the Platforms’ assertion\nthat they will choose to remove noncompliant third-party\nlistings on their website as a consequence of the Ordinance. 3\nNonetheless, their choice to remove listings is insufficient to\nimplicate the CDA.\n\n On its face, the Ordinance does not proscribe, mandate,\nor even discuss the content of the listings that the Platforms\ndisplay on their websites. See Santa Monica Mun. Code\n§§ 6.20.010–6.20.100. It requires only that transactions\ninvolve licensed properties. We acknowledge that, as the\nPlatforms explain in Airbnb’s complaint and in the briefing\non appeal, removal of these listings would be the best option\n“from a business standpoint.” But, as in Internet Brands, the\nunderlying duty “could have been satisfied without changes\nto content posted by the website’s users.” See 824 F.3d\nat 851. Even assuming that removing certain listings may be\nthe Platforms’ most practical compliance option, allowing\ninternet companies to claim CDA immunity under these\ncircumstances would risk exempting them from most local\nregulations and would, as this court feared in\nRoommates.com, 521 F.3d at 1164, “create a lawless no-\nman’s-land on the Internet.” We hold that the Ordinance is\nnot “inconsistent” with the CDA, and is therefore not\nexpressly preempted by its terms. See 47 U.S.C. § 230(e)(3).\n\n Finally, the Platforms argue that, even if the Ordinance\nis not expressly preempted by the CDA, the Ordinance\nimposes “an obstacle to the accomplishment and execution\n 3\n The Platforms argued below that the district court must accept as\ntrue their allegation that they would “have to” monitor and screen\nlistings. As a matter of law, the Ordinance does not require them to do\nso. Courts are “not bound to accept as true a legal conclusion couched\nas a factual allegation.” Bell Atlantic Corp. v. Twombly, 550 U.S. 544,\n555 (2007) (quoting Papasan v. Allain, 478 U.S. 265, 286 (1986)).\n\f16 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\nof the full purposes and objectives of Congress.” See Crosby\nv. Nat’l Foreign Trade Council, 530 U.S. 363, 372–73\n(2000). Reading the CDA expansively, they argue that the\nOrdinance conflicts with the CDA’s goal “to preserve the\nvibrant and competitive free market that presently exists for\nthe Internet . . . unfettered by Federal or State regulation.”\nSee § 230(b)(2). We have consistently eschewed an\nexpansive reading of the statute that would render unlawful\nconduct “magically . . . lawful when [conducted] online,”\nand therefore “giv[ing] online businesses an unfair\nadvantage over their real-world counterparts.” See\nRoommates.com, 521 F.3d at 1164, 1164–65 n.15. For the\nsame reasons, while we acknowledge the Platforms’\nconcerns about the difficulties of complying with numerous\nstate and local regulations, the CDA does not provide\ninternet companies with a one-size-fits-all body of law. Like\ntheir brick-and-mortar counterparts, internet companies\nmust also comply with any number of local regulations\nconcerning, for example, employment, tax, or zoning.\nBecause the Ordinance would not pose an obstacle to\nCongress’s aim to encourage self-monitoring of third-party\ncontent, we hold that obstacle preemption does not preclude\nSanta Monica from enforcing the Ordinance.\n\n Fundamentally, the parties dispute how broadly to\nconstrue the CDA so as to continue serving the purposes\nCongress envisioned while allowing state and local\ngovernments breathing room to address the pressing issues\nfaced by their communities. We have previously\nacknowledged that the CDA’s immunity reaches beyond the\ninitial state court decision that sparked its enactment. See\nFair Hous. Council v. Roommates.com, LLC, 521 F.3d 1157,\n1163 (9th Cir. 2008) (en banc) (discussing Stratton\nOakmont, Inc. v. Prodigy Servs. Co., which held an internet\ncompany liable for defamation when it removed some, but\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 17\n\nnot all, harmful content from its public message boards,\n1995 WL 323710 (N.Y. Sup. Ct. May 24, 1995)\n(unpublished)). As the Platforms correctly note, the Act’s\npolicy statements broadly promote “the vibrant and\ncompetitive free market that presently exists for the Internet\n. . . unfettered by Federal or State regulation.” See 47 U.S.C.\n§ 230(b)(2). “[A] law’s scope often differs from its genesis,”\nand we have repeatedly held the scope of immunity to reach\nbeyond defamation cases. Barnes, 570 F.3d at 1101 (quoting\nChicago Lawyers’ Comm. for Civil Rights Under Law, Inc.\nv. Craigslist, Inc., 519 F.3d 666, 671 (7th Cir. 2008), as\namended (May 2, 2008)) (citing cases applying immunity for\ncauses of action including discrimination, fraud, and\nnegligence).\n\n At the same time, our cases have hewn closely to the\nstatutory language of the CDA and have limited the\nexpansion of its immunity beyond the protection Congress\nenvisioned. As we have observed, “the [relevant] section is\ntitled ‘Protection for “good Samaritan” blocking and\nscreening of offensive material.’” Roommates.com,\n521 F.3d at 1163–64 (quoting 47 U.S.C. § 230(c)); see also\nInternet Brands, 824 F.3d at 852. Congress intended to\n“spare interactive computer services [the] grim choice”\nbetween voluntarily filtering content and being subject to\nliability on the one hand, and “ignoring all problematic posts\naltogether [to] escape liability.” Roommates.com, 521 F.3d\nat 1163–64. In contrast, the Platforms face no liability for\nthe content of the bookings; rather, any liability arises only\nfrom unlicensed bookings. We do not discount the\nPlatforms’ concerns about the administrative burdens of\nstate and local regulations, but we nonetheless disagree that\n§ 230(c)(1) of the CDA may be read as broadly as they\nadvocate, or that we may ourselves expand its provisions\nbeyond what Congress initially intended.\n\f18 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\n In sum, neither express preemption nor obstacle\npreemption apply to the Ordinance. We therefore affirm the\ndistrict court’s dismissal for failure to state a claim under the\nCDA.\n\n II. First Amendment\n\n The Platforms also contend that the district court erred in\ndismissing their First Amendment claims. They argue that,\neven if the plain language of the Ordinance only reaches\n“conduct,” i.e., booking unlicensed properties, the law\neffectively imposes a “content-based financial burden” on\ncommercial speech and is thus subject to First Amendment\nscrutiny. The district court concluded that the Ordinance\n“regulates conduct, not speech, and that the conduct banned\n. . . does not have such a ‘significant expressive element’ as\nto draw First Amendment protection.” We agree.\n\n That the Ordinance regulates “conduct” is not alone\ndispositive. The Supreme Court has previously applied First\nAmendment scrutiny when “‘speech’ and ‘nonspeech’\nelements are combined in the same course of conduct.” See\nUnited States v. O’Brien, 391 U.S. 367, 376 (1968). But\n“restrictions on protected expression are distinct from\nrestrictions on economic activity or, more generally, on\nnonexpressive conduct.” Sorrell v. IMS Health Inc.,\n564 U.S. 552, 567 (2011). While the former is entitled to\nprotection, “the First Amendment does not prevent\nrestrictions directed at commerce or conduct from imposing\nincidental burdens on speech.” Id.\n\n To determine whether the First Amendment applies, we\nmust first ask the “threshold question [of] whether conduct\nwith a ‘significant expressive element’ drew the legal\nremedy or the ordinance has the inevitable effect of ‘singling\nout those engaged in expressive activity.’” Int’l Franchise\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 19\n\nAss’n v. City of Seattle, 803 F.3d 389, 408 (9th Cir. 2015)\n(quoting Arcara v. Cloud Books, Inc., 478 U.S. 697, 706–07\n(1986)). A court may consider the “inevitable effect of a\nstatute on its face,” as well as a statute’s “stated purpose.”\nSorrell, 564 U.S. at 565. However, absent narrow\ncircumstances, a court may not conduct an inquiry into\nlegislative purpose or motive beyond what is stated within\nthe statute itself. See O’Brien, 391 U.S. at 383 n.30.\nBecause the conduct at issue—completing booking\ntransactions for unlawful rentals—consists only of\nnonspeech, nonexpressive conduct, we hold that the\nOrdinance does not implicate the First Amendment.\n\n First, the prohibitions here did not target conduct with “a\nsignificant expressive element.” See Arcara, 478 U.S.\nat 706. Our decision in International Franchise Ass’n is\nanalogous. There, the plaintiff challenged a minimum wage\nordinance that would have accelerated the raising of the\nminimum wage to $15 per hour for franchise owners and\nother large employers. 803 F.3d at 389. In denying a\npreliminary injunction, the district court held that the\nplaintiffs were not likely to succeed on their First\nAmendment argument that the ordinance treated them\ndifferently based on their “speech and association” decisions\nto operate within a franchise relationship framework. Id. at\n408–09. We agreed, concluding that the “business\nagreement or business dealings” were not conduct with a\n“significant expressive element.” Id. at 408. Instead,\n“Seattle’s minimum wage ordinance [was] plainly an\neconomic regulation that [did] not target speech or\nexpressive conduct.” Id.\n\n Similarly, here, the Ordinance is plainly a housing and\nrental regulation. The “inevitable effect of the [Ordinance]\non its face” is to regulate nonexpressive conduct—namely,\n\f20 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\nbooking transactions—not speech. See Sorrell, 564 U.S.\nat 565. As in International Franchise Ass’n, the “business\nagreement or business dealings” associated with processing\na booking is not conduct with a “significant expressive\nelement.” See 803 F.3d at 408 (citation and quotation marks\nomitted). Contrary to the Platforms’ claim, the Ordinance\ndoes not “require” that they monitor or screen\nadvertisements. It instead leaves them to decide how best to\ncomply with the prohibition on booking unlawful\ntransactions.\n\n Nor can the Platforms rely on the Ordinance’s “stated\npurpose” to argue that it intends to regulate speech. The\nOrdinance itself makes clear that the City’s “central and\nsignificant goal . . . is preservation of its housing stock and\npreserving the quality and nature of residential\nneighborhoods.” As such, with respect to the Platforms, the\nonly inevitable effect, and the stated purpose, of the\nOrdinance is to prohibit them from completing booking\ntransactions for unlawful rentals.\n\n As for the second prong of our inquiry, whether the\nOrdinance has the effect of “singling out those engaged in\nexpressive activity,” Arcara, 478 U.S. at 706–07, we\nconclude that it does not. As the Platforms point out,\nwebsites like Craigslist “advertise the very same properties,”\nbut do not process transactions. Unlike the Platforms, those\nwebsites would not be subject to the Ordinance,\nunderscoring that the Ordinance does not target websites that\npost listings, but rather companies that engage in unlawful\nbooking transactions.\n\n Moreover, the incidental impacts on speech cited by the\nPlatforms raise minimal concerns. The Platforms argue that\nthe Ordinance chills commercial speech, namely,\nadvertisements for third-party rentals. But even accepting\n\f HOMEAWAY.COM V. CITY OF SANTA MONICA 21\n\nthat the Platforms will need to engage in efforts to validate\ntransactions before completing them, incidental burdens like\nthese are not always sufficient to trigger First Amendment\nscrutiny. See Int’l Franchise Ass’n, 803 F.3d at 408\n(“[S]ubjecting every incidental impact on speech to First\nAmendment scrutiny ‘would lead to the absurd result that\nany government action that had some conceivable speech\ninhibiting consequences . . . would require analysis under\nthe First Amendment.’” (quoting Arcara, 478 U.S. at 708\n(O’Connor, J., concurring))). Furthermore, to the extent that\nthe speech chilled advertises unlawful rentals, “[a]ny First\nAmendment interest . . . is altogether absent when the\ncommercial activity itself is illegal and the restriction on\nadvertising is incidental to a valid limitation on economic\nactivity.” See Pittsburgh Press Co. v. Pittsburgh Comm’n\non Human Relations, 413 U.S. 376, 389 (1973).\n\n Finally, because the Ordinance does not implicate speech\nprotected by the First Amendment, we similarly reject the\nPlatforms’ argument that the Ordinance is unconstitutional\nwithout a scienter requirement. In most cases, there is no\n“closed definition” on when a criminal statute must contain\na scienter requirement. See Morissette v. United States,\n342 U.S. 246, 260 (1952). However, the Supreme Court has\ndrawn a bright line in certain contexts, such as holding that\nthe First Amendment requires statutes imposing criminal\nliability for obscenity or child pornography to contain a\nscienter requirement. See New York v. Ferber, 458 U.S. 747,\n765 (1982). Such a requirement prevents “a severe\nlimitation on the public’s access to constitutionally protected\nmatter” as would result from inflexible laws criminalizing\n“bookshops and periodical stands.” Smith v. California,\n361 U.S. 147, 153 (1959).\n\f22 HOMEAWAY.COM V. CITY OF SANTA MONICA\n\n Here, even assuming that the Ordinance would lead the\nPlatforms to voluntarily remove some advertisements for\nlawful rentals, there would not be a “severe limitation on the\npublic’s access” to lawful advertisements, especially\nconsidering the existence of alternative channels like\nCraigslist. Id. Such an incidental burden is far from “a\nsubstantial restriction on the freedom of speech” that would\nnecessitate a scienter requirement. Id. at 150. Otherwise,\n“[t]here is no specific constitutional inhibition against\nmaking the distributors of good[s] the strictest censors of\ntheir merchandise.” Id. at 152.\n\n III. Remaining Claims\n\n On appeal, the Platforms do not challenge dismissal of\ntheir other federal law claims “in light of the district court’s\ninterpretation of the Ordinance as only requiring disclosure\nof information pursuant to requests that comply with the\nFourth Amendment and Stored Communications Act.”\nSimilarly, the parties specified that they would “not\nchallenge the district court’s decision to decline\nsupplemental jurisdiction if all the Platforms’ federal claims\nwere properly dismissed.” Accordingly, we need not\nconsider the remaining claims.\n\n * * *\n\n Because the district court properly dismissed the\nPlatforms’ complaints for failure to state a claim, we dismiss\nas moot the appeals from the denial of preliminary injunctive\nrelief.\n\n AFFIRMED in part, DISMISSED in part.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: homeaway.com, Inc. v. City of Santa Monica\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-13\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nHOMEAWAY.COM, INC.; AIRBNB No. 18-55367\\nINC.,\\n Plaintiffs-Appellants, D.C. Nos.\\n 2:16-cv-06641-\\n v. ODW-AFM\\n 2:16-cv-06645-\\nCITY OF SANTA MONICA, ODW-AFM\\n Defendant-Appellee.\\n\\n\\n\\nHOMEAWAY.COM, INC., No. 18-55805\\n Plaintiff-Appellant,\\n D.C. Nos.\\n and 2:16-cv-06641-\\n ODW-AFM\\nAIRBNB INC., 2:16-cv-06645-\\n Plaintiff, ODW-AFM\\n\\n v.\\n\\nCITY OF SANTA MONICA,\\n Defendant-Appellee.\\n\\f2 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\n\\n AIRBNB INC., No. 18-55806\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:16-cv-06645-\\n ODW-AFM\\n CITY OF SANTA MONICA,\\n Defendant-Appellee. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Otis D. Wright II, District Judge, Presiding\\n\\n Argued and Submitted October 12, 2018\\n Pasadena, California\\n\\n Filed March 13, 2019\\n\\n Before: Mary M. Schroeder and Jacqueline H. Nguyen,\\n Circuit Judges, and Michael H. Simon, * District Judge.\\n\\n Opinion by Judge Nguyen\\n\\n\\n\\n\\n *\\n The Honorable Michael H. Simon, United States District Judge for\\nthe District of Oregon, sitting by designation.\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 3\\n\\n SUMMARY **\\n\\n\\n Civil Rights\\n\\n The panel affirmed the district court’s dismissal, for\\nfailure to state a claim, of a complaint brought by\\nHomeAway.com and Airbnb Inc. challenging the City of\\nSanta Monica’s Ordinance 2535, which imposes various\\nobligations on companies that host online platforms for\\nshort-term vacation rentals.\\n\\n The Ordinance, as amended in 2017, imposes four\\nobligations on hosting platforms: (1) collecting and\\nremitting Transient Occupancy Taxes; (2) regularly\\ndisclosing listings and booking information to the City; (3)\\nrefraining from booking properties not licensed and listed on\\nthe City’s registry; (4) and refraining from collecting a fee\\nfor ancillary services.\\n\\n The panel rejected plaintiffs’ assertion that the\\nOrdinance violated the Communications Decency Act of\\n1996, 47 U.S.C. § 230 because it required them to monitor\\nand remove third-party content, and therefore interfered with\\nfederal policy protecting internet companies from liability\\nfor posting third-party content. The panel stated that the\\nOrdinance prohibits processing transactions for unregistered\\nproperties. It does not require the Platforms to review the\\ncontent provided by the hosts of listings on their websites.\\nRather, the panel noted that the only monitoring that\\nappeared necessary in order to comply with the Ordinance\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f4 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nrelated to incoming requests to complete a booking\\ntransaction—content that, while resulting from the third-\\nparty listings, was distinct, internal, and nonpublic. The\\npanel concluded that the Ordinance was not inconsistent\\nwith the Communications Decency Act, and therefore was\\nnot expressly preempted by its terms. The panel further\\nconcluded that the Ordinance would not pose an obstacle to\\nCongress’s aim to encourage self-monitoring of third-party\\ncontent, and therefore obstacle preemption did not preclude\\nSanta Monica from enforcing the Ordinance.\\n\\n The panel held that the Ordinance did not implicate\\nspeech protected by the First Amendment, concluding that\\nthe Ordinance’s prohibitions regulate nonexpressive\\nconduct, specifically booking transactions, and do not single\\nout those engaged in expressive activity.\\n\\n\\n COUNSEL\\n\\nDonald B. Verrilli, Jr. (argued) and Chad Golder, Munger,\\nTolles & Olson LLP, Washington, D.C.; Joseph W. Cotchett\\nand Alexandra P. Summer, Cotchett, Pitre & McCarthy,\\nLLP, Santa Monica, California; Jonathan H. Blavin and\\nJoshua Patashnik, Munger, Tolles & Olson LLP, San\\nFrancisco, California; John B. Major, Munger, Tolles &\\nOlson, Los Angeles, California; for Plaintiff-Appellant\\nAirbnb, Inc.\\n\\nStephen M. Rummage and Ambika K. Doran, Davis Wright\\nTremaine LLP, Seattle, Washington, for Plaintiff-Appellant\\nHomeAway.com, Inc.\\n\\nGeorge S. Cardona (argued), Deputy City Attorney; Lane\\nDilg, City Attorney; Yibin Shen, Chief Deputy City\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 5\\n\\nAttorney; Heidi Von Tongeln and Michael R. Cobden,\\nDeputy City Attorneys; Santa Monica City Attorney’s\\nOffice, Santa Monica, California, for Defendant-Appellee.\\n\\nDavid Salmons, Bryan Killian, and James Nelson, Morgan,\\nLewis & Bockius LLP, Washington, D.C., for Amici Curiae\\nChris Cox and NetChoice.\\n\\nCatherine R. Gellis, Esq., Sausalito, California, for Amicus\\nCuriae Floor64, Inc. d/b/a The Copia Institute.\\n\\nIan C. Ballon and Lori Chang, Greenberg Traurig, LLP, Los\\nAngeles, California, for Amici Curiae Ebay Inc., Glassdoor,\\nInc., Lyft, Inc., Offerup, Inc., Taskrabbit, Inc., Thumbtack,\\nInc., Uber Technologies, Inc., and Upwork, Inc.\\n\\nChristi Hogin, Best Best & Krieger LLP, Manhattan Beach,\\nCalifornia, for Amici Curiae League of California Cities,\\nInternational Municipal Lawyers Association, and\\nCalifornia State Association of Counties.\\n\\nJeremy B. Rosen, Eric S. Boorstin, and Ryan C. Chapman,\\nHorvitz & Levy LLP, Burbank, California; Michael T.\\nWilliams and Allison R. Mclaughlin, Wheeler Trigg\\nO’Donnell LLP, Denver, Colorado; David C. Frederick,\\nBrendan J. Crimmins, and Rachel P. May, Kellogg, Hansen,\\nTodd, Figel & Frederick, P.L.L.C., Washington, D.C., for\\nAmicus Curiae Apartment Investment and Management\\nCompany.\\n\\nHeidi Palutke, Sacramento, California, for Amicus Curiae\\nCalifornia Apartment Association.\\n\\nEdward M. Schulman, Ballston Tower, Arlington, Virginia,\\nfor Amicus Curiae AvalonBay Communities, Inc.\\n\\f6 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nGary S. Kessler, Kulik Gottesman Siegel & Ware LLP,\\nSherman Oaks, California, for Amicus Curiae Community\\nAssociations Institute.\\n\\nAbbey R. Stemler and Matthew C. Turk, Indiana University\\nKelley School of Business, Bloomington, Indiana; Jahan C.\\nSagafi and Relic Sun, Outten & Golden LLP, San Francisco,\\nCalifornia; Peter Romer-Friedman, Outten & Golden LLP,\\nWashington, D.C., for Amici Curiae Internet, Business, and\\nLocal Government Law Professors.\\n\\nRichard G. McCracken and Paul L. More, McCracken,\\nStemerman & Holsberry, LLP, San Francisco, California,\\nfor Amicus Curiae UNITE HERE International Union.\\n\\nDennis J. Herrera, San Francisco City Attorney; Christine\\nVan Aken, Chief of Appellate Litigation; Yvonne R. Meré,\\nChief of Complex and Affirmative Litigation; Sara J.\\nEisenberg, Deputy City Attorney; San Francisco, California\\nfor Amicus Curiae City and County of San Francisco.\\n\\nKarl A. Racine, Attorney General for the District of\\nColumbia, Washington, D.C., for Amicus Curiae District of\\nColumbia.\\n\\nAndre M. Davis, City Solicitor, City of Baltimore,\\nBaltimore, Maryland, for Amicus Curiae Mayor and City\\nCouncil of Baltimore.\\n\\nZach Klein, Columbus City Attorney, Columbus, Ohio, for\\nAmicus Curiae City of Columbus.\\n\\nBarbara J. Parker, Oakland City Attorney; Maria Bee,\\nSpecial Counsel; Erin Bernstein, Supervising Deputy City\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 7\\n\\nAttorney; Oakland, California; for Amicus Curiae City of\\nOakland.\\n\\nPeter S. Holmes, Seattle City Attorney, Seattle, Washington,\\nfor Amicus Curiae City of Seattle.\\n\\nJill E. Habig, Founder & President; Joanna Pearl, Legal\\nDirector; Oakland, California; for Amicus Curiae Public\\nRights Project, A Project of Tides Center.\\n\\n\\n OPINION\\n\\nNGUYEN, Circuit Judge:\\n\\n Located on the coast of Southern California, the city of\\nSanta Monica consists of only about eight square miles but\\nserves 90,000 residents and as many as 500,000 visitors on\\nweekends and holidays. Similar to other popular tourist\\ndestinations, Santa Monica is struggling to manage the\\ndisruptions brought about by the rise of short-term rentals\\nfacilitated by innovative startups such as Appellants\\nHomeAway.com, Inc. and Airbnb Inc. (the “Platforms”).\\nWebsites like those operated by the Platforms are essentially\\nonline marketplaces that allow “guests” seeking\\naccommodations and “hosts” offering accommodations to\\nconnect and enter into rental agreements with one another. 1\\nAs of February 2018, Airbnb had approximately 1,400\\nlistings in Santa Monica, of which about 30 percent are in\\n\\n 1\\n The Platforms do not own, lease, or manage any of the properties\\nlisted on their websites, nor are they parties to the rental agreements.\\nInstead, the content provided alongside the listings—such as description,\\nprice, and availability—are provided by the hosts. For their services, the\\nPlatforms collect a fee from each successful booking.\\n\\f8 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nthe “coastal zone” covered by the California Coastal Act,\\nwhile HomeAway.com had approximately 300 live listings\\nin Santa Monica, of which approximately 40 percent are in\\nthe coastal zone.\\n\\n Santa Monica’s council reported that the proliferation of\\nshort-term rentals had negatively impacted the quality and\\ncharacter of its neighborhoods by “bringing commercial\\nactivity and removing residential housing stock from the\\nmarket” at a time when California is already suffering from\\nsevere housing shortages. In response, the city passed an\\nordinance regulating the short-term vacation rental market\\nby authorizing licensed “home-sharing” (rentals where\\nresidents remain on-site with guests) but prohibiting all other\\nshort-term home rentals of 30 consecutive days or less.\\n\\n The Platforms filed suit, alleging that the city ordinance\\nis preempted by the Communications Decency Act and\\nimpermissibly infringes upon their First Amendment rights.\\nThe district court denied preliminary injunctive relief, and\\ndismissed the Platforms’ complaints for failure to state a\\nclaim under the Communications Decency Act and the First\\nAmendment. We affirm.\\n\\n BACKGROUND\\n\\n In May 2015, Santa Monica passed its initial ordinance\\nregulating the short-term vacation rental market by\\nauthorizing licensed “home-sharing” (rentals where\\nresidents remain on-site with guests) but prohibiting all other\\nforms of short-term rentals for 30 consecutive days or less.\\nSanta Monica Ordinance 2484 (May 12, 2015), codified as\\namended, Santa Monica Mun. Code §§ 6.20.010–6.20.100.\\nThe ordinance reflected the city’s housing goals of\\n“preserving its housing stock and preserving the quality and\\ncharacter of its existing single and multi-family residential\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 9\\n\\nneighborhoods.” Id. As originally enacted, the ordinance\\nprohibited hosting platforms from acting to “undertake,\\nmaintain, authorize, aid, facilitate or advertise any Home-\\nSharing activity” that was not authorized by the city.\\nHosting platforms also were required to collect and remit\\ntaxes, and to regularly disclose listings and booking\\ninformation to the city.\\n\\n The Platforms each filed a complaint in the Central\\nDistrict of California challenging the initial ordinance, and\\nthe district court consolidated the cases for discovery and\\npretrial matters. On September 21, 2016, the parties\\nstipulated to stay the case while the city considered\\namendments to the local ordinance. During the stay period,\\nthe district court for the Northern District of California\\ndenied a preliminary injunction requested by the plaintiffs in\\na separate case challenging a similar ordinance in San\\nFrancisco. See Airbnb Inc. v. City & County of San\\nFrancisco, 217 F. Supp. 3d 1066 (N.D. Cal. 2016). That\\ncase ended in a settlement in which the Platforms agreed to\\ncomply with an amended version of San Francisco’s\\nordinance that prohibited booking unlawful transactions but\\nprovided a safe harbor wherein any platform that complies\\nwith the responsibilities set out in the Ordinance will be\\npresumed to be in compliance with the law.\\n\\n In January 2017, Santa Monica likewise amended its\\nown ordinance. The version challenged here, Ordinance\\n2535 (the “Ordinance”), retains its prohibitions on most\\ntypes of short-term rentals, with the exception of licensed\\nhome-shares. In addition, the Ordinance imposes four\\nobligations on hosting platforms directly: (1) collecting and\\nremitting “Transient Occupancy Taxes,” (2) disclosing\\ncertain listing and booking information regularly,\\n(3) refraining from completing any booking transaction for\\n\\f10 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nproperties not licensed and listed on the City’s registry, and\\n(4) refraining from collecting or receiving a fee for\\n“facilitating or providing services ancillary to a vacation\\nrental or unregistered home-share.” If a housing platform\\noperates in compliance with these obligations, the Ordinance\\nprovides a safe harbor by presuming the platform to be in\\ncompliance with the law. Otherwise, violations are\\npunishable by a fine of up to $500 and/or imprisonment for\\nup to six months.\\n\\n After the district court lifted the stay, the Platforms\\namended their complaint to challenge the revised ordinance\\nand moved for a preliminary injunction. Santa Monica\\nmoved to dismiss the amended complaint. The court denied\\nthe Platforms’ motion for preliminary injunctive relief and\\nsubsequently granted Santa Monica’s motion to dismiss on\\nthe ground that the Platforms failed to state a claim under\\nfederal law, including the Communications Decency Act of\\n1996 and the First Amendment. The district court also\\ndeclined to exercise supplemental jurisdiction over their\\nremaining state-law claims. 2 The Platforms timely appealed\\nthese decisions, and we consolidated the appeals.\\n\\n\\n\\n\\n 2\\n The Platforms do not appeal the district court’s dismissal of other\\nfederal claims brought under the Fourth Amendment and the Stored\\nCommunications Act. Similarly, they do not challenge the court’s\\ndecision not to exercise supplemental jurisdiction over the state-law\\nclaims under the California Coastal Act if we affirm the dismissal of their\\nfederal claims. Because we affirm the district court’s dismissal, we need\\nnot consider the state-law claims. We deny Santa Monica’s motion for\\njudicial notice of its prior enforcement actions because the dispute as to\\nits prior actions relates only to the state-law claims.\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 11\\n\\n JURISDICTION AND STANDARD OF REVIEW\\n\\n We have jurisdiction under 28 U.S.C. § 1291. We\\nreview the district court’s order of dismissal de novo,\\n“accepting all factual allegations in the complaint as true and\\nconstruing them in the light most favorable to the\\nnonmoving party.” Yagman v. Garcetti, 852 F.3d 859, 863\\n(2017) (quoting Ebner v. Fresh, Inc., 838 F.3d 958, 962 (9th\\nCir. 2016)).\\n\\n DISCUSSION\\n\\n I. Communications Decency Act\\n\\n The Communications Decency Act of 1996 (“CDA” or\\nthe “Act”), 47 U.S.C. § 230, provides internet companies\\nwith immunity from certain claims in furtherance of its\\nstated policy “to promote the continued development of the\\nInternet and other interactive computer services.” Id.\\n§ 230(b)(1). Construing this immunity broadly, the\\nPlatforms argue that the Ordinance requires them to monitor\\nand remove third-party content, and therefore violates the\\nCDA by interfering with federal policy protecting internet\\ncompanies from liability for posting third-party content.\\nSanta Monica, on the other hand, argues that the Ordinance\\ndoes not implicate the CDA because it imposes no obligation\\non the Platforms to monitor or edit any listings provided by\\nhosts. Santa Monica contends that the Ordinance is simply\\nan exercise of its right to enact regulations to preserve\\nhousing by curtailing “incentives for landlords to evade rent\\ncontrol laws, evict tenants, and convert residential units into\\nde facto hotels.”\\n\\n We begin our analysis with the text of the CDA. See BP\\nAmerica Prod. Co. v. Burton, 549 U.S. 84, 91 (2006).\\nSection 230(c)(1) states that “[n]o provider or user of an\\n\\f12 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\ninteractive computer service shall be treated as the publisher\\nor speaker of any information provided by another\\ninformation content provider.” Id. § 230(c)(1). The CDA\\nexplicitly preempts inconsistent state laws: “Nothing in this\\nsection shall be construed to prevent any State from\\nenforcing any State law that is consistent with this section.\\nNo cause of action may be brought and no liability may be\\nimposed under any State or local law that is inconsistent with\\nthis section.” Id. § 230(e)(3).\\n\\n We have construed these provisions to extend immunity\\nto “(1) a provider or user of an interactive computer service\\n(2) whom a plaintiff seeks to treat, under a state law cause of\\naction, as a publisher or speaker (3) of information provided\\nby another information content provider.” Barnes v. Yahoo!,\\nInc., 570 F.3d 1096, 1100–01 (9th Cir. 2009). Only the\\nsecond element is at issue here: whether the Ordinance treats\\nthe Platforms as a “publisher or speaker” in a manner that is\\nbarred by the CDA. Although the CDA does not define\\n“publisher,” we have defined “publication” in this context to\\n“involve[] reviewing, editing, and deciding whether to\\npublish or to withdraw from publication third-party\\ncontent.” Id. at 1102 (citing Fair Hous. Council v.\\nRoommates.com, LLC, 521 F.3d 1157, 1170–71 (9th Cir.\\n2008) (en banc)).\\n\\n The Platforms offer two different theories as to how the\\nOrdinance in fact reaches “publication” activities. First, the\\nPlatforms claim that the Ordinance is expressly preempted\\nby the CDA because, as they argue, it implicitly requires\\nthem “to monitor the content of a third-party listing and\\ncompare it against the City’s short-term rental registry\\nbefore allowing any booking to proceed.” Relying on Doe\\nv. Internet Brands, 824 F.3d 846, 851 (9th Cir. 2016), the\\nPlatforms take the view that CDA immunity follows\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 13\\n\\nwhenever a legal duty “affects” how an internet company\\n“monitors” a website.\\n\\n However, the Platforms read Internet Brands too\\nbroadly. In that case, two individuals used the defendant’s\\nwebsite to message and lure the plaintiff to sham auditions\\nwhere she was drugged and raped. Id. at 848. We held that,\\nwhere the website provider was alleged to have known\\nindependently of the ongoing scheme beforehand, the CDA\\ndid not bar an action under state law for failure to warn. Id.\\nat 854. We observed that a duty to warn would not\\n“otherwise affect how [the defendant] publishes or\\nmonitors” user content. Id. at 851. Though the defendant\\ndid, in its business, act as a publisher of third-party content,\\nthe underlying legal duty at issue did not seek to hold the\\ndefendant liable as a “publisher or speaker” of third-party\\ncontent. Id. at 853; see 47 U.S.C. § 230(c)(1). We therefore\\ndeclined to extend CDA immunity to the defendant for the\\nplaintiff’s failure-to-warn claim. Internet Brands, 824 F.3d\\nat 854.\\n\\n We do not read Internet Brands to suggest that CDA\\nimmunity attaches any time a legal duty might lead a\\ncompany to respond with monitoring or other publication\\nactivities. It is not enough that third-party content is\\ninvolved; Internet Brands rejected use of a “but-for” test that\\nwould provide immunity under the CDA solely because a\\ncause of action would not otherwise have accrued but for the\\nthird-party content. Id. at 853. We look instead to what the\\nduty at issue actually requires: specifically, whether the duty\\nwould necessarily require an internet company to monitor\\nthird-party content. See id. at 851, 853.\\n\\n Here, the Ordinance does not require the Platforms to\\nmonitor third-party content and thus falls outside of the\\nCDA’s immunity. The Ordinance prohibits processing\\n\\f14 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\ntransactions for unregistered properties. It does not require\\nthe Platforms to review the content provided by the hosts of\\nlistings on their websites. Rather, the only monitoring that\\nappears necessary in order to comply with the Ordinance\\nrelates to incoming requests to complete a booking\\ntransaction—content that, while resulting from the third-\\nparty listings, is distinct, internal, and nonpublic. As in\\nInternet Brands, it is not enough that the third-party listings\\nare a “but-for” cause of such internal monitoring. See\\n824 F.3d at 853. The text of the CDA is “clear that neither\\nthis subsection nor any other declares a general immunity\\nfrom liability deriving from third-party content.” Barnes,\\n570 F.3d at 1100. To provide broad immunity “every time a\\nwebsite uses data initially obtained from third parties would\\neviscerate [the CDA].” Barnes, 570 F.3d at 1100 (quoting\\nRoommates.com, 521 F.3d at 1171 (9th Cir. 2008) (en\\nbanc)). That is not the result that Congress intended.\\n\\n Nor could a duty to cross-reference bookings against\\nSanta Monica’s property registry give rise to CDA\\nimmunity. While keeping track of the city’s registry is\\n“monitoring” third-party content in the most basic sense,\\nsuch conduct cannot be fairly classified as “publication” of\\nthird-party content. The Platforms have no editorial control\\nover the registry whatsoever. As with tax regulations or\\ncriminal statutes, the Ordinance can fairly charge parties\\nwith keeping abreast of the law without running afoul of the\\nCDA.\\n\\n Second, the Platforms argue that the Ordinance “in\\noperation and effect . . . forces [them] to remove third-party\\ncontent.” Although it is clear that the Ordinance does not\\nexpressly mandate that they do so, the Platforms claim that\\n“common sense explains” that they cannot “leave in place a\\nwebsite chock-full of un-bookable listings.” For purposes of\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 15\\n\\nour review, we accept at face value the Platforms’ assertion\\nthat they will choose to remove noncompliant third-party\\nlistings on their website as a consequence of the Ordinance. 3\\nNonetheless, their choice to remove listings is insufficient to\\nimplicate the CDA.\\n\\n On its face, the Ordinance does not proscribe, mandate,\\nor even discuss the content of the listings that the Platforms\\ndisplay on their websites. See Santa Monica Mun. Code\\n§§ 6.20.010–6.20.100. It requires only that transactions\\ninvolve licensed properties. We acknowledge that, as the\\nPlatforms explain in Airbnb’s complaint and in the briefing\\non appeal, removal of these listings would be the best option\\n“from a business standpoint.” But, as in Internet Brands, the\\nunderlying duty “could have been satisfied without changes\\nto content posted by the website’s users.” See 824 F.3d\\nat 851. Even assuming that removing certain listings may be\\nthe Platforms’ most practical compliance option, allowing\\ninternet companies to claim CDA immunity under these\\ncircumstances would risk exempting them from most local\\nregulations and would, as this court feared in\\nRoommates.com, 521 F.3d at 1164, “create a lawless no-\\nman’s-land on the Internet.” We hold that the Ordinance is\\nnot “inconsistent” with the CDA, and is therefore not\\nexpressly preempted by its terms. See 47 U.S.C. § 230(e)(3).\\n\\n Finally, the Platforms argue that, even if the Ordinance\\nis not expressly preempted by the CDA, the Ordinance\\nimposes “an obstacle to the accomplishment and execution\\n 3\\n The Platforms argued below that the district court must accept as\\ntrue their allegation that they would “have to” monitor and screen\\nlistings. As a matter of law, the Ordinance does not require them to do\\nso. Courts are “not bound to accept as true a legal conclusion couched\\nas a factual allegation.” Bell Atlantic Corp. v. Twombly, 550 U.S. 544,\\n555 (2007) (quoting Papasan v. Allain, 478 U.S. 265, 286 (1986)).\\n\\f16 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nof the full purposes and objectives of Congress.” See Crosby\\nv. Nat’l Foreign Trade Council, 530 U.S. 363, 372–73\\n(2000). Reading the CDA expansively, they argue that the\\nOrdinance conflicts with the CDA’s goal “to preserve the\\nvibrant and competitive free market that presently exists for\\nthe Internet . . . unfettered by Federal or State regulation.”\\nSee § 230(b)(2). We have consistently eschewed an\\nexpansive reading of the statute that would render unlawful\\nconduct “magically . . . lawful when [conducted] online,”\\nand therefore “giv[ing] online businesses an unfair\\nadvantage over their real-world counterparts.” See\\nRoommates.com, 521 F.3d at 1164, 1164–65 n.15. For the\\nsame reasons, while we acknowledge the Platforms’\\nconcerns about the difficulties of complying with numerous\\nstate and local regulations, the CDA does not provide\\ninternet companies with a one-size-fits-all body of law. Like\\ntheir brick-and-mortar counterparts, internet companies\\nmust also comply with any number of local regulations\\nconcerning, for example, employment, tax, or zoning.\\nBecause the Ordinance would not pose an obstacle to\\nCongress’s aim to encourage self-monitoring of third-party\\ncontent, we hold that obstacle preemption does not preclude\\nSanta Monica from enforcing the Ordinance.\\n\\n Fundamentally, the parties dispute how broadly to\\nconstrue the CDA so as to continue serving the purposes\\nCongress envisioned while allowing state and local\\ngovernments breathing room to address the pressing issues\\nfaced by their communities. We have previously\\nacknowledged that the CDA’s immunity reaches beyond the\\ninitial state court decision that sparked its enactment. See\\nFair Hous. Council v. Roommates.com, LLC, 521 F.3d 1157,\\n1163 (9th Cir. 2008) (en banc) (discussing Stratton\\nOakmont, Inc. v. Prodigy Servs. Co., which held an internet\\ncompany liable for defamation when it removed some, but\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 17\\n\\nnot all, harmful content from its public message boards,\\n1995 WL 323710 (N.Y. Sup. Ct. May 24, 1995)\\n(unpublished)). As the Platforms correctly note, the Act’s\\npolicy statements broadly promote “the vibrant and\\ncompetitive free market that presently exists for the Internet\\n. . . unfettered by Federal or State regulation.” See 47 U.S.C.\\n§ 230(b)(2). “[A] law’s scope often differs from its genesis,”\\nand we have repeatedly held the scope of immunity to reach\\nbeyond defamation cases. Barnes, 570 F.3d at 1101 (quoting\\nChicago Lawyers’ Comm. for Civil Rights Under Law, Inc.\\nv. Craigslist, Inc., 519 F.3d 666, 671 (7th Cir. 2008), as\\namended (May 2, 2008)) (citing cases applying immunity for\\ncauses of action including discrimination, fraud, and\\nnegligence).\\n\\n At the same time, our cases have hewn closely to the\\nstatutory language of the CDA and have limited the\\nexpansion of its immunity beyond the protection Congress\\nenvisioned. As we have observed, “the [relevant] section is\\ntitled ‘Protection for “good Samaritan” blocking and\\nscreening of offensive material.’” Roommates.com,\\n521 F.3d at 1163–64 (quoting 47 U.S.C. § 230(c)); see also\\nInternet Brands, 824 F.3d at 852. Congress intended to\\n“spare interactive computer services [the] grim choice”\\nbetween voluntarily filtering content and being subject to\\nliability on the one hand, and “ignoring all problematic posts\\naltogether [to] escape liability.” Roommates.com, 521 F.3d\\nat 1163–64. In contrast, the Platforms face no liability for\\nthe content of the bookings; rather, any liability arises only\\nfrom unlicensed bookings. We do not discount the\\nPlatforms’ concerns about the administrative burdens of\\nstate and local regulations, but we nonetheless disagree that\\n§ 230(c)(1) of the CDA may be read as broadly as they\\nadvocate, or that we may ourselves expand its provisions\\nbeyond what Congress initially intended.\\n\\f18 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\n In sum, neither express preemption nor obstacle\\npreemption apply to the Ordinance. We therefore affirm the\\ndistrict court’s dismissal for failure to state a claim under the\\nCDA.\\n\\n II. First Amendment\\n\\n The Platforms also contend that the district court erred in\\ndismissing their First Amendment claims. They argue that,\\neven if the plain language of the Ordinance only reaches\\n“conduct,” i.e., booking unlicensed properties, the law\\neffectively imposes a “content-based financial burden” on\\ncommercial speech and is thus subject to First Amendment\\nscrutiny. The district court concluded that the Ordinance\\n“regulates conduct, not speech, and that the conduct banned\\n. . . does not have such a ‘significant expressive element’ as\\nto draw First Amendment protection.” We agree.\\n\\n That the Ordinance regulates “conduct” is not alone\\ndispositive. The Supreme Court has previously applied First\\nAmendment scrutiny when “‘speech’ and ‘nonspeech’\\nelements are combined in the same course of conduct.” See\\nUnited States v. O’Brien, 391 U.S. 367, 376 (1968). But\\n“restrictions on protected expression are distinct from\\nrestrictions on economic activity or, more generally, on\\nnonexpressive conduct.” Sorrell v. IMS Health Inc.,\\n564 U.S. 552, 567 (2011). While the former is entitled to\\nprotection, “the First Amendment does not prevent\\nrestrictions directed at commerce or conduct from imposing\\nincidental burdens on speech.” Id.\\n\\n To determine whether the First Amendment applies, we\\nmust first ask the “threshold question [of] whether conduct\\nwith a ‘significant expressive element’ drew the legal\\nremedy or the ordinance has the inevitable effect of ‘singling\\nout those engaged in expressive activity.’” Int’l Franchise\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 19\\n\\nAss’n v. City of Seattle, 803 F.3d 389, 408 (9th Cir. 2015)\\n(quoting Arcara v. Cloud Books, Inc., 478 U.S. 697, 706–07\\n(1986)). A court may consider the “inevitable effect of a\\nstatute on its face,” as well as a statute’s “stated purpose.”\\nSorrell, 564 U.S. at 565. However, absent narrow\\ncircumstances, a court may not conduct an inquiry into\\nlegislative purpose or motive beyond what is stated within\\nthe statute itself. See O’Brien, 391 U.S. at 383 n.30.\\nBecause the conduct at issue—completing booking\\ntransactions for unlawful rentals—consists only of\\nnonspeech, nonexpressive conduct, we hold that the\\nOrdinance does not implicate the First Amendment.\\n\\n First, the prohibitions here did not target conduct with “a\\nsignificant expressive element.” See Arcara, 478 U.S.\\nat 706. Our decision in International Franchise Ass’n is\\nanalogous. There, the plaintiff challenged a minimum wage\\nordinance that would have accelerated the raising of the\\nminimum wage to $15 per hour for franchise owners and\\nother large employers. 803 F.3d at 389. In denying a\\npreliminary injunction, the district court held that the\\nplaintiffs were not likely to succeed on their First\\nAmendment argument that the ordinance treated them\\ndifferently based on their “speech and association” decisions\\nto operate within a franchise relationship framework. Id. at\\n408–09. We agreed, concluding that the “business\\nagreement or business dealings” were not conduct with a\\n“significant expressive element.” Id. at 408. Instead,\\n“Seattle’s minimum wage ordinance [was] plainly an\\neconomic regulation that [did] not target speech or\\nexpressive conduct.” Id.\\n\\n Similarly, here, the Ordinance is plainly a housing and\\nrental regulation. The “inevitable effect of the [Ordinance]\\non its face” is to regulate nonexpressive conduct—namely,\\n\\f20 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\nbooking transactions—not speech. See Sorrell, 564 U.S.\\nat 565. As in International Franchise Ass’n, the “business\\nagreement or business dealings” associated with processing\\na booking is not conduct with a “significant expressive\\nelement.” See 803 F.3d at 408 (citation and quotation marks\\nomitted). Contrary to the Platforms’ claim, the Ordinance\\ndoes not “require” that they monitor or screen\\nadvertisements. It instead leaves them to decide how best to\\ncomply with the prohibition on booking unlawful\\ntransactions.\\n\\n Nor can the Platforms rely on the Ordinance’s “stated\\npurpose” to argue that it intends to regulate speech. The\\nOrdinance itself makes clear that the City’s “central and\\nsignificant goal . . . is preservation of its housing stock and\\npreserving the quality and nature of residential\\nneighborhoods.” As such, with respect to the Platforms, the\\nonly inevitable effect, and the stated purpose, of the\\nOrdinance is to prohibit them from completing booking\\ntransactions for unlawful rentals.\\n\\n As for the second prong of our inquiry, whether the\\nOrdinance has the effect of “singling out those engaged in\\nexpressive activity,” Arcara, 478 U.S. at 706–07, we\\nconclude that it does not. As the Platforms point out,\\nwebsites like Craigslist “advertise the very same properties,”\\nbut do not process transactions. Unlike the Platforms, those\\nwebsites would not be subject to the Ordinance,\\nunderscoring that the Ordinance does not target websites that\\npost listings, but rather companies that engage in unlawful\\nbooking transactions.\\n\\n Moreover, the incidental impacts on speech cited by the\\nPlatforms raise minimal concerns. The Platforms argue that\\nthe Ordinance chills commercial speech, namely,\\nadvertisements for third-party rentals. But even accepting\\n\\f HOMEAWAY.COM V. CITY OF SANTA MONICA 21\\n\\nthat the Platforms will need to engage in efforts to validate\\ntransactions before completing them, incidental burdens like\\nthese are not always sufficient to trigger First Amendment\\nscrutiny. See Int’l Franchise Ass’n, 803 F.3d at 408\\n(“[S]ubjecting every incidental impact on speech to First\\nAmendment scrutiny ‘would lead to the absurd result that\\nany government action that had some conceivable speech\\ninhibiting consequences . . . would require analysis under\\nthe First Amendment.’” (quoting Arcara, 478 U.S. at 708\\n(O’Connor, J., concurring))). Furthermore, to the extent that\\nthe speech chilled advertises unlawful rentals, “[a]ny First\\nAmendment interest . . . is altogether absent when the\\ncommercial activity itself is illegal and the restriction on\\nadvertising is incidental to a valid limitation on economic\\nactivity.” See Pittsburgh Press Co. v. Pittsburgh Comm’n\\non Human Relations, 413 U.S. 376, 389 (1973).\\n\\n Finally, because the Ordinance does not implicate speech\\nprotected by the First Amendment, we similarly reject the\\nPlatforms’ argument that the Ordinance is unconstitutional\\nwithout a scienter requirement. In most cases, there is no\\n“closed definition” on when a criminal statute must contain\\na scienter requirement. See Morissette v. United States,\\n342 U.S. 246, 260 (1952). However, the Supreme Court has\\ndrawn a bright line in certain contexts, such as holding that\\nthe First Amendment requires statutes imposing criminal\\nliability for obscenity or child pornography to contain a\\nscienter requirement. See New York v. Ferber, 458 U.S. 747,\\n765 (1982). Such a requirement prevents “a severe\\nlimitation on the public’s access to constitutionally protected\\nmatter” as would result from inflexible laws criminalizing\\n“bookshops and periodical stands.” Smith v. California,\\n361 U.S. 147, 153 (1959).\\n\\f22 HOMEAWAY.COM V. CITY OF SANTA MONICA\\n\\n Here, even assuming that the Ordinance would lead the\\nPlatforms to voluntarily remove some advertisements for\\nlawful rentals, there would not be a “severe limitation on the\\npublic’s access” to lawful advertisements, especially\\nconsidering the existence of alternative channels like\\nCraigslist. Id. Such an incidental burden is far from “a\\nsubstantial restriction on the freedom of speech” that would\\nnecessitate a scienter requirement. Id. at 150. Otherwise,\\n“[t]here is no specific constitutional inhibition against\\nmaking the distributors of good[s] the strictest censors of\\ntheir merchandise.” Id. at 152.\\n\\n III. Remaining Claims\\n\\n On appeal, the Platforms do not challenge dismissal of\\ntheir other federal law claims “in light of the district court’s\\ninterpretation of the Ordinance as only requiring disclosure\\nof information pursuant to requests that comply with the\\nFourth Amendment and Stored Communications Act.”\\nSimilarly, the parties specified that they would “not\\nchallenge the district court’s decision to decline\\nsupplemental jurisdiction if all the Platforms’ federal claims\\nwere properly dismissed.” Accordingly, we need not\\nconsider the remaining claims.\\n\\n * * *\\n\\n Because the district court properly dismissed the\\nPlatforms’ complaints for failure to state a claim, we dismiss\\nas moot the appeals from the denial of preliminary injunctive\\nrelief.\\n\\n AFFIRMED in part, DISMISSED in part.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1288,"cells":{"id":{"kind":"number","value":9022991,"string":"9,022,991"},"name":{"kind":"string","value":"ADVANCED BUILDING & FABRICATION, INC., a California Corporation Robert Honan v. CALIFORNIA HIGHWAY PATROL John Wilson, and Curtis Ayers"},"name_abbreviation":{"kind":"string","value":"Advanced Bldg. & Fabrication, Inc. v. Cal. Highway Patrol"},"decision_date":{"kind":"string","value":"2019-03-13"},"docket_number":{"kind":"string","value":"No. 17-16618"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Nguyen\", \"parties\": \"\", \"opinions\": [{\"author\": \"\", \"type\": \"020lead\", \"text\": \"\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/9016229/\", \"author_raw\": \"\"}]}"},"judge1":{"kind":"string","value":"NGUYEN"},"judge2":{"kind":"string","value":"NGUYEN"},"judge3":{"kind":"string","value":""},"exactly_three_caps":{"kind":"number","value":0,"string":"0"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/9022991/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":0,"string":"0"},"020lead_count":{"kind":"number","value":1,"string":"1"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: ADVANCED BUILDING & FABRICATION, INC., a California Corporation Robert Honan v. CALIFORNIA HIGHWAY PATROL John Wilson, and Curtis Ayers\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-13\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\n\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: ADVANCED BUILDING & FABRICATION, INC., a California Corporation Robert Honan v. CALIFORNIA HIGHWAY PATROL John Wilson, and Curtis Ayers\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-13\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\n\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1289,"cells":{"id":{"kind":"number","value":4603416,"string":"4,603,416"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. Sha-Ron Haines"},"decision_date":{"kind":"string","value":"2019-03-14"},"docket_number":{"kind":"string","value":"17-10059"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: J. Clifford Wallace and Michelle T. Friedland, Circuit Judges, and Lynn S. Adelman, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"ADELMAN, District Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-10059\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 2:14-cr-00264-APG-\\n VCF-2\\nSHA-RON HAINES,\\n Defendant-Appellant. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the District of Nevada\\n Andrew P. Gordon, District Judge, Presiding\\n\\n Argued and Submitted January 17, 2019\\n San Francisco, California\\n\\n Filed March 14, 2019\\n\\n Before: J. Clifford Wallace and Michelle T. Friedland,\\n Circuit Judges, and Lynn S. Adelman, * District Judge.\\n\\n Opinion by Judge Adelman\\n\\n\\n\\n\\n *\\n The Honorable Lynn S. Adelman, United States District Judge for\\nthe Eastern District of Wisconsin, sitting by designation.\\n\\f2 UNITED STATES V. HAINES\\n\\n SUMMARY **\\n\\n\\n Criminal Law\\n\\n The panel affirmed the district court’s judgment in a case\\nin which the defendant was convicted of conspiracy to\\ncommit sex trafficking of a minor, sex trafficking of a minor,\\nconspiracy to transport a minor to engage in prostitution, and\\ntransporting a minor to engage in prostitution.\\n\\n At trial, the defendant sought to question minor J.C.\\nabout her prior prostitution activities (which apparently did\\nnot involve a pimp), arguing that this evidence was relevant\\nto, among other things, whether he recruited or encouraged\\nher to engage in prostitution on this occasion. The panel held\\nthat the district court did not err by excluding the testimony\\nunder Fed. R. Evid. 412, the “rape shield” rule. The panel\\nrejected the defendant’s contention that evidence of J.C.’s\\nprior prostitution activities should have been admitted under\\nthe exception in Rule 412(b)(1)(C) for “evidence whose\\nexclusion would violate [his] constitutional rights”—here,\\nhis due process right to present a complete defense and his\\nSixth Amendment right to confront witnesses. The panel\\nsaw no reason to depart from persuasive authorities holding\\nthat a defense such as the one the defendant sought to\\npresent—that he had no intent to, and did not, pimp out\\nJ.C.—triggers the exception. The panel also held that the\\napplicability of Rule 412 should not depend on the alleged\\nvictim’s desire to testify. The panel concluded that even if\\nthe district court misapplied Rule 412, any error would be\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f UNITED STATES V. HAINES 3\\n\\nharmless. The panel held that the defendant’s arguments that\\nthe government opened the door to testimony about J.C.’s\\nprior activities lacked merit.\\n\\n The panel addressed other arguments in a separate\\nmemorandum disposition.\\n\\n\\n COUNSEL\\n\\nKaren A. Connolly (argued), Karen A. Connolly, Ltd., Las\\nVegas, Nevada, for Defendant-Appellant.\\n\\nVijay Shanker (argued), United States Department of\\nJustice, Criminal Division, Appellate Section, Washington\\nD.C., for Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nADELMAN, District Judge:\\n\\n Sha-Ron Haines appeals his convictions for conspiracy\\nto commit sex trafficking of a minor, sex trafficking of a\\nminor, conspiracy to transport a minor to engage in\\nprostitution, and transporting a minor to engage in\\nprostitution. We have jurisdiction under 28 U.S.C. § 1291.\\nWe affirm.\\n\\n I.\\n\\n The government alleged that Haines and his friend Tyral\\nKing transported two minor females, J.C. (age 15) and A.S.\\n(age 17), from Nevada to California to prostitute them, with\\nJ.C. working for Haines and A.S. working for King. The\\n\\f4 UNITED STATES V. HAINES\\n\\ngirls found “dates” by walking a “track” where men picked\\nup prostitutes and through ads posted on a website called\\n“Backpage.com.” King and Haines would drop the girls off\\nat their dates and return to pick them up afterwards.\\n\\n J.C. initially cooperated with the government’s\\ninvestigation, albeit reluctantly. She testified before the\\ngrand jury that ultimately indicted Haines and King that she\\nworked for Haines and gave him the money she earned from\\nprostitution. Prior to trial, however, her account changed.\\nJ.C. then claimed that she initially implicated Haines due to\\npressure from the investigating detective to testify in\\nexchange for release from juvenile detention. This change\\nmay have been prompted by a jailhouse phone call in which\\nHaines advised J.C. to make herself unavailable to testify at\\ntrial, of which the government later found a recording.\\n\\n Whatever the reason, at trial J.C. testified that she\\nworked independently, that she kept her earnings, and that\\nher prior grand jury testimony to the contrary was false. The\\ngovernment impeached J.C. with her previous testimony.\\nSee Fed. R. Evid. 801(d)(1)(A) (authorizing the admission\\nof prior inconsistent statements by testifying witnesses as\\nsubstantive evidence if the prior statements were given under\\noath). The government also presented testimony from King,\\nwho pleaded guilty and agreed to cooperate with the\\ngovernment, that J.C. worked for Haines and gave Haines\\nher prostitution earnings.\\n\\n The jury convicted Haines on all counts. The district\\ncourt sentenced him to 156 months in prison.\\n\\n II.\\n\\n Haines’s defense at trial was that he was merely along\\nfor the ride and did not act as J.C.’s pimp. In support of that\\n\\f UNITED STATES V. HAINES 5\\n\\ndefense, he sought to question J.C. about her prior\\nprostitution activities (which apparently did not involve a\\npimp), arguing that this evidence was relevant to, among\\nother things, whether he recruited or encouraged her to\\nengage in prostitution on this occasion. The district court\\nexcluded the testimony under Federal Rule of Evidence 412,\\nthe “rape shield” rule.\\n\\n We review a district court’s evidentiary rulings for abuse\\nof discretion, though we review de novo the district court’s\\ninterpretation of the Federal Rules of Evidence. United\\nStates v. Kahre, 737 F.3d 554, 565 (9th Cir. 2013). We also\\nreview de novo whether a district court’s evidentiary rulings\\nviolated a defendant’s constitutional rights. United States v.\\nLaursen, 847 F.3d 1026, 1031 (9th Cir. 2017).\\n\\n Rule 412 provides, in pertinent part:\\n\\n (a) Prohibited Uses. The following evidence\\n is not admissible in a civil or criminal\\n proceeding involving alleged sexual\\n misconduct:\\n\\n (1) evidence offered to prove that a victim\\n engaged in other sexual behavior; or\\n\\n (2) evidence offered to prove a victim’s\\n sexual predisposition.\\n\\n (b) Exceptions.\\n\\n (1) Criminal Cases. The court may admit\\n the following evidence in a criminal case:\\n\\n (A) evidence of specific instances of\\n a victim’s sexual behavior, if offered\\n\\f6 UNITED STATES V. HAINES\\n\\n to prove that someone other than the\\n defendant was the source of semen,\\n injury, or other physical evidence;\\n\\n (B) evidence of specific instances of\\n a victim’s sexual behavior with\\n respect to the person accused of the\\n sexual misconduct, if offered by the\\n defendant to prove consent or if\\n offered by the prosecutor; and\\n\\n (C) evidence whose exclusion would\\n violate the defendant’s constitutional\\n rights.\\n\\nFed. R. Evid. 412.\\n\\n The district court correctly determined that the Rule\\napplied. As our sister circuits have noted, sex trafficking\\ncases involve “alleged sexual misconduct,” see United States\\nv. Wardlow, 830 F.3d 817, 820 (8th Cir. 2016) (applying\\nRule 412 in § 2423 prosecution); United States v. Elbert, 561\\nF.3d 771, 776 (8th Cir. 2009) (applying Rule 412 in § 1591\\nprosecution), and evidence of a trafficking victim’s pre- or\\npost-indictment involvement in prostitution implicates her\\n“other sexual behavior” or “sexual predisposition,” see\\nUnited States v. Lockhart, 844 F.3d 501, 509 (5th Cir. 2016).\\nConsistent with this construction of the Rule, courts have\\nroutinely barred evidence of a sex trafficking victim’s other\\nprostitution activities. See, e.g., United States v. Betts, 911\\nF.3d 523, 528–29 (8th Cir. 2018); United States v. Groce,\\n891 F.3d 260, 267–68 (7th Cir. 2018); United States v.\\nCarson, 870 F.3d 584, 593 (7th Cir. 2017), cert. denied, 138\\nS. Ct. 2011 (2018); United States v. Gemma, 818 F.3d 23,\\n34–35 (1st Cir. 2016); Lockhart, 844 F.3d at 510; Elbert, 561\\nF.3d at 777; cf. United States v. Backman, 817 F.3d 662, 670\\n\\f UNITED STATES V. HAINES 7\\n\\n(9th Cir. 2016) (affirming exclusion for failure to comply\\nwith the Rule’s procedural requirements but expressing\\n“doubt that evidence that the victim engaged in commercial\\nsex acts after she had been coerced into prostitution has a\\nbearing on whether Defendant earlier took coercive\\nactions”).\\n\\n Haines argues that evidence of J.C.’s prior prostitution\\nactivities should have been admitted under the exception to\\nRule 412 for “evidence whose exclusion would violate the\\ndefendant’s constitutional rights”—here, his due process\\nright to present a complete defense and his Sixth\\nAmendment right to confront witnesses. But in cases\\ninvolving adult victims forced or coerced into prostitution,\\ncourts have rejected such arguments, concluding that\\nevidence of other prostitution activity has little or no\\nrelevance. Courts have reasoned that just because a victim\\nagreed to engage in sex for money on other occasions does\\nnot mean she consented to, e.g., being beaten or having her\\nearnings confiscated by the defendant. See United States v.\\nRivera, 799 F.3d 180, 185–86 (2d Cir. 2015); United States\\nv. Cephus, 684 F.3d 703, 708 (7th Cir. 2012).\\n\\n Haines argues that his case is different because his\\ndefense was not consent but rather that he had no intent to,\\nand did not, pimp out J.C. However, he makes no attempt to\\ndistinguish Elbert, a case, like his, involving a minor victim\\n(as to which the government need not show force, fraud, or\\ncoercion to prove a violation). The Elbert court rejected the\\nargument that evidence of a child-victim’s other sexual\\nbehavior should be admitted to rebut the allegation that the\\ndefendant recruited the victim to engage in commercial sex\\nacts. The court explained:\\n\\f8 UNITED STATES V. HAINES\\n\\n What Elbert fails to recognize is the evidence\\n he wishes to admit does not provide a defense\\n for the crime with which he was charged and\\n convicted. Elbert repeatedly argues evidence\\n of the victims’ prior acts of prostitution\\n demonstrates he did not cause them to engage\\n in commercial sex acts. This argument relies\\n upon an improper construction of the phrase\\n “caused to engage.” 18 U.S.C. § 1591(a).\\n Because the victims were minors and could\\n not legally consent, the government did not\\n need to prove the elements of fraud, force, or\\n coercion, which are required for adult\\n victims. Id. Instead, the government was only\\n required to prove Elbert knowingly recruited,\\n enticed, harbored, transported, provided, or\\n obtained a minor, knowing the minor would\\n be caused to engage in commercial sex acts.\\n Id. Whether the children engaged in acts of\\n prostitution before or after their encounters\\n with Elbert is irrelevant, and would only\\n prove other people may be guilty of similar\\n offenses of recruiting, enticing, or causing\\n these victims to engage in a commercial sex\\n act.\\n\\n561 F.3d at 777. The court thus rejected Elbert’s argument\\nthat this was “evidence [whose] exclusion . . . would violate\\nthe defendant’s constitutional rights” and thus within the\\nexception contained in Rule 412. Id. at 776–77 (quoting\\nFed. R. Evid. 412(b)(1)(C)).\\n\\n That J.C. may have prostituted on other occasions on her\\nown does not change the result. In United States v.\\nShamsud-Din, No. 10 CR 927, 2011 U.S. Dist. LEXIS\\n\\f UNITED STATES V. HAINES 9\\n\\n124449, at *3 (N.D. Ill. Oct. 27, 2011), the defendant sought\\nto admit evidence that the alleged victims prostituted at other\\ntimes without a pimp. In a passage that applies equally to\\nthis case, the court, ruling pre-trial, stated:\\n\\n Defendant contends that his defense is\\n broader than that in Elbert because his\\n proffered evidence here offers a complete\\n defense to the charges. He specifically asserts\\n that he is arguing that the proffered evidence\\n shows that Victims A and B engaged in the\\n prostitution on their own and without his\\n involvement or knowledge. His argument is a\\n distinction without a difference. Defendant is\\n free to explore whether the Victims engaged\\n in the prostitution activities at issue in this\\n case on their own, rather than with his\\n assistance. He can also question the Victims\\n about their computer skills and whether or\\n not they are familiar with Craigslist, without\\n asking about using it for advertising prior or\\n subsequent prostitution activities. Defendant\\n cannot, however, inquire into other\\n prostitution activities. Such evidence of prior\\n and post prostitution activities is the\\n equivalent of propensity evidence and\\n irrelevant to the charges.\\n\\nId. at *11.\\n\\n Haines cites no case holding that a defense such as the\\none he sought to present here triggers the exception in Rule\\n\\f10 UNITED STATES V. HAINES\\n\\n412. And we see no reason to depart from the persuasive\\nauthorities set forth above that held to the contrary. 1\\n\\n What does make this case somewhat different from those\\ncited above is J.C.’s posture at trial. As Haines notes, Rule\\n412 aims to “safeguard the alleged victim against the\\ninvasion of privacy, potential embarrassment and sexual\\nstereotyping that is associated with public disclosure of\\nintimate sexual details.” See Fed. R. Evid. 412, Advisory\\nCommittee’s Notes to 1994 Amendments (“Advisory\\nCommittee Notes”). Haines contends that in his case J.C.\\nactually wanted to testify about her prior acts of prostitution,\\nand that the government used Rule 412 not as a shield to\\nprotect a cooperative victim’s privacy, but as a sword to\\nobtain a conviction by precluding him from eliciting\\nfavorable testimony from a recalcitrant witness. The parties\\ndo not cite—and we have not found—a case discussing\\napplicability of the Rule to a witness hostile to the\\ngovernment.\\n\\n For several reasons, however, we conclude that the\\napplicability of the Rule should not depend on the alleged\\nvictim’s desire to testify. First, Rule 412 is a rule of\\nexclusion containing three specific exceptions in criminal\\ncases; the victim’s desire to testify or waive the protections\\nof the Rule is not one of them. Second, to the extent that\\nRule 412 also serves the purpose of keeping irrelevant,\\n\\n\\n 1\\n For the same reasons, we reject Haines’s argument that his right to\\npresent a defense was violated when, after the government challenged\\nthe credibility of J.C.’s trial testimony, he was not permitted to\\n“rehabilitate” her with irrelevant and otherwise inadmissible evidence\\nabout her prior sexual behavior. See United States v. Scheffer, 523 U.S.\\n303, 308 (1998) (explaining that the right to present a defense is not\\nunfettered, and law-makers retain broad latitude to establish rules\\nexcluding evidence from criminal trials).\\n\\f UNITED STATES V. HAINES 11\\n\\nprejudicial, and/or inflammatory evidence from the jury, it\\nshould not matter whether the witness wants its protection;\\nthe district court enforces the Rule to ensure that the jury\\ndecides the case based on proper considerations. See\\nAdvisory Committee Notes (noting that the Rule prevents\\n“the infusion of sexual innuendo into the factfinding\\nprocess”); Privacy of Rape Victims: Hearing Before the\\nSubcomm. on Crim. J. of the H. Comm. on the Judiciary on\\nH.R. 14666 and Other Bills, 94th Cong. 41 (1976) (statement\\nof Mary Ann Largen, National Organization of Women)\\n(“[Proposed Rule 412] assures that highly inflammatory and\\narguably irrelevant matters will not be injected.”). Third,\\nallowing the victim’s wishes to control opens the door to\\nmischief. Indeed, in this case the government suspected that\\nJ.C. changed her testimony based on pressure from Haines.\\n\\n Finally, even if the district court misapplied Rule 412\\nhere, and we do not believe it did, any error would be\\nharmless. See United States v. Preston, 873 F.3d 829, 835\\n(9th Cir. 2017) (noting that evidentiary rulings are subject to\\nharmless error review); United States v. Nielson, 371 F.3d\\n574, 581 (9th Cir. 2004) (“Confrontation Clause violations\\nare subject to harmless error analysis[.]”); see also United\\nStates v. Torres, 794 F.3d 1053, 1062–63 (9th Cir. 2015)\\n(holding that an evidentiary ruling excluding evidence\\ndesired by the defense did not violate the right to present a\\ncomplete defense because the evidence would not have\\nsubstantially furthered the defense’s trial theories). Haines\\nwas able to present his theory of the case through J.C.’s\\ntestimony that he was not her pimp during the trip to\\nCalifornia, that she engaged in the charged acts of\\nprostitution on her own and without his encouragement or\\ninvolvement, and that she kept the money she earned.\\nHaines was also able to argue to the jury in closing that this\\ntrip to California was not J.C.’s “first rodeo,” that she knew\\n\\f12 UNITED STATES V. HAINES\\n\\nwhere the track was, and that she knew how to place ads on\\nBackpage. The district court also gave the jury a “mere\\npresence” instruction in support of the defense, explaining\\nthat to convict in this case “[t]he defendant must be a\\nparticipant and not merely a knowing spectator,” which\\nHaines incorporated into his argument. It is hard to see how\\nadditional testimony about J.C.’s other “solo” prostitution-\\nrelated activities would have materially aided the defense.\\nSee United States v. Hofus, 598 F.3d 1171, 1180 (9th Cir.\\n2010) (holding that limitation on testimony from key witness\\ndid not violate the defendant’s ability to present a defense,\\nwhere counsel was still able to argue his theory of the case).\\n\\n We have considered Haines’s additional arguments that\\nthe government opened the door to testimony about J.C.’s\\nprior activities, but none has merit. The government did not\\nelicit testimony suggesting that Haines introduced J.C. to\\nprostitution or that she was, in the district court’s words, “an\\ninnocent lamb led to the slaughter.” The district court took\\nthe issue seriously, warning the government that it could\\nopen the door depending on how it presented its case. We\\nhold there was no abuse of discretion in its rulings.\\n\\n Haines argues that his defense was prejudiced by the\\ndistrict court’s refusal to allow him to recall J.C. to ask her\\nabout her prior prostitution activities after King testified that\\nHaines, A.S., and J.C. came up with the idea to prostitute\\nwhile they were at a pool party together. As the district court\\nnoted in denying Haines’s request, J.C. had already testified\\nthat she and A.S. came up with the idea, without Haines’s\\ninvolvement, and the jury would have to decide who was\\ntelling the truth, J.C. or King. The government did not open\\nthe door simply by presenting evidence contrary to J.C.’s\\ntrial testimony.\\n\\f UNITED STATES V. HAINES 13\\n\\n III.\\n\\n For the foregoing reasons, and those stated in our\\nseparate memorandum disposition addressing Haines’s other\\narguments, we affirm the district court’s judgment.\\n\\n AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4380669/\", \"author_raw\": \"ADELMAN, District Judge:\"}]}"},"judge1":{"kind":"string","value":"J CLIFFORD WALLACE"},"judge2":{"kind":"string","value":"MICHELLE T FRIEDLAND"},"judge3":{"kind":"string","value":"LYNN S ADELMAN"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4603416/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. Sha-Ron Haines\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-14\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 17-10059\n Plaintiff-Appellee,\n D.C. No.\n v. 2:14-cr-00264-APG-\n VCF-2\nSHA-RON HAINES,\n Defendant-Appellant. OPINION\n\n\n\n Appeal from the United States District Court\n for the District of Nevada\n Andrew P. Gordon, District Judge, Presiding\n\n Argued and Submitted January 17, 2019\n San Francisco, California\n\n Filed March 14, 2019\n\n Before: J. Clifford Wallace and Michelle T. Friedland,\n Circuit Judges, and Lynn S. Adelman, * District Judge.\n\n Opinion by Judge Adelman\n\n\n\n\n *\n The Honorable Lynn S. Adelman, United States District Judge for\nthe Eastern District of Wisconsin, sitting by designation.\n\f2 UNITED STATES V. HAINES\n\n SUMMARY **\n\n\n Criminal Law\n\n The panel affirmed the district court’s judgment in a case\nin which the defendant was convicted of conspiracy to\ncommit sex trafficking of a minor, sex trafficking of a minor,\nconspiracy to transport a minor to engage in prostitution, and\ntransporting a minor to engage in prostitution.\n\n At trial, the defendant sought to question minor J.C.\nabout her prior prostitution activities (which apparently did\nnot involve a pimp), arguing that this evidence was relevant\nto, among other things, whether he recruited or encouraged\nher to engage in prostitution on this occasion. The panel held\nthat the district court did not err by excluding the testimony\nunder Fed. R. Evid. 412, the “rape shield” rule. The panel\nrejected the defendant’s contention that evidence of J.C.’s\nprior prostitution activities should have been admitted under\nthe exception in Rule 412(b)(1)(C) for “evidence whose\nexclusion would violate [his] constitutional rights”—here,\nhis due process right to present a complete defense and his\nSixth Amendment right to confront witnesses. The panel\nsaw no reason to depart from persuasive authorities holding\nthat a defense such as the one the defendant sought to\npresent—that he had no intent to, and did not, pimp out\nJ.C.—triggers the exception. The panel also held that the\napplicability of Rule 412 should not depend on the alleged\nvictim’s desire to testify. The panel concluded that even if\nthe district court misapplied Rule 412, any error would be\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f UNITED STATES V. HAINES 3\n\nharmless. The panel held that the defendant’s arguments that\nthe government opened the door to testimony about J.C.’s\nprior activities lacked merit.\n\n The panel addressed other arguments in a separate\nmemorandum disposition.\n\n\n COUNSEL\n\nKaren A. Connolly (argued), Karen A. Connolly, Ltd., Las\nVegas, Nevada, for Defendant-Appellant.\n\nVijay Shanker (argued), United States Department of\nJustice, Criminal Division, Appellate Section, Washington\nD.C., for Plaintiff-Appellee.\n\n\n OPINION\n\nADELMAN, District Judge:\n\n Sha-Ron Haines appeals his convictions for conspiracy\nto commit sex trafficking of a minor, sex trafficking of a\nminor, conspiracy to transport a minor to engage in\nprostitution, and transporting a minor to engage in\nprostitution. We have jurisdiction under 28 U.S.C. § 1291.\nWe affirm.\n\n I.\n\n The government alleged that Haines and his friend Tyral\nKing transported two minor females, J.C. (age 15) and A.S.\n(age 17), from Nevada to California to prostitute them, with\nJ.C. working for Haines and A.S. working for King. The\n\f4 UNITED STATES V. HAINES\n\ngirls found “dates” by walking a “track” where men picked\nup prostitutes and through ads posted on a website called\n“Backpage.com.” King and Haines would drop the girls off\nat their dates and return to pick them up afterwards.\n\n J.C. initially cooperated with the government’s\ninvestigation, albeit reluctantly. She testified before the\ngrand jury that ultimately indicted Haines and King that she\nworked for Haines and gave him the money she earned from\nprostitution. Prior to trial, however, her account changed.\nJ.C. then claimed that she initially implicated Haines due to\npressure from the investigating detective to testify in\nexchange for release from juvenile detention. This change\nmay have been prompted by a jailhouse phone call in which\nHaines advised J.C. to make herself unavailable to testify at\ntrial, of which the government later found a recording.\n\n Whatever the reason, at trial J.C. testified that she\nworked independently, that she kept her earnings, and that\nher prior grand jury testimony to the contrary was false. The\ngovernment impeached J.C. with her previous testimony.\nSee Fed. R. Evid. 801(d)(1)(A) (authorizing the admission\nof prior inconsistent statements by testifying witnesses as\nsubstantive evidence if the prior statements were given under\noath). The government also presented testimony from King,\nwho pleaded guilty and agreed to cooperate with the\ngovernment, that J.C. worked for Haines and gave Haines\nher prostitution earnings.\n\n The jury convicted Haines on all counts. The district\ncourt sentenced him to 156 months in prison.\n\n II.\n\n Haines’s defense at trial was that he was merely along\nfor the ride and did not act as J.C.’s pimp. In support of that\n\f UNITED STATES V. HAINES 5\n\ndefense, he sought to question J.C. about her prior\nprostitution activities (which apparently did not involve a\npimp), arguing that this evidence was relevant to, among\nother things, whether he recruited or encouraged her to\nengage in prostitution on this occasion. The district court\nexcluded the testimony under Federal Rule of Evidence 412,\nthe “rape shield” rule.\n\n We review a district court’s evidentiary rulings for abuse\nof discretion, though we review de novo the district court’s\ninterpretation of the Federal Rules of Evidence. United\nStates v. Kahre, 737 F.3d 554, 565 (9th Cir. 2013). We also\nreview de novo whether a district court’s evidentiary rulings\nviolated a defendant’s constitutional rights. United States v.\nLaursen, 847 F.3d 1026, 1031 (9th Cir. 2017).\n\n Rule 412 provides, in pertinent part:\n\n (a) Prohibited Uses. The following evidence\n is not admissible in a civil or criminal\n proceeding involving alleged sexual\n misconduct:\n\n (1) evidence offered to prove that a victim\n engaged in other sexual behavior; or\n\n (2) evidence offered to prove a victim’s\n sexual predisposition.\n\n (b) Exceptions.\n\n (1) Criminal Cases. The court may admit\n the following evidence in a criminal case:\n\n (A) evidence of specific instances of\n a victim’s sexual behavior, if offered\n\f6 UNITED STATES V. HAINES\n\n to prove that someone other than the\n defendant was the source of semen,\n injury, or other physical evidence;\n\n (B) evidence of specific instances of\n a victim’s sexual behavior with\n respect to the person accused of the\n sexual misconduct, if offered by the\n defendant to prove consent or if\n offered by the prosecutor; and\n\n (C) evidence whose exclusion would\n violate the defendant’s constitutional\n rights.\n\nFed. R. Evid. 412.\n\n The district court correctly determined that the Rule\napplied. As our sister circuits have noted, sex trafficking\ncases involve “alleged sexual misconduct,” see United States\nv. Wardlow, 830 F.3d 817, 820 (8th Cir. 2016) (applying\nRule 412 in § 2423 prosecution); United States v. Elbert, 561\nF.3d 771, 776 (8th Cir. 2009) (applying Rule 412 in § 1591\nprosecution), and evidence of a trafficking victim’s pre- or\npost-indictment involvement in prostitution implicates her\n“other sexual behavior” or “sexual predisposition,” see\nUnited States v. Lockhart, 844 F.3d 501, 509 (5th Cir. 2016).\nConsistent with this construction of the Rule, courts have\nroutinely barred evidence of a sex trafficking victim’s other\nprostitution activities. See, e.g., United States v. Betts, 911\nF.3d 523, 528–29 (8th Cir. 2018); United States v. Groce,\n891 F.3d 260, 267–68 (7th Cir. 2018); United States v.\nCarson, 870 F.3d 584, 593 (7th Cir. 2017), cert. denied, 138\nS. Ct. 2011 (2018); United States v. Gemma, 818 F.3d 23,\n34–35 (1st Cir. 2016); Lockhart, 844 F.3d at 510; Elbert, 561\nF.3d at 777; cf. United States v. Backman, 817 F.3d 662, 670\n\f UNITED STATES V. HAINES 7\n\n(9th Cir. 2016) (affirming exclusion for failure to comply\nwith the Rule’s procedural requirements but expressing\n“doubt that evidence that the victim engaged in commercial\nsex acts after she had been coerced into prostitution has a\nbearing on whether Defendant earlier took coercive\nactions”).\n\n Haines argues that evidence of J.C.’s prior prostitution\nactivities should have been admitted under the exception to\nRule 412 for “evidence whose exclusion would violate the\ndefendant’s constitutional rights”—here, his due process\nright to present a complete defense and his Sixth\nAmendment right to confront witnesses. But in cases\ninvolving adult victims forced or coerced into prostitution,\ncourts have rejected such arguments, concluding that\nevidence of other prostitution activity has little or no\nrelevance. Courts have reasoned that just because a victim\nagreed to engage in sex for money on other occasions does\nnot mean she consented to, e.g., being beaten or having her\nearnings confiscated by the defendant. See United States v.\nRivera, 799 F.3d 180, 185–86 (2d Cir. 2015); United States\nv. Cephus, 684 F.3d 703, 708 (7th Cir. 2012).\n\n Haines argues that his case is different because his\ndefense was not consent but rather that he had no intent to,\nand did not, pimp out J.C. However, he makes no attempt to\ndistinguish Elbert, a case, like his, involving a minor victim\n(as to which the government need not show force, fraud, or\ncoercion to prove a violation). The Elbert court rejected the\nargument that evidence of a child-victim’s other sexual\nbehavior should be admitted to rebut the allegation that the\ndefendant recruited the victim to engage in commercial sex\nacts. The court explained:\n\f8 UNITED STATES V. HAINES\n\n What Elbert fails to recognize is the evidence\n he wishes to admit does not provide a defense\n for the crime with which he was charged and\n convicted. Elbert repeatedly argues evidence\n of the victims’ prior acts of prostitution\n demonstrates he did not cause them to engage\n in commercial sex acts. This argument relies\n upon an improper construction of the phrase\n “caused to engage.” 18 U.S.C. § 1591(a).\n Because the victims were minors and could\n not legally consent, the government did not\n need to prove the elements of fraud, force, or\n coercion, which are required for adult\n victims. Id. Instead, the government was only\n required to prove Elbert knowingly recruited,\n enticed, harbored, transported, provided, or\n obtained a minor, knowing the minor would\n be caused to engage in commercial sex acts.\n Id. Whether the children engaged in acts of\n prostitution before or after their encounters\n with Elbert is irrelevant, and would only\n prove other people may be guilty of similar\n offenses of recruiting, enticing, or causing\n these victims to engage in a commercial sex\n act.\n\n561 F.3d at 777. The court thus rejected Elbert’s argument\nthat this was “evidence [whose] exclusion . . . would violate\nthe defendant’s constitutional rights” and thus within the\nexception contained in Rule 412. Id. at 776–77 (quoting\nFed. R. Evid. 412(b)(1)(C)).\n\n That J.C. may have prostituted on other occasions on her\nown does not change the result. In United States v.\nShamsud-Din, No. 10 CR 927, 2011 U.S. Dist. LEXIS\n\f UNITED STATES V. HAINES 9\n\n124449, at *3 (N.D. Ill. Oct. 27, 2011), the defendant sought\nto admit evidence that the alleged victims prostituted at other\ntimes without a pimp. In a passage that applies equally to\nthis case, the court, ruling pre-trial, stated:\n\n Defendant contends that his defense is\n broader than that in Elbert because his\n proffered evidence here offers a complete\n defense to the charges. He specifically asserts\n that he is arguing that the proffered evidence\n shows that Victims A and B engaged in the\n prostitution on their own and without his\n involvement or knowledge. His argument is a\n distinction without a difference. Defendant is\n free to explore whether the Victims engaged\n in the prostitution activities at issue in this\n case on their own, rather than with his\n assistance. He can also question the Victims\n about their computer skills and whether or\n not they are familiar with Craigslist, without\n asking about using it for advertising prior or\n subsequent prostitution activities. Defendant\n cannot, however, inquire into other\n prostitution activities. Such evidence of prior\n and post prostitution activities is the\n equivalent of propensity evidence and\n irrelevant to the charges.\n\nId. at *11.\n\n Haines cites no case holding that a defense such as the\none he sought to present here triggers the exception in Rule\n\f10 UNITED STATES V. HAINES\n\n412. And we see no reason to depart from the persuasive\nauthorities set forth above that held to the contrary. 1\n\n What does make this case somewhat different from those\ncited above is J.C.’s posture at trial. As Haines notes, Rule\n412 aims to “safeguard the alleged victim against the\ninvasion of privacy, potential embarrassment and sexual\nstereotyping that is associated with public disclosure of\nintimate sexual details.” See Fed. R. Evid. 412, Advisory\nCommittee’s Notes to 1994 Amendments (“Advisory\nCommittee Notes”). Haines contends that in his case J.C.\nactually wanted to testify about her prior acts of prostitution,\nand that the government used Rule 412 not as a shield to\nprotect a cooperative victim’s privacy, but as a sword to\nobtain a conviction by precluding him from eliciting\nfavorable testimony from a recalcitrant witness. The parties\ndo not cite—and we have not found—a case discussing\napplicability of the Rule to a witness hostile to the\ngovernment.\n\n For several reasons, however, we conclude that the\napplicability of the Rule should not depend on the alleged\nvictim’s desire to testify. First, Rule 412 is a rule of\nexclusion containing three specific exceptions in criminal\ncases; the victim’s desire to testify or waive the protections\nof the Rule is not one of them. Second, to the extent that\nRule 412 also serves the purpose of keeping irrelevant,\n\n\n 1\n For the same reasons, we reject Haines’s argument that his right to\npresent a defense was violated when, after the government challenged\nthe credibility of J.C.’s trial testimony, he was not permitted to\n“rehabilitate” her with irrelevant and otherwise inadmissible evidence\nabout her prior sexual behavior. See United States v. Scheffer, 523 U.S.\n303, 308 (1998) (explaining that the right to present a defense is not\nunfettered, and law-makers retain broad latitude to establish rules\nexcluding evidence from criminal trials).\n\f UNITED STATES V. HAINES 11\n\nprejudicial, and/or inflammatory evidence from the jury, it\nshould not matter whether the witness wants its protection;\nthe district court enforces the Rule to ensure that the jury\ndecides the case based on proper considerations. See\nAdvisory Committee Notes (noting that the Rule prevents\n“the infusion of sexual innuendo into the factfinding\nprocess”); Privacy of Rape Victims: Hearing Before the\nSubcomm. on Crim. J. of the H. Comm. on the Judiciary on\nH.R. 14666 and Other Bills, 94th Cong. 41 (1976) (statement\nof Mary Ann Largen, National Organization of Women)\n(“[Proposed Rule 412] assures that highly inflammatory and\narguably irrelevant matters will not be injected.”). Third,\nallowing the victim’s wishes to control opens the door to\nmischief. Indeed, in this case the government suspected that\nJ.C. changed her testimony based on pressure from Haines.\n\n Finally, even if the district court misapplied Rule 412\nhere, and we do not believe it did, any error would be\nharmless. See United States v. Preston, 873 F.3d 829, 835\n(9th Cir. 2017) (noting that evidentiary rulings are subject to\nharmless error review); United States v. Nielson, 371 F.3d\n574, 581 (9th Cir. 2004) (“Confrontation Clause violations\nare subject to harmless error analysis[.]”); see also United\nStates v. Torres, 794 F.3d 1053, 1062–63 (9th Cir. 2015)\n(holding that an evidentiary ruling excluding evidence\ndesired by the defense did not violate the right to present a\ncomplete defense because the evidence would not have\nsubstantially furthered the defense’s trial theories). Haines\nwas able to present his theory of the case through J.C.’s\ntestimony that he was not her pimp during the trip to\nCalifornia, that she engaged in the charged acts of\nprostitution on her own and without his encouragement or\ninvolvement, and that she kept the money she earned.\nHaines was also able to argue to the jury in closing that this\ntrip to California was not J.C.’s “first rodeo,” that she knew\n\f12 UNITED STATES V. HAINES\n\nwhere the track was, and that she knew how to place ads on\nBackpage. The district court also gave the jury a “mere\npresence” instruction in support of the defense, explaining\nthat to convict in this case “[t]he defendant must be a\nparticipant and not merely a knowing spectator,” which\nHaines incorporated into his argument. It is hard to see how\nadditional testimony about J.C.’s other “solo” prostitution-\nrelated activities would have materially aided the defense.\nSee United States v. Hofus, 598 F.3d 1171, 1180 (9th Cir.\n2010) (holding that limitation on testimony from key witness\ndid not violate the defendant’s ability to present a defense,\nwhere counsel was still able to argue his theory of the case).\n\n We have considered Haines’s additional arguments that\nthe government opened the door to testimony about J.C.’s\nprior activities, but none has merit. The government did not\nelicit testimony suggesting that Haines introduced J.C. to\nprostitution or that she was, in the district court’s words, “an\ninnocent lamb led to the slaughter.” The district court took\nthe issue seriously, warning the government that it could\nopen the door depending on how it presented its case. We\nhold there was no abuse of discretion in its rulings.\n\n Haines argues that his defense was prejudiced by the\ndistrict court’s refusal to allow him to recall J.C. to ask her\nabout her prior prostitution activities after King testified that\nHaines, A.S., and J.C. came up with the idea to prostitute\nwhile they were at a pool party together. As the district court\nnoted in denying Haines’s request, J.C. had already testified\nthat she and A.S. came up with the idea, without Haines’s\ninvolvement, and the jury would have to decide who was\ntelling the truth, J.C. or King. The government did not open\nthe door simply by presenting evidence contrary to J.C.’s\ntrial testimony.\n\f UNITED STATES V. HAINES 13\n\n III.\n\n For the foregoing reasons, and those stated in our\nseparate memorandum disposition addressing Haines’s other\narguments, we affirm the district court’s judgment.\n\n AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. Sha-Ron Haines\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-14\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-10059\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 2:14-cr-00264-APG-\\n VCF-2\\nSHA-RON HAINES,\\n Defendant-Appellant. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the District of Nevada\\n Andrew P. Gordon, District Judge, Presiding\\n\\n Argued and Submitted January 17, 2019\\n San Francisco, California\\n\\n Filed March 14, 2019\\n\\n Before: J. Clifford Wallace and Michelle T. Friedland,\\n Circuit Judges, and Lynn S. Adelman, * District Judge.\\n\\n Opinion by Judge Adelman\\n\\n\\n\\n\\n *\\n The Honorable Lynn S. Adelman, United States District Judge for\\nthe Eastern District of Wisconsin, sitting by designation.\\n\\f2 UNITED STATES V. HAINES\\n\\n SUMMARY **\\n\\n\\n Criminal Law\\n\\n The panel affirmed the district court’s judgment in a case\\nin which the defendant was convicted of conspiracy to\\ncommit sex trafficking of a minor, sex trafficking of a minor,\\nconspiracy to transport a minor to engage in prostitution, and\\ntransporting a minor to engage in prostitution.\\n\\n At trial, the defendant sought to question minor J.C.\\nabout her prior prostitution activities (which apparently did\\nnot involve a pimp), arguing that this evidence was relevant\\nto, among other things, whether he recruited or encouraged\\nher to engage in prostitution on this occasion. The panel held\\nthat the district court did not err by excluding the testimony\\nunder Fed. R. Evid. 412, the “rape shield” rule. The panel\\nrejected the defendant’s contention that evidence of J.C.’s\\nprior prostitution activities should have been admitted under\\nthe exception in Rule 412(b)(1)(C) for “evidence whose\\nexclusion would violate [his] constitutional rights”—here,\\nhis due process right to present a complete defense and his\\nSixth Amendment right to confront witnesses. The panel\\nsaw no reason to depart from persuasive authorities holding\\nthat a defense such as the one the defendant sought to\\npresent—that he had no intent to, and did not, pimp out\\nJ.C.—triggers the exception. The panel also held that the\\napplicability of Rule 412 should not depend on the alleged\\nvictim’s desire to testify. The panel concluded that even if\\nthe district court misapplied Rule 412, any error would be\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f UNITED STATES V. HAINES 3\\n\\nharmless. The panel held that the defendant’s arguments that\\nthe government opened the door to testimony about J.C.’s\\nprior activities lacked merit.\\n\\n The panel addressed other arguments in a separate\\nmemorandum disposition.\\n\\n\\n COUNSEL\\n\\nKaren A. Connolly (argued), Karen A. Connolly, Ltd., Las\\nVegas, Nevada, for Defendant-Appellant.\\n\\nVijay Shanker (argued), United States Department of\\nJustice, Criminal Division, Appellate Section, Washington\\nD.C., for Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nADELMAN, District Judge:\\n\\n Sha-Ron Haines appeals his convictions for conspiracy\\nto commit sex trafficking of a minor, sex trafficking of a\\nminor, conspiracy to transport a minor to engage in\\nprostitution, and transporting a minor to engage in\\nprostitution. We have jurisdiction under 28 U.S.C. § 1291.\\nWe affirm.\\n\\n I.\\n\\n The government alleged that Haines and his friend Tyral\\nKing transported two minor females, J.C. (age 15) and A.S.\\n(age 17), from Nevada to California to prostitute them, with\\nJ.C. working for Haines and A.S. working for King. The\\n\\f4 UNITED STATES V. HAINES\\n\\ngirls found “dates” by walking a “track” where men picked\\nup prostitutes and through ads posted on a website called\\n“Backpage.com.” King and Haines would drop the girls off\\nat their dates and return to pick them up afterwards.\\n\\n J.C. initially cooperated with the government’s\\ninvestigation, albeit reluctantly. She testified before the\\ngrand jury that ultimately indicted Haines and King that she\\nworked for Haines and gave him the money she earned from\\nprostitution. Prior to trial, however, her account changed.\\nJ.C. then claimed that she initially implicated Haines due to\\npressure from the investigating detective to testify in\\nexchange for release from juvenile detention. This change\\nmay have been prompted by a jailhouse phone call in which\\nHaines advised J.C. to make herself unavailable to testify at\\ntrial, of which the government later found a recording.\\n\\n Whatever the reason, at trial J.C. testified that she\\nworked independently, that she kept her earnings, and that\\nher prior grand jury testimony to the contrary was false. The\\ngovernment impeached J.C. with her previous testimony.\\nSee Fed. R. Evid. 801(d)(1)(A) (authorizing the admission\\nof prior inconsistent statements by testifying witnesses as\\nsubstantive evidence if the prior statements were given under\\noath). The government also presented testimony from King,\\nwho pleaded guilty and agreed to cooperate with the\\ngovernment, that J.C. worked for Haines and gave Haines\\nher prostitution earnings.\\n\\n The jury convicted Haines on all counts. The district\\ncourt sentenced him to 156 months in prison.\\n\\n II.\\n\\n Haines’s defense at trial was that he was merely along\\nfor the ride and did not act as J.C.’s pimp. In support of that\\n\\f UNITED STATES V. HAINES 5\\n\\ndefense, he sought to question J.C. about her prior\\nprostitution activities (which apparently did not involve a\\npimp), arguing that this evidence was relevant to, among\\nother things, whether he recruited or encouraged her to\\nengage in prostitution on this occasion. The district court\\nexcluded the testimony under Federal Rule of Evidence 412,\\nthe “rape shield” rule.\\n\\n We review a district court’s evidentiary rulings for abuse\\nof discretion, though we review de novo the district court’s\\ninterpretation of the Federal Rules of Evidence. United\\nStates v. Kahre, 737 F.3d 554, 565 (9th Cir. 2013). We also\\nreview de novo whether a district court’s evidentiary rulings\\nviolated a defendant’s constitutional rights. United States v.\\nLaursen, 847 F.3d 1026, 1031 (9th Cir. 2017).\\n\\n Rule 412 provides, in pertinent part:\\n\\n (a) Prohibited Uses. The following evidence\\n is not admissible in a civil or criminal\\n proceeding involving alleged sexual\\n misconduct:\\n\\n (1) evidence offered to prove that a victim\\n engaged in other sexual behavior; or\\n\\n (2) evidence offered to prove a victim’s\\n sexual predisposition.\\n\\n (b) Exceptions.\\n\\n (1) Criminal Cases. The court may admit\\n the following evidence in a criminal case:\\n\\n (A) evidence of specific instances of\\n a victim’s sexual behavior, if offered\\n\\f6 UNITED STATES V. HAINES\\n\\n to prove that someone other than the\\n defendant was the source of semen,\\n injury, or other physical evidence;\\n\\n (B) evidence of specific instances of\\n a victim’s sexual behavior with\\n respect to the person accused of the\\n sexual misconduct, if offered by the\\n defendant to prove consent or if\\n offered by the prosecutor; and\\n\\n (C) evidence whose exclusion would\\n violate the defendant’s constitutional\\n rights.\\n\\nFed. R. Evid. 412.\\n\\n The district court correctly determined that the Rule\\napplied. As our sister circuits have noted, sex trafficking\\ncases involve “alleged sexual misconduct,” see United States\\nv. Wardlow, 830 F.3d 817, 820 (8th Cir. 2016) (applying\\nRule 412 in § 2423 prosecution); United States v. Elbert, 561\\nF.3d 771, 776 (8th Cir. 2009) (applying Rule 412 in § 1591\\nprosecution), and evidence of a trafficking victim’s pre- or\\npost-indictment involvement in prostitution implicates her\\n“other sexual behavior” or “sexual predisposition,” see\\nUnited States v. Lockhart, 844 F.3d 501, 509 (5th Cir. 2016).\\nConsistent with this construction of the Rule, courts have\\nroutinely barred evidence of a sex trafficking victim’s other\\nprostitution activities. See, e.g., United States v. Betts, 911\\nF.3d 523, 528–29 (8th Cir. 2018); United States v. Groce,\\n891 F.3d 260, 267–68 (7th Cir. 2018); United States v.\\nCarson, 870 F.3d 584, 593 (7th Cir. 2017), cert. denied, 138\\nS. Ct. 2011 (2018); United States v. Gemma, 818 F.3d 23,\\n34–35 (1st Cir. 2016); Lockhart, 844 F.3d at 510; Elbert, 561\\nF.3d at 777; cf. United States v. Backman, 817 F.3d 662, 670\\n\\f UNITED STATES V. HAINES 7\\n\\n(9th Cir. 2016) (affirming exclusion for failure to comply\\nwith the Rule’s procedural requirements but expressing\\n“doubt that evidence that the victim engaged in commercial\\nsex acts after she had been coerced into prostitution has a\\nbearing on whether Defendant earlier took coercive\\nactions”).\\n\\n Haines argues that evidence of J.C.’s prior prostitution\\nactivities should have been admitted under the exception to\\nRule 412 for “evidence whose exclusion would violate the\\ndefendant’s constitutional rights”—here, his due process\\nright to present a complete defense and his Sixth\\nAmendment right to confront witnesses. But in cases\\ninvolving adult victims forced or coerced into prostitution,\\ncourts have rejected such arguments, concluding that\\nevidence of other prostitution activity has little or no\\nrelevance. Courts have reasoned that just because a victim\\nagreed to engage in sex for money on other occasions does\\nnot mean she consented to, e.g., being beaten or having her\\nearnings confiscated by the defendant. See United States v.\\nRivera, 799 F.3d 180, 185–86 (2d Cir. 2015); United States\\nv. Cephus, 684 F.3d 703, 708 (7th Cir. 2012).\\n\\n Haines argues that his case is different because his\\ndefense was not consent but rather that he had no intent to,\\nand did not, pimp out J.C. However, he makes no attempt to\\ndistinguish Elbert, a case, like his, involving a minor victim\\n(as to which the government need not show force, fraud, or\\ncoercion to prove a violation). The Elbert court rejected the\\nargument that evidence of a child-victim’s other sexual\\nbehavior should be admitted to rebut the allegation that the\\ndefendant recruited the victim to engage in commercial sex\\nacts. The court explained:\\n\\f8 UNITED STATES V. HAINES\\n\\n What Elbert fails to recognize is the evidence\\n he wishes to admit does not provide a defense\\n for the crime with which he was charged and\\n convicted. Elbert repeatedly argues evidence\\n of the victims’ prior acts of prostitution\\n demonstrates he did not cause them to engage\\n in commercial sex acts. This argument relies\\n upon an improper construction of the phrase\\n “caused to engage.” 18 U.S.C. § 1591(a).\\n Because the victims were minors and could\\n not legally consent, the government did not\\n need to prove the elements of fraud, force, or\\n coercion, which are required for adult\\n victims. Id. Instead, the government was only\\n required to prove Elbert knowingly recruited,\\n enticed, harbored, transported, provided, or\\n obtained a minor, knowing the minor would\\n be caused to engage in commercial sex acts.\\n Id. Whether the children engaged in acts of\\n prostitution before or after their encounters\\n with Elbert is irrelevant, and would only\\n prove other people may be guilty of similar\\n offenses of recruiting, enticing, or causing\\n these victims to engage in a commercial sex\\n act.\\n\\n561 F.3d at 777. The court thus rejected Elbert’s argument\\nthat this was “evidence [whose] exclusion . . . would violate\\nthe defendant’s constitutional rights” and thus within the\\nexception contained in Rule 412. Id. at 776–77 (quoting\\nFed. R. Evid. 412(b)(1)(C)).\\n\\n That J.C. may have prostituted on other occasions on her\\nown does not change the result. In United States v.\\nShamsud-Din, No. 10 CR 927, 2011 U.S. Dist. LEXIS\\n\\f UNITED STATES V. HAINES 9\\n\\n124449, at *3 (N.D. Ill. Oct. 27, 2011), the defendant sought\\nto admit evidence that the alleged victims prostituted at other\\ntimes without a pimp. In a passage that applies equally to\\nthis case, the court, ruling pre-trial, stated:\\n\\n Defendant contends that his defense is\\n broader than that in Elbert because his\\n proffered evidence here offers a complete\\n defense to the charges. He specifically asserts\\n that he is arguing that the proffered evidence\\n shows that Victims A and B engaged in the\\n prostitution on their own and without his\\n involvement or knowledge. His argument is a\\n distinction without a difference. Defendant is\\n free to explore whether the Victims engaged\\n in the prostitution activities at issue in this\\n case on their own, rather than with his\\n assistance. He can also question the Victims\\n about their computer skills and whether or\\n not they are familiar with Craigslist, without\\n asking about using it for advertising prior or\\n subsequent prostitution activities. Defendant\\n cannot, however, inquire into other\\n prostitution activities. Such evidence of prior\\n and post prostitution activities is the\\n equivalent of propensity evidence and\\n irrelevant to the charges.\\n\\nId. at *11.\\n\\n Haines cites no case holding that a defense such as the\\none he sought to present here triggers the exception in Rule\\n\\f10 UNITED STATES V. HAINES\\n\\n412. And we see no reason to depart from the persuasive\\nauthorities set forth above that held to the contrary. 1\\n\\n What does make this case somewhat different from those\\ncited above is J.C.’s posture at trial. As Haines notes, Rule\\n412 aims to “safeguard the alleged victim against the\\ninvasion of privacy, potential embarrassment and sexual\\nstereotyping that is associated with public disclosure of\\nintimate sexual details.” See Fed. R. Evid. 412, Advisory\\nCommittee’s Notes to 1994 Amendments (“Advisory\\nCommittee Notes”). Haines contends that in his case J.C.\\nactually wanted to testify about her prior acts of prostitution,\\nand that the government used Rule 412 not as a shield to\\nprotect a cooperative victim’s privacy, but as a sword to\\nobtain a conviction by precluding him from eliciting\\nfavorable testimony from a recalcitrant witness. The parties\\ndo not cite—and we have not found—a case discussing\\napplicability of the Rule to a witness hostile to the\\ngovernment.\\n\\n For several reasons, however, we conclude that the\\napplicability of the Rule should not depend on the alleged\\nvictim’s desire to testify. First, Rule 412 is a rule of\\nexclusion containing three specific exceptions in criminal\\ncases; the victim’s desire to testify or waive the protections\\nof the Rule is not one of them. Second, to the extent that\\nRule 412 also serves the purpose of keeping irrelevant,\\n\\n\\n 1\\n For the same reasons, we reject Haines’s argument that his right to\\npresent a defense was violated when, after the government challenged\\nthe credibility of J.C.’s trial testimony, he was not permitted to\\n“rehabilitate” her with irrelevant and otherwise inadmissible evidence\\nabout her prior sexual behavior. See United States v. Scheffer, 523 U.S.\\n303, 308 (1998) (explaining that the right to present a defense is not\\nunfettered, and law-makers retain broad latitude to establish rules\\nexcluding evidence from criminal trials).\\n\\f UNITED STATES V. HAINES 11\\n\\nprejudicial, and/or inflammatory evidence from the jury, it\\nshould not matter whether the witness wants its protection;\\nthe district court enforces the Rule to ensure that the jury\\ndecides the case based on proper considerations. See\\nAdvisory Committee Notes (noting that the Rule prevents\\n“the infusion of sexual innuendo into the factfinding\\nprocess”); Privacy of Rape Victims: Hearing Before the\\nSubcomm. on Crim. J. of the H. Comm. on the Judiciary on\\nH.R. 14666 and Other Bills, 94th Cong. 41 (1976) (statement\\nof Mary Ann Largen, National Organization of Women)\\n(“[Proposed Rule 412] assures that highly inflammatory and\\narguably irrelevant matters will not be injected.”). Third,\\nallowing the victim’s wishes to control opens the door to\\nmischief. Indeed, in this case the government suspected that\\nJ.C. changed her testimony based on pressure from Haines.\\n\\n Finally, even if the district court misapplied Rule 412\\nhere, and we do not believe it did, any error would be\\nharmless. See United States v. Preston, 873 F.3d 829, 835\\n(9th Cir. 2017) (noting that evidentiary rulings are subject to\\nharmless error review); United States v. Nielson, 371 F.3d\\n574, 581 (9th Cir. 2004) (“Confrontation Clause violations\\nare subject to harmless error analysis[.]”); see also United\\nStates v. Torres, 794 F.3d 1053, 1062–63 (9th Cir. 2015)\\n(holding that an evidentiary ruling excluding evidence\\ndesired by the defense did not violate the right to present a\\ncomplete defense because the evidence would not have\\nsubstantially furthered the defense’s trial theories). Haines\\nwas able to present his theory of the case through J.C.’s\\ntestimony that he was not her pimp during the trip to\\nCalifornia, that she engaged in the charged acts of\\nprostitution on her own and without his encouragement or\\ninvolvement, and that she kept the money she earned.\\nHaines was also able to argue to the jury in closing that this\\ntrip to California was not J.C.’s “first rodeo,” that she knew\\n\\f12 UNITED STATES V. HAINES\\n\\nwhere the track was, and that she knew how to place ads on\\nBackpage. The district court also gave the jury a “mere\\npresence” instruction in support of the defense, explaining\\nthat to convict in this case “[t]he defendant must be a\\nparticipant and not merely a knowing spectator,” which\\nHaines incorporated into his argument. It is hard to see how\\nadditional testimony about J.C.’s other “solo” prostitution-\\nrelated activities would have materially aided the defense.\\nSee United States v. Hofus, 598 F.3d 1171, 1180 (9th Cir.\\n2010) (holding that limitation on testimony from key witness\\ndid not violate the defendant’s ability to present a defense,\\nwhere counsel was still able to argue his theory of the case).\\n\\n We have considered Haines’s additional arguments that\\nthe government opened the door to testimony about J.C.’s\\nprior activities, but none has merit. The government did not\\nelicit testimony suggesting that Haines introduced J.C. to\\nprostitution or that she was, in the district court’s words, “an\\ninnocent lamb led to the slaughter.” The district court took\\nthe issue seriously, warning the government that it could\\nopen the door depending on how it presented its case. We\\nhold there was no abuse of discretion in its rulings.\\n\\n Haines argues that his defense was prejudiced by the\\ndistrict court’s refusal to allow him to recall J.C. to ask her\\nabout her prior prostitution activities after King testified that\\nHaines, A.S., and J.C. came up with the idea to prostitute\\nwhile they were at a pool party together. As the district court\\nnoted in denying Haines’s request, J.C. had already testified\\nthat she and A.S. came up with the idea, without Haines’s\\ninvolvement, and the jury would have to decide who was\\ntelling the truth, J.C. or King. The government did not open\\nthe door simply by presenting evidence contrary to J.C.’s\\ntrial testimony.\\n\\f UNITED STATES V. HAINES 13\\n\\n III.\\n\\n For the foregoing reasons, and those stated in our\\nseparate memorandum disposition addressing Haines’s other\\narguments, we affirm the district court’s judgment.\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1290,"cells":{"id":{"kind":"number","value":4603417,"string":"4,603,417"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. Jonathan Read"},"decision_date":{"kind":"string","value":"2019-03-14"},"docket_number":{"kind":"string","value":"17-10439"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Michael Daly Hawkins, Milan D. Smith, Jr., and Andrew D. Hurwitz, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"Opinion by Judge Hawkins\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-10439\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 2:15-cr-00933-\\n DLR-1\\nJONATHAN LEE READ,\\n Defendant-Appellant. OPINION\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Douglas L. Rayes, District Judge, Presiding\\n\\n Argued and Submitted February 8, 2019\\n Arizona State University, Phoenix\\n\\n Filed March 14, 2019\\n\\nBefore: Michael Daly Hawkins, Milan D. Smith, Jr., and\\n Andrew D. Hurwitz, Circuit Judges.\\n\\n Opinion by Judge Hawkins\\n\\f2 UNITED STATES V. READ\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n The panel reversed a criminal judgment and remanded in\\na case in which the defendant—who stabbed his cellmate\\nwhile serving a sentence in the Federal Correctional Institute\\nin Phoenix, Arizona—was convicted of assault with a deadly\\nweapon with intent to bodily harm, and assault with a deadly\\nweapon resulting in serious bodily injury, in violation of 18\\nU.S.C. § 113(a).\\n\\n The panel rejected the defendant’s arguments that the\\ngovernment did not sufficiently allege and prove the\\njurisdiction element of § 113(a), which prohibits assaults\\n“within the special maritime and territorial jurisdiction of the\\nUnited States.”\\n\\n The panel held that in light of McCoy v. Louisiana, 138\\nS. Ct. 1500 (2018), a district court commits reversible error\\nby permitting defense counsel to present a defense of\\ninsanity over a competent defendant’s clear rejection of that\\ndefense.\\n\\n The panel held that the district court did not err by\\nrevoking his pro se status under Indiana v. Edwards, 554\\nU.S. 164 (2008), on the basis of his “decidedly bizarre” and\\n“nonsensical” arguments.\\n\\n\\n\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f UNITED STATES V. READ 3\\n\\n The panel held that the defendant waived his claimed\\nthat the indictment must be dismissed because his trial did\\nnot comply with the Speedy Trial Act.\\n\\n\\n COUNSEL\\n\\nDavina T. Chen (argued), Glendale, California, for\\nDefendant-Appellant.\\n\\nPeter S. Kozinets (argued), Assistant United States Attorney;\\nKrissa M. Lanham, Deputy Appellate Chief; Elizabeth A.\\nStrange, First Assistant United States Attorney, District of\\nArizona; United States Attorney’s Office Phoenix, Arizona;\\nfor Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nHAWKINS, Senior Circuit Judge:\\n\\n We address a question of first impression: whether a\\ncriminal defendant has the Sixth Amendment right to\\ndemand that counsel not present an insanity defense. We\\nhold that McCoy v. Louisiana, 138 S. Ct. 1500 (2018),\\nrequires under the facts of this case that the demand be\\nhonored. We therefore reverse and remand for a new trial.\\n\\n FACTS AND PROCEEDINGS BELOW\\n\\n a. Read was charged with assaulting his cellmate.\\n\\n While serving a sentence for attempted robbery in the\\nFederal Correctional Institute in Phoenix, Arizona (“FCI-\\nPhoenix”), Jonathan Lee Read stabbed his cellmate thirteen\\ntimes with a homemade knife. Read’s relationship with his\\n\\f4 UNITED STATES V. READ\\n\\ncellmates had previously been without tension. When\\ndetained, Read claimed he had no memory of the attack.\\n\\n Read was charged with one count each of assault with a\\ndeadly weapon with intent to do bodily harm, and assault\\nwith a deadly weapon resulting in serious bodily injury, both\\nin violation of 18 U.S.C. § 113(a), which prohibits assaults\\n“within the special maritime and territorial jurisdiction of the\\nUnited States.” Id.\\n\\n b. Read was ultimately deemed competent to stand\\n trial.\\n\\n Upon the motion of appointed counsel, Read was\\ninitially admitted to the Metropolitan Detention Center in\\nLos Angeles for evaluation of his competency. Two months\\nlater, Dr. Lesli Johnson, Ph.D., a forensic psychologist,\\nissued a report diagnosing Read with schizophrenia and\\nsevere cannabis use disorder. Dr. Johnson’s diagnosis of\\nschizophrenia was based on Read’s delusional thoughts\\nregarding Christianity, Satan, and demonization. She noted\\nthat Read appeared to respond to internal stimuli and that his\\nthought and speech patterns were disorganized when he\\ndiscussed his delusionally-focused beliefs. The district court\\naccepted the report, found Read incompetent to stand trial,\\nand ordered him committed for hospitalization, treatment,\\nand restoration, pursuant to 18 U.S.C. §§ 4241 and 4247.\\nRead was admitted to the Federal Medical Center in\\nSpringfield, Missouri.\\n\\n Four months later, Dr. Allison Schenk, Ph.D., a clinical\\npsychologist, determined that Read was competent to stand\\ntrial. Dr. Schenk also took note of Read’s unusual beliefs,\\nreporting that Read claimed he was experiencing the\\nsensation of having his stomach cut with a knife, caused by\\na person’s use of a “voodoo doll” against him. Dr. Schenk\\n\\f UNITED STATES V. READ 5\\n\\ndetermined that Read suffered from schizotypal personality\\ndisorder and cannabis use disorder, but nonetheless\\ndetermined that Read was competent to stand trial.\\n\\n Based on Dr. Schenk’s evaluation, and without objection\\nfrom defense counsel, the court found Read competent to\\nstand trial.\\n\\n c. Read was evaluated to determine his sanity during\\n the assault.\\n\\n Read’s counsel arranged for an examination by Dr. John\\nR. Walker III, Psy.D., a neuropsychologist, to assess his state\\nof mind at the time of the alleged assault. Dr. Walker\\nreported many of the same behaviors as had Drs. Johnson\\nand Schenk. He concluded that Read’s psychosis rendered\\nhim unable to form criminal intent, and that Read was likely\\nstill psychotic at the time of the examination.\\n\\n Read’s appointed counsel then filed a Notice of Insanity\\nDefense. In response, the government requested an\\nexamination of Read pursuant to 18 U.S.C. § 4242(a). The\\ncourt granted the government’s motion.\\n\\n Read was admitted to the Federal Medical Center in\\nButner, North Carolina (“FMC-Butner”) for the\\nexamination. FMC-Butner eventually issued a report\\nconcluding that Read was not insane at the time of the\\nalleged offense. The report was prepared by Sumandeep\\nKaur, a doctoral psychology intern, under the supervision of\\nforensic psychologist Dr. Angela Walden Weaver, Ph.D.\\nMs. Kaur reported that, during the evaluation, Read claimed\\nhe was suffering from “demonization” rather than mental\\nillness. She diagnosed Read with schizotypal personality\\ndisorder and cannabis use disorder, and opined that Read\\nwas able to appreciate the nature, quality, and wrongfulness\\n\\f6 UNITED STATES V. READ\\n\\nof his alleged criminal acts. The district court ordered\\nRead’s return to the District of Arizona.\\n\\n d. Read successfully moved to proceed without counsel.\\n\\n Read then asked to proceed without counsel. Following\\na Faretta hearing, 1 the court held that Read had a right to do\\nso, finding that he “knowingly and voluntarily waived the\\nright to counsel.” Read’s appointed counsel was named as\\nstandby counsel.\\n\\n e. The court vacated its order and reappointed counsel\\n for Read.\\n\\n Prior to the final trial management conference, advisory\\ncounsel told the government that Read might abandon an\\ninsanity defense in favor of a defense based on demonic\\npossession. At the final pretrial conference, Read did just\\nthat. The judge subsequently asked Read if he intended to\\ncall any witnesses to discuss his mental condition. Read said\\nthat he would call Dr. Walker, the neuropathologist who had\\nopined that Read was insane at the time of the alleged\\nassault. However, shortly thereafter, Read told the court, “I\\ncompletely withdraw the insanity. That’s not an option for\\nme.” The judge asked Read to clarify, and Read confirmed\\nthat he did not wish to present an insanity defense.\\n\\n The government responded that, if Read was\\nwithdrawing his insanity defense, then Dr. Walker could\\noffer no relevant testimony. When the court asked him what\\ndefense he intended to pursue, Read responded:\\n\\n\\n\\n 1\\n See Faretta v. California, 422 U.S. 806 (1975).\\n\\f UNITED STATES V. READ 7\\n\\n My civil rights have been violated. I’ve been\\n tied in with so many other people that I have\\n nothing to do with. Incarcerated. My faith,\\n my belief system is not—what is that—\\n extreme Islamic. I believe in—I’m a believer\\n in Christianity. And there’s been a lot of\\n things going on in prison that the prison\\n system itself has brought upon itself.\\n\\nThe court responded that Dr. Walker’s testimony about his\\nmental condition would not be relevant to such a defense.\\nRead then said he wanted Dr. Walker to testify about cases\\nof demonic possession that he had observed in inmates\\nbesides Read. The court affirmed that such testimony would\\nnot be relevant. Read responded, “All right. I guess we can\\nscratch—scratch that.”\\n\\n The court then asked Read if he wished to continue\\nrepresenting himself. Read said that he did. The court asked\\nRead’s standby counsel, if he had any concerns about Read’s\\ncompetence to proceed without counsel. Mr. Williams said\\nthat he was concerned because Read did not seem to\\nunderstand the legal distinction between a defense of\\ninsanity and his proposed defense. The court again asked\\nstandby counsel if he thought Read was competent to\\nrepresent himself. Counsel said that he did not know.\\n\\n The court responded that it would consider whether\\nRead’s standby counsel should be reappointed. Standby\\ncounsel explained that he would present an insanity defense\\nif reappointed; he noted that the very reason that Read had\\nwanted to proceed pro se in the first place was because he\\ndid not want an insanity defense.\\n\\n Over Read’s objection, the district court reappointed\\nstandby counsel to act as Read’s counsel. The court noted,\\n\\f8 UNITED STATES V. READ\\n\\n“[T]he Constitution permits [judges] to insist upon\\nrepresentation by counsel for those competent enough to\\nstand trial[,] . . . but who still suffer from severe mental\\nillness to the point where they are not competent to conduct\\ntrial proceedings by themselves.” The court reasoned that\\n“[t]his standard is met where the defendant’s behavior is\\n‘decidedly bizarre’ and his arguments in defense to the\\ncharges against him are nonsensical.” The court found that\\nRead’s “beliefs are bizarre and his representation will be\\nwholly ineffective,” because “[h]is anticipated defense, that\\nhe is possessed by demons and that other inmates are also\\npossessed, is not a legal defense and is based on his bizarre\\nbeliefs.” Despite these comments, at no point did the district\\ncourt revisit Read’s competency.\\n\\n Counsel ably, but unsuccessfully, proceeded to present\\nan insanity defense at trial. Read was convicted and\\nsentenced to concurrent 82-month terms. This timely appeal\\nfollowed.\\n\\n DISCUSSION\\n\\nI. Whether the Jurisdictional Element Was Alleged and\\n Proved\\n\\n Read was charged under 18 U.S.C. § 113(a), which\\nprohibits an assault “within the special maritime and\\nterritorial jurisdiction of the United States.” Id. This\\njurisdiction includes “[a]ny lands reserved or acquired for\\nthe use of the United States, and under the exclusive or\\nconcurrent jurisdiction thereof, or any place purchased or\\notherwise acquired by the United States by consent of the\\nlegislature of the State in which the same shall be, for the\\nerection of a fort, magazine, arsenal, dockyard, or other\\nneedful building.” 18 U.S.C. § 7(3).\\n\\f UNITED STATES V. READ 9\\n\\n Read argues that the government presented insufficient\\nevidence that the assault took place “within the special\\nmaritime and territorial jurisdiction of the United States.”\\nRead also claims the indictment must be dismissed for\\nfailing to state the jurisdictional element of the offenses.\\nNeither argument is persuasive.\\n\\n a. Sufficiency of the Evidence\\n\\n The existence of federal jurisdiction over the place in\\nwhich the offense occurred is an element of the offenses\\ndefined at 18 U.S.C. § 113(a), which must be proved to the\\njury beyond a reasonable doubt. See United States v.\\nGaudin, 515 U.S. 506, 510 (1995). Read claims that the\\ngovernment’s failure to submit historical documents\\nestablishing FCI-Phoenix’s jurisdictional status means that\\nit failed to prove the jurisdictional component of the charged\\noffenses. But, while historical documents can be sufficient\\nto prove that land is subject to the “special maritime and\\nterritorial jurisdiction of the United States,” they are not\\nnecessary. Our cases addressing convictions under statutes\\nwhose jurisdictional component demands proof that a bank\\nis federally insured are instructive; we have stressed in that\\ncontext that while documentary evidence can be used to\\nestablish jurisdiction, “bank employees’ uncontradicted\\ntestimony of a bank’s insured status can sufficiently support\\nthe jury’s conclusion that this element was proven beyond a\\nreasonable doubt.” United States v. Corbin, 972 F.2d 271,\\n272 (9th Cir. 1992) (per curiam) (citing United States v.\\nCampbell, 616 F.2d 1151, 1153 (9th Cir. 1980). Likewise,\\nuncontradicted testimony from inmates or employees at a\\nfederal prison can establish the jurisdictional element of 18\\nU.S.C. § 113.\\n\\n The government elicited precisely this type of testimony\\nat trial. The victim testified he was an inmate “of the\\n\\f10 UNITED STATES V. READ\\n\\nPhoenix federal prison” at the time of the assault. Phillip\\nLopez, the officer who responded to the assault, testified he\\nworked at the Federal Bureau of Prisons’ male facility in\\nPhoenix, and was employed by the United States\\nDepartment of Justice, Federal Bureau of Prisons. None of\\nthe relevant testimony was objected to or contradicted at\\ntrial; nor, indeed, has Read ever alleged that FCI-Phoenix is\\nnot, in fact, subject to federal jurisdiction. Nor does Read\\ndeny that the assault took place at the prison. A reasonable\\njuror could conclude from these statements that FCI-Phoenix\\nwas under federal jurisdiction at the time Read allegedly\\ncommitted assault. See Jackson v. Virginia, 443 U.S. 307,\\n319 (1979) (noting that we must affirm a conviction if, “after\\nviewing the evidence in the light most favorable to the\\nprosecution, any rational trier of fact could have found the\\nessential elements of the crime beyond a reasonable\\ndoubt.”).\\n\\n b. Sufficiency of the Indictment\\n\\n Read also argues that the indictment should be dismissed\\nbecause it omitted the jurisdictional element of the two\\ncrimes charged. “When the sufficiency of the indictment is\\nchallenged after trial, it is only required that the necessary\\nfacts appear in any form or by fair construction can be found\\nwithin the terms of the indictment.” United States v.\\nVelasco-Medina, 305 F.3d 839, 846 (9th Cir. 2002)\\n(citations, brackets, and quotation marks omitted). “A\\ndefendant is not prejudiced where her counsel has notice of\\nthe omitted element and the jury is properly instructed\\nregarding the missing element.” United States v. Arnt, 474\\nF.3d 1159, 1162 (9th Cir. 2007).\\n\\n Here, Read’s counsel and the jury were adequately\\ninformed of the missing element. The indictment’s express\\nreference to 18 U.S.C. § 113(a) put Read’s counsel on notice\\n\\f UNITED STATES V. READ 11\\n\\nof the jurisdictional element. See Arnt, 474 F.3d at 1162\\n(counsel “had notice” of the element “from the statute itself,\\nspecifically cited in the indictment[]”). And, the jury was\\nproperly instructed that the government had to prove the\\nassaults took place in FCI-Phoenix. See United States v.\\nWarren, 984 F.2d 325, 327–28 (9th Cir. 1993). Thus, the\\nindictment was sufficient to support Read’s conviction.\\n\\nII. Whether the District Court Erred by Permitting\\n Counsel to Present an Insanity Defense\\n\\n Read claims the district court violated his Sixth\\nAmendment right to present a defense of his own choosing\\nby terminating self-representation and permitting counsel to\\nmake an insanity defense. Reviewing de novo, see United\\nStates v. Brown, 859 F.3d 730, 733 (9th Cir. 2017), we hold\\nthat a district court commits reversible error by permitting\\ndefense counsel to present a defense of insanity over a\\ncompetent defendant’s clear rejection of that defense.\\n\\n The trial judge here undoubtedly faced a difficult\\ndilemma: whether to permit a defendant, competent and\\nallowed self-representation but clearly mentally ill, to\\neschew a plausible defense of insanity in favor of one based\\nin delusion and certain to fail. Although several of our sister\\ncircuits and many state courts recognize a defendant’s right\\nto refuse a defense of insanity, 2 we today face a question of\\n\\n 2\\n These discussions of the insanity defense arise in a variety of\\nprocedural postures; some are dicta. See Petrovich v. Leonardo, 229\\nF.3d 384, 386–87 (2d Cir. 2000); United States v. Marble, 940 F.2d\\n1543, 1547 (D.C. Cir. 1991); Lowenfield v. Phelps, 817 F.2d 285, 292\\n(5th Cir. 1987); Foster v. Strickland, 707 F.2d 1339, 1343 & n.3 (11th\\nCir. 1983); Snider v. Cunningham, 292 F.2d 683, 685 (4th Cir. 1961);\\nState v. Fayle, 658 P.2d 218, 228–29 (Ariz. Ct. App. 1982); People v.\\nFrierson, 705 P.2d 396, 401–05 (Cal. 1985); People v. Redmond, 94 Cal.\\n\\f12 UNITED STATES V. READ\\n\\nfirst impression in this circuit. 3 The trial judge faced this\\nquestion without the benefit of McCoy, which was not\\ndecided until after Read filed his opening brief in this appeal.\\nWe hold that, in light of McCoy, Read’s Sixth Amendment\\nrights were violated when the trial judge permitted counsel\\nto present an insanity defense against Read’s clear objection.\\n\\n McCoy affirmed the defendant’s autonomy to determine\\nthe “objectives” of a defense, 138 S. Ct. at 1508, a right the\\nSupreme Court recognized in Faretta v. California, 422 U.S.\\n806 (1975). A represented defendant surrenders control over\\ntactical decisions at trial while retaining the right to be the\\n“master” of his or her own defense. See Faretta, 422 U.S. at\\n820. Counsel can choose “the objections to make, the\\nwitnesses to call, and the arguments to advance.” Gonzalez\\nv. United States, 553 U.S. 242, 249 (2008). But the\\n\\nRptr. 543, 548–49 (Cal. Ct. App. 1971); Jacobs v. Commonwealth, 870\\nS.W.2d 412, 418 (Ky. 1994), overruled on other grounds by St. Clair v.\\nCommonwealth, 451 S.W.3d 597 (Ky. 2014); State v. Lowenfield, 495\\nSo. 2d 1245, 1252 (La. 1985); Commonwealth v. Simpson, 689 N.E. 2d\\n824, 830–31 (Mass. Ct. App. 1998), rev’d on other grounds, 704 N.E.2d\\n1131 (1999); Treece v. State, 547 A.2d 1054, 1062 (Md. Ct. App. 1988);\\nState v. Gorthy, 145 A.3d 146, 157 (N.J. 2016); People v. Morton, 570\\nN.Y.S.2d 846 (N.Y. App. Div. 1991); People v. MacDowell, 508 N.Y.S.\\n2d 870 (N.Y. Sup. Ct. 1986); State v. Payne, 808 S.E. 2d 476, 486 (N.C.\\nCt. App. 2017); State v. Tenace, 700 N.E. 2d 899, 908 (Ohio Ct. App.\\n1997); State v. Peterson, 689 P.2d 985, 992 (Or. Ct. App. 1984); State v.\\nBean, 762 A.2d 1259, 1266–67 (Vt. 2000); State v. Jones, 664 P.2d 1216,\\n1219–21 (Wash. 1983); State v. Higa, 685 P.2d 1117, 1119–20 (Wash.\\nCt. App. 1984); State v. Felton, 329 N.W.2d 161, 174–75 (Wis. 1983);\\nMcLaren v. State, 407 P.3d 1200, 1212–16 (Wy. 2017); but see\\nHendricks v. People, 10 P.3d 1231, 1241–44 (Colo. 2000) (permitting\\ncounsel, with leave of court, to impose insanity defense on defendant\\nover defendant’s objection).\\n 3\\n We expressly declined to decide whether such a right exists in\\nUnited States v. Kaczynski, 239 F.3d 1108, 1118–19 (9th Cir. 2001).\\n\\f UNITED STATES V. READ 13\\n\\ndefendant retains “ultimate authority to make certain\\nfundamental decisions regarding the case, as to whether to\\nplead guilty, waive a jury, testify in his or her own behalf, or\\ntake an appeal[.]” Jones v. Barnes, 463 U.S. 745, 751\\n(1983). As the Court explained in McCoy, the latter category\\nof decisions “are not strategic choices about how best to\\nachieve a client’s objectives; they are choices about what the\\nclient’s objectives in fact are.” 138 S. Ct. at 1508–09.\\n\\n In McCoy, the Court held that the decision of whether to\\nadmit guilt, even in the face of overwhelming evidence, is\\none of the choices that must remain with the defendant.\\nMcCoy was charged with murdering his estranged wife’s\\nmother, stepfather, and son; his counsel concluded the only\\nway to avoid the death penalty would be to concede\\nMcCoy’s guilt at both the guilt and penalty phases of his\\ntrial. Id. at 1505–07. McCoy clearly instructed his counsel\\nnot to concede his guilt, preferring a defense—every bit as\\nbizarre as Read’s—that “the victims were killed by the local\\npolice and that he had been framed by a farflung conspiracy\\nof state and federal officials, reaching from Louisiana to\\nIdaho” and “that even his attorney and the trial judge had\\njoined the plot.” Id. at 1513 (Alito, J., dissenting). Counsel\\nnonetheless told the guilt and penalty phase juries that\\nMcCoy was guilty, and the jury returned three death\\nverdicts. Id. at 1506–07.\\n\\n The Supreme Court held that counsel’s actions violated\\nMcCoy’s Sixth Amendment “autonomy” right and\\nremanded the case for a new trial. Id. at 1510–11. The Court\\nnoted that, although a concession of guilt might have been\\nMcCoy’s best chance at avoiding the death penalty,\\n\\n the client may not share that objective. He\\n may wish to avoid, above all else, the\\n opprobrium that comes with admitting he\\n\\f14 UNITED STATES V. READ\\n\\n killed family members. Or he may hold life\\n in prison not worth living and prefer to risk\\n death for any hope, however small, of\\n exoneration. When a client expressly asserts\\n that the objective of “his defence” is to\\n maintain innocence of the charged criminal\\n acts, his lawyer must abide by that objective\\n and may not override it by conceding guilt.\\n\\nId. at 1508–09 (citations omitted).\\n\\n McCoy’s emphasis on the defendant’s autonomy\\nstrongly suggests that counsel cannot impose an insanity\\ndefense on a non-consenting defendant. An insanity defense\\nis tantamount to a concession of guilt. See McLaren v. State,\\n407 P.3d 1200, 1213 (Wyo. 2017). Moreover, a defense of\\ninsanity, like a concession of guilt, carries grave personal\\nconsequences that go beyond the sphere of trial tactics. A\\ndefendant may not wish to plead insane because of a firmly\\nheld “feeling that he was not mentally ill at the time of the\\ncrime.” Id. Just as conceding guilt might carry\\n“opprobrium” that a defendant might “wish to avoid, above\\nall else,” McCoy, 138 S. Ct. at 1508, “a defendant, with good\\nreason, may choose to avoid the stigma of insanity.”\\nFrendak v. United States, 408 A.2d 364, 377 (D.C. Ct. App.\\n1979). A defendant may also prefer a remote chance of\\nexoneration to the prospect of “indefinite commitment to a\\nstate institution.” Treece v. State, 547 A.2d 1054, 1060 (Md.\\n1988).\\n\\n The government’s arguments to the contrary are not\\npersuasive. First, the government claims that this case does\\nnot implicate the McCoy “objectives” because Read and his\\ncounsel “agreed on the fundamental objective of the defense:\\nto persuade the jury that [Read] was not capable of being\\nmentally responsible for the assault.” This argument goes\\n\\f UNITED STATES V. READ 15\\n\\ntoo far. While the best way to present the argument that “the\\ndevil made me do it” may well be through the “rubric” of an\\ninsanity defense, Read’s goal was not merely to persuade the\\njury, in the best way possible, that he was not responsible for\\nthe alleged assaults. To the contrary, he repeatedly\\nemphasized that an insanity defense was inconsistent with\\nhis goals.\\n\\n Second, the government argues there is no right to refuse\\nan insanity defense beyond the “core” McCoy right to\\nmaintain factual innocence. Because an insanity defense is\\ntantamount to an admission of guilt, the government\\nconcedes that imposing it on a defendant who wishes to\\nmaintain factual innocence might violate the right\\nannounced in McCoy. But, the government argues, when the\\ndefendant’s preferred defense is not one of factual\\ninnocence, no separate right to refuse an insanity defense\\napplies.\\n\\n This argument fails because pleading insanity has grave,\\npersonal implications that are separate from its functional\\nequivalence to a guilty plea. True, one reason that an\\ninsanity defense should not be imposed on a defendant is that\\nit can sometimes directly violate the McCoy right to maintain\\ninnocence. However, even where this concern is absent, the\\ndefendant’s choice to avoid contradicting his own deeply\\npersonal belief that he is sane, as well as to avoid the risk of\\nconfinement in a mental institution and the social stigma\\nassociated with an assertion or adjudication of insanity, are\\nstill present. These considerations go beyond mere trial\\ntactics and so must be left with the defendant.\\n\\n The district court’s violation of a defendant’s Sixth\\nAmendment right to choose his or her defense is a structural\\nerror, and the proper remedy is a new trial. See McCoy, 138\\nS. Ct. at 1511.\\n\\f16 UNITED STATES V. READ\\n\\nIII. Whether the District Court Erred by Appointing\\n Counsel\\n\\n Read also argues that the district court should not have\\nrevoked his pro se status under Indiana v. Edwards, 554 U.S.\\n164 (2008), on the basis of his “decidedly bizarre” behavior\\nand “nonsensical” arguments. We find that the district\\ncourt’s Edwards holding was not erroneous.\\n\\n Trial judges may “insist upon representation by counsel\\nfor those competent enough to stand trial . . . but who still\\nsuffer from severe mental illness to the point where they are\\nnot competent to conduct trial proceedings by themselves.”\\nId. at 178. The Court specified no single standard for such\\n“gray-area” cases because a trial judge “will often prove best\\nable to make more fine-tuned mental capacity decisions,\\ntailored to the individualized circumstances of a particular\\ndefendant.” Id.at 177.\\n\\n However, we have identified several considerations that\\nbear on the Edwards inquiry. One is the psychiatric\\nevidence. Reports evaluating whether a defendant meets the\\nlesser standard of competency to stand trial “are of limited\\nvalue,” United States v. Ferguson, 560 F.3d 1060, 1068 (9th\\nCir. 2009), but reports evaluating whether the defendant can\\nrepresent himself or herself are more helpful, see United\\nStates v. Thompson, 587 F.3d 1165, 1173 (9th Cir. 2009).\\nAnother issue is the defendant’s behavior. Behavior that is\\n“decidedly bizarre,” Ferguson, 560 F.3d at 1068, or\\n“disruptive or defiant,” United States v. Johnson, 610 F.3d\\n1138, 1144 (9th Cir. 2010), weighs in favor of appointing\\ncounsel. Conduct indicating the defendant is “acutely aware\\nof what was occurring” cuts the other way. Thompson, 587\\nF.3d at 1173. A defendant’s “bizarre and wholly ineffective\\nbehavior” while proceeding pro se indicates that counsel\\nshould be appointed. Ferguson, 560 F.3d at 1069.\\n\\f UNITED STATES V. READ 17\\n\\n Read’s claim that the district court applied the wrong\\nlegal standard fails. The district court explained that, in the\\ncontext of a mentally ill defendant, the Edwards standard is\\nmet when “the defendant’s behavior is ‘decidedly bizarre’\\nand his arguments in defense to the charges against him are\\nnonsensical.’” The court discussed the considerations raised\\nin Ferguson, Thompson, and Johnson: Read’s psychiatric\\nevaluations, behavior at trial, and ability to present an\\neffective defense.\\n\\n Nor did the district court abuse its discretion by\\nappointing counsel. The district court noted Read’s\\ndiagnosis of schizophrenia and his unusual beliefs. It also\\nnoted that Read’s behavior was “bizarre,” and that his\\nproposed defense would likely be “wholly ineffective.”\\nRead cites to Johnson for the proposition that a defendant\\nhas a right to espouse “unorthodox defenses . . . to the bitter\\nend.” See 610 F.3d at 1147. But there, we found “no\\nevidence that [the defendant] was suffering from any mental\\ndisorder.” Id. at 1146. Considering Read’s mental illness\\nand behavior, the district court did not abuse its discretion\\nby appointing counsel.\\n\\nIV. Whether Read’s Trial Violated the Speedy Trial Act\\n\\n Finally, Read claims the indictment must be dismissed\\nbecause his trial did not comply with the Speedy Trial Act.\\nWe find this claim waived.\\n\\n An indictment may be dismissed if trial does not\\ncommence within seventy days from the filing of the\\nindictment, not counting days properly excluded under the\\nAct. 18 U.S.C. §§ 3161–62. However, the “[f]ailure of the\\ndefendant to move for dismissal prior to trial . . . shall\\nconstitute a waiver of the right to dismissal under this\\nsection.” Id. § 3162(a)(2).\\n\\f18 UNITED STATES V. READ\\n\\n Read waived his Speedy Trial Act claim. The sole\\ninstance in which the issue was arguably raised occurred at\\na status conference during which the government sought an\\nadditional psychiatric evaluation. Read’s counsel said, “Mr.\\nRead has indicated that he would like to proceed to trial and\\nhe would prefer no further delays.” But, Read’s counsel\\ndeclined to object to the evaluation, noting that, because\\nRead would be in custody until 2020 on his robbery\\nsentence, there would be no prejudice in granting the motion.\\n\\n This objection did not constitute a motion to dismiss the\\nindictment. A defendant’s “passing reference” to the Speedy\\nTrial Act is inadequate to preserve the statutory claim. See\\nUnited States v. Brown, 761 F.2d 1272, 1276–77 (9th Cir.\\n1985). Read’s objection was even more cursory; it was\\ndirected more to the mental health evaluation than to delay,\\nand did not mention his Speedy Trial rights at all. Cf. Hall,\\n181 F.3d at 1060 (defendant preserved his claim with an in\\npropria persona motion to dismiss stating,\\n“Notwithstanding the actions of appointed counsel,\\ndefendant has never waived or given up any of the rights to\\na Speedy Trial . . .”). 4\\n\\n\\n\\n\\n 4\\n To the extent Read suggests his counsel reiterated his objection in\\na February 27, 2017 filing, his claim is meritless. That filing expressed\\nno objection at all to a subsequent delay, notwithstanding Read’s claim\\nthat the filing was “carefully worded” to imply Read’s objection.\\n\\f UNITED STATES V. READ 19\\n\\n CONCLUSION\\n\\n The district court committed structural error by\\npermitting counsel to present an insanity defense over\\nRead’s clear objections. 5, 6\\n\\n REVERSED and REMANDED.\\n\\n\\n\\n\\n 5\\n At this time, we decline the government’s invitation to limit a\\ndefendant’s right to refuse an insanity defense in cases where the\\ndefendant is mentally ill. But see Hendricks v. People, 10 P.3d 1231,\\n1241–44 (Colo. 2000) (on statutory grounds, permitting counsel to\\npresent an insanity defense over defendant’s clear objection where the\\ndefendant’s reasons for rejecting the defense did not pass a test of “basic\\nrationality.”).\\n\\n 6\\n All pending motions are denied as moot.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4380670/\", \"author_raw\": \"Opinion by Judge Hawkins\"}]}"},"judge1":{"kind":"string","value":"MICHAEL DALY HAWKINS"},"judge2":{"kind":"string","value":"MILAN D SMITH JR"},"judge3":{"kind":"string","value":"ANDREW D HURWITZ"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4603417/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. Jonathan Read\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-14\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 17-10439\n Plaintiff-Appellee,\n D.C. No.\n v. 2:15-cr-00933-\n DLR-1\nJONATHAN LEE READ,\n Defendant-Appellant. OPINION\n\n\n Appeal from the United States District Court\n for the District of Arizona\n Douglas L. Rayes, District Judge, Presiding\n\n Argued and Submitted February 8, 2019\n Arizona State University, Phoenix\n\n Filed March 14, 2019\n\nBefore: Michael Daly Hawkins, Milan D. Smith, Jr., and\n Andrew D. Hurwitz, Circuit Judges.\n\n Opinion by Judge Hawkins\n\f2 UNITED STATES V. READ\n\n SUMMARY *\n\n\n Criminal Law\n\n The panel reversed a criminal judgment and remanded in\na case in which the defendant—who stabbed his cellmate\nwhile serving a sentence in the Federal Correctional Institute\nin Phoenix, Arizona—was convicted of assault with a deadly\nweapon with intent to bodily harm, and assault with a deadly\nweapon resulting in serious bodily injury, in violation of 18\nU.S.C. § 113(a).\n\n The panel rejected the defendant’s arguments that the\ngovernment did not sufficiently allege and prove the\njurisdiction element of § 113(a), which prohibits assaults\n“within the special maritime and territorial jurisdiction of the\nUnited States.”\n\n The panel held that in light of McCoy v. Louisiana, 138\nS. Ct. 1500 (2018), a district court commits reversible error\nby permitting defense counsel to present a defense of\ninsanity over a competent defendant’s clear rejection of that\ndefense.\n\n The panel held that the district court did not err by\nrevoking his pro se status under Indiana v. Edwards, 554\nU.S. 164 (2008), on the basis of his “decidedly bizarre” and\n“nonsensical” arguments.\n\n\n\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f UNITED STATES V. READ 3\n\n The panel held that the defendant waived his claimed\nthat the indictment must be dismissed because his trial did\nnot comply with the Speedy Trial Act.\n\n\n COUNSEL\n\nDavina T. Chen (argued), Glendale, California, for\nDefendant-Appellant.\n\nPeter S. Kozinets (argued), Assistant United States Attorney;\nKrissa M. Lanham, Deputy Appellate Chief; Elizabeth A.\nStrange, First Assistant United States Attorney, District of\nArizona; United States Attorney’s Office Phoenix, Arizona;\nfor Plaintiff-Appellee.\n\n\n OPINION\n\nHAWKINS, Senior Circuit Judge:\n\n We address a question of first impression: whether a\ncriminal defendant has the Sixth Amendment right to\ndemand that counsel not present an insanity defense. We\nhold that McCoy v. Louisiana, 138 S. Ct. 1500 (2018),\nrequires under the facts of this case that the demand be\nhonored. We therefore reverse and remand for a new trial.\n\n FACTS AND PROCEEDINGS BELOW\n\n a. Read was charged with assaulting his cellmate.\n\n While serving a sentence for attempted robbery in the\nFederal Correctional Institute in Phoenix, Arizona (“FCI-\nPhoenix”), Jonathan Lee Read stabbed his cellmate thirteen\ntimes with a homemade knife. Read’s relationship with his\n\f4 UNITED STATES V. READ\n\ncellmates had previously been without tension. When\ndetained, Read claimed he had no memory of the attack.\n\n Read was charged with one count each of assault with a\ndeadly weapon with intent to do bodily harm, and assault\nwith a deadly weapon resulting in serious bodily injury, both\nin violation of 18 U.S.C. § 113(a), which prohibits assaults\n“within the special maritime and territorial jurisdiction of the\nUnited States.” Id.\n\n b. Read was ultimately deemed competent to stand\n trial.\n\n Upon the motion of appointed counsel, Read was\ninitially admitted to the Metropolitan Detention Center in\nLos Angeles for evaluation of his competency. Two months\nlater, Dr. Lesli Johnson, Ph.D., a forensic psychologist,\nissued a report diagnosing Read with schizophrenia and\nsevere cannabis use disorder. Dr. Johnson’s diagnosis of\nschizophrenia was based on Read’s delusional thoughts\nregarding Christianity, Satan, and demonization. She noted\nthat Read appeared to respond to internal stimuli and that his\nthought and speech patterns were disorganized when he\ndiscussed his delusionally-focused beliefs. The district court\naccepted the report, found Read incompetent to stand trial,\nand ordered him committed for hospitalization, treatment,\nand restoration, pursuant to 18 U.S.C. §§ 4241 and 4247.\nRead was admitted to the Federal Medical Center in\nSpringfield, Missouri.\n\n Four months later, Dr. Allison Schenk, Ph.D., a clinical\npsychologist, determined that Read was competent to stand\ntrial. Dr. Schenk also took note of Read’s unusual beliefs,\nreporting that Read claimed he was experiencing the\nsensation of having his stomach cut with a knife, caused by\na person’s use of a “voodoo doll” against him. Dr. Schenk\n\f UNITED STATES V. READ 5\n\ndetermined that Read suffered from schizotypal personality\ndisorder and cannabis use disorder, but nonetheless\ndetermined that Read was competent to stand trial.\n\n Based on Dr. Schenk’s evaluation, and without objection\nfrom defense counsel, the court found Read competent to\nstand trial.\n\n c. Read was evaluated to determine his sanity during\n the assault.\n\n Read’s counsel arranged for an examination by Dr. John\nR. Walker III, Psy.D., a neuropsychologist, to assess his state\nof mind at the time of the alleged assault. Dr. Walker\nreported many of the same behaviors as had Drs. Johnson\nand Schenk. He concluded that Read’s psychosis rendered\nhim unable to form criminal intent, and that Read was likely\nstill psychotic at the time of the examination.\n\n Read’s appointed counsel then filed a Notice of Insanity\nDefense. In response, the government requested an\nexamination of Read pursuant to 18 U.S.C. § 4242(a). The\ncourt granted the government’s motion.\n\n Read was admitted to the Federal Medical Center in\nButner, North Carolina (“FMC-Butner”) for the\nexamination. FMC-Butner eventually issued a report\nconcluding that Read was not insane at the time of the\nalleged offense. The report was prepared by Sumandeep\nKaur, a doctoral psychology intern, under the supervision of\nforensic psychologist Dr. Angela Walden Weaver, Ph.D.\nMs. Kaur reported that, during the evaluation, Read claimed\nhe was suffering from “demonization” rather than mental\nillness. She diagnosed Read with schizotypal personality\ndisorder and cannabis use disorder, and opined that Read\nwas able to appreciate the nature, quality, and wrongfulness\n\f6 UNITED STATES V. READ\n\nof his alleged criminal acts. The district court ordered\nRead’s return to the District of Arizona.\n\n d. Read successfully moved to proceed without counsel.\n\n Read then asked to proceed without counsel. Following\na Faretta hearing, 1 the court held that Read had a right to do\nso, finding that he “knowingly and voluntarily waived the\nright to counsel.” Read’s appointed counsel was named as\nstandby counsel.\n\n e. The court vacated its order and reappointed counsel\n for Read.\n\n Prior to the final trial management conference, advisory\ncounsel told the government that Read might abandon an\ninsanity defense in favor of a defense based on demonic\npossession. At the final pretrial conference, Read did just\nthat. The judge subsequently asked Read if he intended to\ncall any witnesses to discuss his mental condition. Read said\nthat he would call Dr. Walker, the neuropathologist who had\nopined that Read was insane at the time of the alleged\nassault. However, shortly thereafter, Read told the court, “I\ncompletely withdraw the insanity. That’s not an option for\nme.” The judge asked Read to clarify, and Read confirmed\nthat he did not wish to present an insanity defense.\n\n The government responded that, if Read was\nwithdrawing his insanity defense, then Dr. Walker could\noffer no relevant testimony. When the court asked him what\ndefense he intended to pursue, Read responded:\n\n\n\n 1\n See Faretta v. California, 422 U.S. 806 (1975).\n\f UNITED STATES V. READ 7\n\n My civil rights have been violated. I’ve been\n tied in with so many other people that I have\n nothing to do with. Incarcerated. My faith,\n my belief system is not—what is that—\n extreme Islamic. I believe in—I’m a believer\n in Christianity. And there’s been a lot of\n things going on in prison that the prison\n system itself has brought upon itself.\n\nThe court responded that Dr. Walker’s testimony about his\nmental condition would not be relevant to such a defense.\nRead then said he wanted Dr. Walker to testify about cases\nof demonic possession that he had observed in inmates\nbesides Read. The court affirmed that such testimony would\nnot be relevant. Read responded, “All right. I guess we can\nscratch—scratch that.”\n\n The court then asked Read if he wished to continue\nrepresenting himself. Read said that he did. The court asked\nRead’s standby counsel, if he had any concerns about Read’s\ncompetence to proceed without counsel. Mr. Williams said\nthat he was concerned because Read did not seem to\nunderstand the legal distinction between a defense of\ninsanity and his proposed defense. The court again asked\nstandby counsel if he thought Read was competent to\nrepresent himself. Counsel said that he did not know.\n\n The court responded that it would consider whether\nRead’s standby counsel should be reappointed. Standby\ncounsel explained that he would present an insanity defense\nif reappointed; he noted that the very reason that Read had\nwanted to proceed pro se in the first place was because he\ndid not want an insanity defense.\n\n Over Read’s objection, the district court reappointed\nstandby counsel to act as Read’s counsel. The court noted,\n\f8 UNITED STATES V. READ\n\n“[T]he Constitution permits [judges] to insist upon\nrepresentation by counsel for those competent enough to\nstand trial[,] . . . but who still suffer from severe mental\nillness to the point where they are not competent to conduct\ntrial proceedings by themselves.” The court reasoned that\n“[t]his standard is met where the defendant’s behavior is\n‘decidedly bizarre’ and his arguments in defense to the\ncharges against him are nonsensical.” The court found that\nRead’s “beliefs are bizarre and his representation will be\nwholly ineffective,” because “[h]is anticipated defense, that\nhe is possessed by demons and that other inmates are also\npossessed, is not a legal defense and is based on his bizarre\nbeliefs.” Despite these comments, at no point did the district\ncourt revisit Read’s competency.\n\n Counsel ably, but unsuccessfully, proceeded to present\nan insanity defense at trial. Read was convicted and\nsentenced to concurrent 82-month terms. This timely appeal\nfollowed.\n\n DISCUSSION\n\nI. Whether the Jurisdictional Element Was Alleged and\n Proved\n\n Read was charged under 18 U.S.C. § 113(a), which\nprohibits an assault “within the special maritime and\nterritorial jurisdiction of the United States.” Id. This\njurisdiction includes “[a]ny lands reserved or acquired for\nthe use of the United States, and under the exclusive or\nconcurrent jurisdiction thereof, or any place purchased or\notherwise acquired by the United States by consent of the\nlegislature of the State in which the same shall be, for the\nerection of a fort, magazine, arsenal, dockyard, or other\nneedful building.” 18 U.S.C. § 7(3).\n\f UNITED STATES V. READ 9\n\n Read argues that the government presented insufficient\nevidence that the assault took place “within the special\nmaritime and territorial jurisdiction of the United States.”\nRead also claims the indictment must be dismissed for\nfailing to state the jurisdictional element of the offenses.\nNeither argument is persuasive.\n\n a. Sufficiency of the Evidence\n\n The existence of federal jurisdiction over the place in\nwhich the offense occurred is an element of the offenses\ndefined at 18 U.S.C. § 113(a), which must be proved to the\njury beyond a reasonable doubt. See United States v.\nGaudin, 515 U.S. 506, 510 (1995). Read claims that the\ngovernment’s failure to submit historical documents\nestablishing FCI-Phoenix’s jurisdictional status means that\nit failed to prove the jurisdictional component of the charged\noffenses. But, while historical documents can be sufficient\nto prove that land is subject to the “special maritime and\nterritorial jurisdiction of the United States,” they are not\nnecessary. Our cases addressing convictions under statutes\nwhose jurisdictional component demands proof that a bank\nis federally insured are instructive; we have stressed in that\ncontext that while documentary evidence can be used to\nestablish jurisdiction, “bank employees’ uncontradicted\ntestimony of a bank’s insured status can sufficiently support\nthe jury’s conclusion that this element was proven beyond a\nreasonable doubt.” United States v. Corbin, 972 F.2d 271,\n272 (9th Cir. 1992) (per curiam) (citing United States v.\nCampbell, 616 F.2d 1151, 1153 (9th Cir. 1980). Likewise,\nuncontradicted testimony from inmates or employees at a\nfederal prison can establish the jurisdictional element of 18\nU.S.C. § 113.\n\n The government elicited precisely this type of testimony\nat trial. The victim testified he was an inmate “of the\n\f10 UNITED STATES V. READ\n\nPhoenix federal prison” at the time of the assault. Phillip\nLopez, the officer who responded to the assault, testified he\nworked at the Federal Bureau of Prisons’ male facility in\nPhoenix, and was employed by the United States\nDepartment of Justice, Federal Bureau of Prisons. None of\nthe relevant testimony was objected to or contradicted at\ntrial; nor, indeed, has Read ever alleged that FCI-Phoenix is\nnot, in fact, subject to federal jurisdiction. Nor does Read\ndeny that the assault took place at the prison. A reasonable\njuror could conclude from these statements that FCI-Phoenix\nwas under federal jurisdiction at the time Read allegedly\ncommitted assault. See Jackson v. Virginia, 443 U.S. 307,\n319 (1979) (noting that we must affirm a conviction if, “after\nviewing the evidence in the light most favorable to the\nprosecution, any rational trier of fact could have found the\nessential elements of the crime beyond a reasonable\ndoubt.”).\n\n b. Sufficiency of the Indictment\n\n Read also argues that the indictment should be dismissed\nbecause it omitted the jurisdictional element of the two\ncrimes charged. “When the sufficiency of the indictment is\nchallenged after trial, it is only required that the necessary\nfacts appear in any form or by fair construction can be found\nwithin the terms of the indictment.” United States v.\nVelasco-Medina, 305 F.3d 839, 846 (9th Cir. 2002)\n(citations, brackets, and quotation marks omitted). “A\ndefendant is not prejudiced where her counsel has notice of\nthe omitted element and the jury is properly instructed\nregarding the missing element.” United States v. Arnt, 474\nF.3d 1159, 1162 (9th Cir. 2007).\n\n Here, Read’s counsel and the jury were adequately\ninformed of the missing element. The indictment’s express\nreference to 18 U.S.C. § 113(a) put Read’s counsel on notice\n\f UNITED STATES V. READ 11\n\nof the jurisdictional element. See Arnt, 474 F.3d at 1162\n(counsel “had notice” of the element “from the statute itself,\nspecifically cited in the indictment[]”). And, the jury was\nproperly instructed that the government had to prove the\nassaults took place in FCI-Phoenix. See United States v.\nWarren, 984 F.2d 325, 327–28 (9th Cir. 1993). Thus, the\nindictment was sufficient to support Read’s conviction.\n\nII. Whether the District Court Erred by Permitting\n Counsel to Present an Insanity Defense\n\n Read claims the district court violated his Sixth\nAmendment right to present a defense of his own choosing\nby terminating self-representation and permitting counsel to\nmake an insanity defense. Reviewing de novo, see United\nStates v. Brown, 859 F.3d 730, 733 (9th Cir. 2017), we hold\nthat a district court commits reversible error by permitting\ndefense counsel to present a defense of insanity over a\ncompetent defendant’s clear rejection of that defense.\n\n The trial judge here undoubtedly faced a difficult\ndilemma: whether to permit a defendant, competent and\nallowed self-representation but clearly mentally ill, to\neschew a plausible defense of insanity in favor of one based\nin delusion and certain to fail. Although several of our sister\ncircuits and many state courts recognize a defendant’s right\nto refuse a defense of insanity, 2 we today face a question of\n\n 2\n These discussions of the insanity defense arise in a variety of\nprocedural postures; some are dicta. See Petrovich v. Leonardo, 229\nF.3d 384, 386–87 (2d Cir. 2000); United States v. Marble, 940 F.2d\n1543, 1547 (D.C. Cir. 1991); Lowenfield v. Phelps, 817 F.2d 285, 292\n(5th Cir. 1987); Foster v. Strickland, 707 F.2d 1339, 1343 & n.3 (11th\nCir. 1983); Snider v. Cunningham, 292 F.2d 683, 685 (4th Cir. 1961);\nState v. Fayle, 658 P.2d 218, 228–29 (Ariz. Ct. App. 1982); People v.\nFrierson, 705 P.2d 396, 401–05 (Cal. 1985); People v. Redmond, 94 Cal.\n\f12 UNITED STATES V. READ\n\nfirst impression in this circuit. 3 The trial judge faced this\nquestion without the benefit of McCoy, which was not\ndecided until after Read filed his opening brief in this appeal.\nWe hold that, in light of McCoy, Read’s Sixth Amendment\nrights were violated when the trial judge permitted counsel\nto present an insanity defense against Read’s clear objection.\n\n McCoy affirmed the defendant’s autonomy to determine\nthe “objectives” of a defense, 138 S. Ct. at 1508, a right the\nSupreme Court recognized in Faretta v. California, 422 U.S.\n806 (1975). A represented defendant surrenders control over\ntactical decisions at trial while retaining the right to be the\n“master” of his or her own defense. See Faretta, 422 U.S. at\n820. Counsel can choose “the objections to make, the\nwitnesses to call, and the arguments to advance.” Gonzalez\nv. United States, 553 U.S. 242, 249 (2008). But the\n\nRptr. 543, 548–49 (Cal. Ct. App. 1971); Jacobs v. Commonwealth, 870\nS.W.2d 412, 418 (Ky. 1994), overruled on other grounds by St. Clair v.\nCommonwealth, 451 S.W.3d 597 (Ky. 2014); State v. Lowenfield, 495\nSo. 2d 1245, 1252 (La. 1985); Commonwealth v. Simpson, 689 N.E. 2d\n824, 830–31 (Mass. Ct. App. 1998), rev’d on other grounds, 704 N.E.2d\n1131 (1999); Treece v. State, 547 A.2d 1054, 1062 (Md. Ct. App. 1988);\nState v. Gorthy, 145 A.3d 146, 157 (N.J. 2016); People v. Morton, 570\nN.Y.S.2d 846 (N.Y. App. Div. 1991); People v. MacDowell, 508 N.Y.S.\n2d 870 (N.Y. Sup. Ct. 1986); State v. Payne, 808 S.E. 2d 476, 486 (N.C.\nCt. App. 2017); State v. Tenace, 700 N.E. 2d 899, 908 (Ohio Ct. App.\n1997); State v. Peterson, 689 P.2d 985, 992 (Or. Ct. App. 1984); State v.\nBean, 762 A.2d 1259, 1266–67 (Vt. 2000); State v. Jones, 664 P.2d 1216,\n1219–21 (Wash. 1983); State v. Higa, 685 P.2d 1117, 1119–20 (Wash.\nCt. App. 1984); State v. Felton, 329 N.W.2d 161, 174–75 (Wis. 1983);\nMcLaren v. State, 407 P.3d 1200, 1212–16 (Wy. 2017); but see\nHendricks v. People, 10 P.3d 1231, 1241–44 (Colo. 2000) (permitting\ncounsel, with leave of court, to impose insanity defense on defendant\nover defendant’s objection).\n 3\n We expressly declined to decide whether such a right exists in\nUnited States v. Kaczynski, 239 F.3d 1108, 1118–19 (9th Cir. 2001).\n\f UNITED STATES V. READ 13\n\ndefendant retains “ultimate authority to make certain\nfundamental decisions regarding the case, as to whether to\nplead guilty, waive a jury, testify in his or her own behalf, or\ntake an appeal[.]” Jones v. Barnes, 463 U.S. 745, 751\n(1983). As the Court explained in McCoy, the latter category\nof decisions “are not strategic choices about how best to\nachieve a client’s objectives; they are choices about what the\nclient’s objectives in fact are.” 138 S. Ct. at 1508–09.\n\n In McCoy, the Court held that the decision of whether to\nadmit guilt, even in the face of overwhelming evidence, is\none of the choices that must remain with the defendant.\nMcCoy was charged with murdering his estranged wife’s\nmother, stepfather, and son; his counsel concluded the only\nway to avoid the death penalty would be to concede\nMcCoy’s guilt at both the guilt and penalty phases of his\ntrial. Id. at 1505–07. McCoy clearly instructed his counsel\nnot to concede his guilt, preferring a defense—every bit as\nbizarre as Read’s—that “the victims were killed by the local\npolice and that he had been framed by a farflung conspiracy\nof state and federal officials, reaching from Louisiana to\nIdaho” and “that even his attorney and the trial judge had\njoined the plot.” Id. at 1513 (Alito, J., dissenting). Counsel\nnonetheless told the guilt and penalty phase juries that\nMcCoy was guilty, and the jury returned three death\nverdicts. Id. at 1506–07.\n\n The Supreme Court held that counsel’s actions violated\nMcCoy’s Sixth Amendment “autonomy” right and\nremanded the case for a new trial. Id. at 1510–11. The Court\nnoted that, although a concession of guilt might have been\nMcCoy’s best chance at avoiding the death penalty,\n\n the client may not share that objective. He\n may wish to avoid, above all else, the\n opprobrium that comes with admitting he\n\f14 UNITED STATES V. READ\n\n killed family members. Or he may hold life\n in prison not worth living and prefer to risk\n death for any hope, however small, of\n exoneration. When a client expressly asserts\n that the objective of “his defence” is to\n maintain innocence of the charged criminal\n acts, his lawyer must abide by that objective\n and may not override it by conceding guilt.\n\nId. at 1508–09 (citations omitted).\n\n McCoy’s emphasis on the defendant’s autonomy\nstrongly suggests that counsel cannot impose an insanity\ndefense on a non-consenting defendant. An insanity defense\nis tantamount to a concession of guilt. See McLaren v. State,\n407 P.3d 1200, 1213 (Wyo. 2017). Moreover, a defense of\ninsanity, like a concession of guilt, carries grave personal\nconsequences that go beyond the sphere of trial tactics. A\ndefendant may not wish to plead insane because of a firmly\nheld “feeling that he was not mentally ill at the time of the\ncrime.” Id. Just as conceding guilt might carry\n“opprobrium” that a defendant might “wish to avoid, above\nall else,” McCoy, 138 S. Ct. at 1508, “a defendant, with good\nreason, may choose to avoid the stigma of insanity.”\nFrendak v. United States, 408 A.2d 364, 377 (D.C. Ct. App.\n1979). A defendant may also prefer a remote chance of\nexoneration to the prospect of “indefinite commitment to a\nstate institution.” Treece v. State, 547 A.2d 1054, 1060 (Md.\n1988).\n\n The government’s arguments to the contrary are not\npersuasive. First, the government claims that this case does\nnot implicate the McCoy “objectives” because Read and his\ncounsel “agreed on the fundamental objective of the defense:\nto persuade the jury that [Read] was not capable of being\nmentally responsible for the assault.” This argument goes\n\f UNITED STATES V. READ 15\n\ntoo far. While the best way to present the argument that “the\ndevil made me do it” may well be through the “rubric” of an\ninsanity defense, Read’s goal was not merely to persuade the\njury, in the best way possible, that he was not responsible for\nthe alleged assaults. To the contrary, he repeatedly\nemphasized that an insanity defense was inconsistent with\nhis goals.\n\n Second, the government argues there is no right to refuse\nan insanity defense beyond the “core” McCoy right to\nmaintain factual innocence. Because an insanity defense is\ntantamount to an admission of guilt, the government\nconcedes that imposing it on a defendant who wishes to\nmaintain factual innocence might violate the right\nannounced in McCoy. But, the government argues, when the\ndefendant’s preferred defense is not one of factual\ninnocence, no separate right to refuse an insanity defense\napplies.\n\n This argument fails because pleading insanity has grave,\npersonal implications that are separate from its functional\nequivalence to a guilty plea. True, one reason that an\ninsanity defense should not be imposed on a defendant is that\nit can sometimes directly violate the McCoy right to maintain\ninnocence. However, even where this concern is absent, the\ndefendant’s choice to avoid contradicting his own deeply\npersonal belief that he is sane, as well as to avoid the risk of\nconfinement in a mental institution and the social stigma\nassociated with an assertion or adjudication of insanity, are\nstill present. These considerations go beyond mere trial\ntactics and so must be left with the defendant.\n\n The district court’s violation of a defendant’s Sixth\nAmendment right to choose his or her defense is a structural\nerror, and the proper remedy is a new trial. See McCoy, 138\nS. Ct. at 1511.\n\f16 UNITED STATES V. READ\n\nIII. Whether the District Court Erred by Appointing\n Counsel\n\n Read also argues that the district court should not have\nrevoked his pro se status under Indiana v. Edwards, 554 U.S.\n164 (2008), on the basis of his “decidedly bizarre” behavior\nand “nonsensical” arguments. We find that the district\ncourt’s Edwards holding was not erroneous.\n\n Trial judges may “insist upon representation by counsel\nfor those competent enough to stand trial . . . but who still\nsuffer from severe mental illness to the point where they are\nnot competent to conduct trial proceedings by themselves.”\nId. at 178. The Court specified no single standard for such\n“gray-area” cases because a trial judge “will often prove best\nable to make more fine-tuned mental capacity decisions,\ntailored to the individualized circumstances of a particular\ndefendant.” Id.at 177.\n\n However, we have identified several considerations that\nbear on the Edwards inquiry. One is the psychiatric\nevidence. Reports evaluating whether a defendant meets the\nlesser standard of competency to stand trial “are of limited\nvalue,” United States v. Ferguson, 560 F.3d 1060, 1068 (9th\nCir. 2009), but reports evaluating whether the defendant can\nrepresent himself or herself are more helpful, see United\nStates v. Thompson, 587 F.3d 1165, 1173 (9th Cir. 2009).\nAnother issue is the defendant’s behavior. Behavior that is\n“decidedly bizarre,” Ferguson, 560 F.3d at 1068, or\n“disruptive or defiant,” United States v. Johnson, 610 F.3d\n1138, 1144 (9th Cir. 2010), weighs in favor of appointing\ncounsel. Conduct indicating the defendant is “acutely aware\nof what was occurring” cuts the other way. Thompson, 587\nF.3d at 1173. A defendant’s “bizarre and wholly ineffective\nbehavior” while proceeding pro se indicates that counsel\nshould be appointed. Ferguson, 560 F.3d at 1069.\n\f UNITED STATES V. READ 17\n\n Read’s claim that the district court applied the wrong\nlegal standard fails. The district court explained that, in the\ncontext of a mentally ill defendant, the Edwards standard is\nmet when “the defendant’s behavior is ‘decidedly bizarre’\nand his arguments in defense to the charges against him are\nnonsensical.’” The court discussed the considerations raised\nin Ferguson, Thompson, and Johnson: Read’s psychiatric\nevaluations, behavior at trial, and ability to present an\neffective defense.\n\n Nor did the district court abuse its discretion by\nappointing counsel. The district court noted Read’s\ndiagnosis of schizophrenia and his unusual beliefs. It also\nnoted that Read’s behavior was “bizarre,” and that his\nproposed defense would likely be “wholly ineffective.”\nRead cites to Johnson for the proposition that a defendant\nhas a right to espouse “unorthodox defenses . . . to the bitter\nend.” See 610 F.3d at 1147. But there, we found “no\nevidence that [the defendant] was suffering from any mental\ndisorder.” Id. at 1146. Considering Read’s mental illness\nand behavior, the district court did not abuse its discretion\nby appointing counsel.\n\nIV. Whether Read’s Trial Violated the Speedy Trial Act\n\n Finally, Read claims the indictment must be dismissed\nbecause his trial did not comply with the Speedy Trial Act.\nWe find this claim waived.\n\n An indictment may be dismissed if trial does not\ncommence within seventy days from the filing of the\nindictment, not counting days properly excluded under the\nAct. 18 U.S.C. §§ 3161–62. However, the “[f]ailure of the\ndefendant to move for dismissal prior to trial . . . shall\nconstitute a waiver of the right to dismissal under this\nsection.” Id. § 3162(a)(2).\n\f18 UNITED STATES V. READ\n\n Read waived his Speedy Trial Act claim. The sole\ninstance in which the issue was arguably raised occurred at\na status conference during which the government sought an\nadditional psychiatric evaluation. Read’s counsel said, “Mr.\nRead has indicated that he would like to proceed to trial and\nhe would prefer no further delays.” But, Read’s counsel\ndeclined to object to the evaluation, noting that, because\nRead would be in custody until 2020 on his robbery\nsentence, there would be no prejudice in granting the motion.\n\n This objection did not constitute a motion to dismiss the\nindictment. A defendant’s “passing reference” to the Speedy\nTrial Act is inadequate to preserve the statutory claim. See\nUnited States v. Brown, 761 F.2d 1272, 1276–77 (9th Cir.\n1985). Read’s objection was even more cursory; it was\ndirected more to the mental health evaluation than to delay,\nand did not mention his Speedy Trial rights at all. Cf. Hall,\n181 F.3d at 1060 (defendant preserved his claim with an in\npropria persona motion to dismiss stating,\n“Notwithstanding the actions of appointed counsel,\ndefendant has never waived or given up any of the rights to\na Speedy Trial . . .”). 4\n\n\n\n\n 4\n To the extent Read suggests his counsel reiterated his objection in\na February 27, 2017 filing, his claim is meritless. That filing expressed\nno objection at all to a subsequent delay, notwithstanding Read’s claim\nthat the filing was “carefully worded” to imply Read’s objection.\n\f UNITED STATES V. READ 19\n\n CONCLUSION\n\n The district court committed structural error by\npermitting counsel to present an insanity defense over\nRead’s clear objections. 5, 6\n\n REVERSED and REMANDED.\n\n\n\n\n 5\n At this time, we decline the government’s invitation to limit a\ndefendant’s right to refuse an insanity defense in cases where the\ndefendant is mentally ill. But see Hendricks v. People, 10 P.3d 1231,\n1241–44 (Colo. 2000) (on statutory grounds, permitting counsel to\npresent an insanity defense over defendant’s clear objection where the\ndefendant’s reasons for rejecting the defense did not pass a test of “basic\nrationality.”).\n\n 6\n All pending motions are denied as moot.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. Jonathan Read\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-14\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-10439\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 2:15-cr-00933-\\n DLR-1\\nJONATHAN LEE READ,\\n Defendant-Appellant. OPINION\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Douglas L. Rayes, District Judge, Presiding\\n\\n Argued and Submitted February 8, 2019\\n Arizona State University, Phoenix\\n\\n Filed March 14, 2019\\n\\nBefore: Michael Daly Hawkins, Milan D. Smith, Jr., and\\n Andrew D. Hurwitz, Circuit Judges.\\n\\n Opinion by Judge Hawkins\\n\\f2 UNITED STATES V. READ\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n The panel reversed a criminal judgment and remanded in\\na case in which the defendant—who stabbed his cellmate\\nwhile serving a sentence in the Federal Correctional Institute\\nin Phoenix, Arizona—was convicted of assault with a deadly\\nweapon with intent to bodily harm, and assault with a deadly\\nweapon resulting in serious bodily injury, in violation of 18\\nU.S.C. § 113(a).\\n\\n The panel rejected the defendant’s arguments that the\\ngovernment did not sufficiently allege and prove the\\njurisdiction element of § 113(a), which prohibits assaults\\n“within the special maritime and territorial jurisdiction of the\\nUnited States.”\\n\\n The panel held that in light of McCoy v. Louisiana, 138\\nS. Ct. 1500 (2018), a district court commits reversible error\\nby permitting defense counsel to present a defense of\\ninsanity over a competent defendant’s clear rejection of that\\ndefense.\\n\\n The panel held that the district court did not err by\\nrevoking his pro se status under Indiana v. Edwards, 554\\nU.S. 164 (2008), on the basis of his “decidedly bizarre” and\\n“nonsensical” arguments.\\n\\n\\n\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f UNITED STATES V. READ 3\\n\\n The panel held that the defendant waived his claimed\\nthat the indictment must be dismissed because his trial did\\nnot comply with the Speedy Trial Act.\\n\\n\\n COUNSEL\\n\\nDavina T. Chen (argued), Glendale, California, for\\nDefendant-Appellant.\\n\\nPeter S. Kozinets (argued), Assistant United States Attorney;\\nKrissa M. Lanham, Deputy Appellate Chief; Elizabeth A.\\nStrange, First Assistant United States Attorney, District of\\nArizona; United States Attorney’s Office Phoenix, Arizona;\\nfor Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nHAWKINS, Senior Circuit Judge:\\n\\n We address a question of first impression: whether a\\ncriminal defendant has the Sixth Amendment right to\\ndemand that counsel not present an insanity defense. We\\nhold that McCoy v. Louisiana, 138 S. Ct. 1500 (2018),\\nrequires under the facts of this case that the demand be\\nhonored. We therefore reverse and remand for a new trial.\\n\\n FACTS AND PROCEEDINGS BELOW\\n\\n a. Read was charged with assaulting his cellmate.\\n\\n While serving a sentence for attempted robbery in the\\nFederal Correctional Institute in Phoenix, Arizona (“FCI-\\nPhoenix”), Jonathan Lee Read stabbed his cellmate thirteen\\ntimes with a homemade knife. Read’s relationship with his\\n\\f4 UNITED STATES V. READ\\n\\ncellmates had previously been without tension. When\\ndetained, Read claimed he had no memory of the attack.\\n\\n Read was charged with one count each of assault with a\\ndeadly weapon with intent to do bodily harm, and assault\\nwith a deadly weapon resulting in serious bodily injury, both\\nin violation of 18 U.S.C. § 113(a), which prohibits assaults\\n“within the special maritime and territorial jurisdiction of the\\nUnited States.” Id.\\n\\n b. Read was ultimately deemed competent to stand\\n trial.\\n\\n Upon the motion of appointed counsel, Read was\\ninitially admitted to the Metropolitan Detention Center in\\nLos Angeles for evaluation of his competency. Two months\\nlater, Dr. Lesli Johnson, Ph.D., a forensic psychologist,\\nissued a report diagnosing Read with schizophrenia and\\nsevere cannabis use disorder. Dr. Johnson’s diagnosis of\\nschizophrenia was based on Read’s delusional thoughts\\nregarding Christianity, Satan, and demonization. She noted\\nthat Read appeared to respond to internal stimuli and that his\\nthought and speech patterns were disorganized when he\\ndiscussed his delusionally-focused beliefs. The district court\\naccepted the report, found Read incompetent to stand trial,\\nand ordered him committed for hospitalization, treatment,\\nand restoration, pursuant to 18 U.S.C. §§ 4241 and 4247.\\nRead was admitted to the Federal Medical Center in\\nSpringfield, Missouri.\\n\\n Four months later, Dr. Allison Schenk, Ph.D., a clinical\\npsychologist, determined that Read was competent to stand\\ntrial. Dr. Schenk also took note of Read’s unusual beliefs,\\nreporting that Read claimed he was experiencing the\\nsensation of having his stomach cut with a knife, caused by\\na person’s use of a “voodoo doll” against him. Dr. Schenk\\n\\f UNITED STATES V. READ 5\\n\\ndetermined that Read suffered from schizotypal personality\\ndisorder and cannabis use disorder, but nonetheless\\ndetermined that Read was competent to stand trial.\\n\\n Based on Dr. Schenk’s evaluation, and without objection\\nfrom defense counsel, the court found Read competent to\\nstand trial.\\n\\n c. Read was evaluated to determine his sanity during\\n the assault.\\n\\n Read’s counsel arranged for an examination by Dr. John\\nR. Walker III, Psy.D., a neuropsychologist, to assess his state\\nof mind at the time of the alleged assault. Dr. Walker\\nreported many of the same behaviors as had Drs. Johnson\\nand Schenk. He concluded that Read’s psychosis rendered\\nhim unable to form criminal intent, and that Read was likely\\nstill psychotic at the time of the examination.\\n\\n Read’s appointed counsel then filed a Notice of Insanity\\nDefense. In response, the government requested an\\nexamination of Read pursuant to 18 U.S.C. § 4242(a). The\\ncourt granted the government’s motion.\\n\\n Read was admitted to the Federal Medical Center in\\nButner, North Carolina (“FMC-Butner”) for the\\nexamination. FMC-Butner eventually issued a report\\nconcluding that Read was not insane at the time of the\\nalleged offense. The report was prepared by Sumandeep\\nKaur, a doctoral psychology intern, under the supervision of\\nforensic psychologist Dr. Angela Walden Weaver, Ph.D.\\nMs. Kaur reported that, during the evaluation, Read claimed\\nhe was suffering from “demonization” rather than mental\\nillness. She diagnosed Read with schizotypal personality\\ndisorder and cannabis use disorder, and opined that Read\\nwas able to appreciate the nature, quality, and wrongfulness\\n\\f6 UNITED STATES V. READ\\n\\nof his alleged criminal acts. The district court ordered\\nRead’s return to the District of Arizona.\\n\\n d. Read successfully moved to proceed without counsel.\\n\\n Read then asked to proceed without counsel. Following\\na Faretta hearing, 1 the court held that Read had a right to do\\nso, finding that he “knowingly and voluntarily waived the\\nright to counsel.” Read’s appointed counsel was named as\\nstandby counsel.\\n\\n e. The court vacated its order and reappointed counsel\\n for Read.\\n\\n Prior to the final trial management conference, advisory\\ncounsel told the government that Read might abandon an\\ninsanity defense in favor of a defense based on demonic\\npossession. At the final pretrial conference, Read did just\\nthat. The judge subsequently asked Read if he intended to\\ncall any witnesses to discuss his mental condition. Read said\\nthat he would call Dr. Walker, the neuropathologist who had\\nopined that Read was insane at the time of the alleged\\nassault. However, shortly thereafter, Read told the court, “I\\ncompletely withdraw the insanity. That’s not an option for\\nme.” The judge asked Read to clarify, and Read confirmed\\nthat he did not wish to present an insanity defense.\\n\\n The government responded that, if Read was\\nwithdrawing his insanity defense, then Dr. Walker could\\noffer no relevant testimony. When the court asked him what\\ndefense he intended to pursue, Read responded:\\n\\n\\n\\n 1\\n See Faretta v. California, 422 U.S. 806 (1975).\\n\\f UNITED STATES V. READ 7\\n\\n My civil rights have been violated. I’ve been\\n tied in with so many other people that I have\\n nothing to do with. Incarcerated. My faith,\\n my belief system is not—what is that—\\n extreme Islamic. I believe in—I’m a believer\\n in Christianity. And there’s been a lot of\\n things going on in prison that the prison\\n system itself has brought upon itself.\\n\\nThe court responded that Dr. Walker’s testimony about his\\nmental condition would not be relevant to such a defense.\\nRead then said he wanted Dr. Walker to testify about cases\\nof demonic possession that he had observed in inmates\\nbesides Read. The court affirmed that such testimony would\\nnot be relevant. Read responded, “All right. I guess we can\\nscratch—scratch that.”\\n\\n The court then asked Read if he wished to continue\\nrepresenting himself. Read said that he did. The court asked\\nRead’s standby counsel, if he had any concerns about Read’s\\ncompetence to proceed without counsel. Mr. Williams said\\nthat he was concerned because Read did not seem to\\nunderstand the legal distinction between a defense of\\ninsanity and his proposed defense. The court again asked\\nstandby counsel if he thought Read was competent to\\nrepresent himself. Counsel said that he did not know.\\n\\n The court responded that it would consider whether\\nRead’s standby counsel should be reappointed. Standby\\ncounsel explained that he would present an insanity defense\\nif reappointed; he noted that the very reason that Read had\\nwanted to proceed pro se in the first place was because he\\ndid not want an insanity defense.\\n\\n Over Read’s objection, the district court reappointed\\nstandby counsel to act as Read’s counsel. The court noted,\\n\\f8 UNITED STATES V. READ\\n\\n“[T]he Constitution permits [judges] to insist upon\\nrepresentation by counsel for those competent enough to\\nstand trial[,] . . . but who still suffer from severe mental\\nillness to the point where they are not competent to conduct\\ntrial proceedings by themselves.” The court reasoned that\\n“[t]his standard is met where the defendant’s behavior is\\n‘decidedly bizarre’ and his arguments in defense to the\\ncharges against him are nonsensical.” The court found that\\nRead’s “beliefs are bizarre and his representation will be\\nwholly ineffective,” because “[h]is anticipated defense, that\\nhe is possessed by demons and that other inmates are also\\npossessed, is not a legal defense and is based on his bizarre\\nbeliefs.” Despite these comments, at no point did the district\\ncourt revisit Read’s competency.\\n\\n Counsel ably, but unsuccessfully, proceeded to present\\nan insanity defense at trial. Read was convicted and\\nsentenced to concurrent 82-month terms. This timely appeal\\nfollowed.\\n\\n DISCUSSION\\n\\nI. Whether the Jurisdictional Element Was Alleged and\\n Proved\\n\\n Read was charged under 18 U.S.C. § 113(a), which\\nprohibits an assault “within the special maritime and\\nterritorial jurisdiction of the United States.” Id. This\\njurisdiction includes “[a]ny lands reserved or acquired for\\nthe use of the United States, and under the exclusive or\\nconcurrent jurisdiction thereof, or any place purchased or\\notherwise acquired by the United States by consent of the\\nlegislature of the State in which the same shall be, for the\\nerection of a fort, magazine, arsenal, dockyard, or other\\nneedful building.” 18 U.S.C. § 7(3).\\n\\f UNITED STATES V. READ 9\\n\\n Read argues that the government presented insufficient\\nevidence that the assault took place “within the special\\nmaritime and territorial jurisdiction of the United States.”\\nRead also claims the indictment must be dismissed for\\nfailing to state the jurisdictional element of the offenses.\\nNeither argument is persuasive.\\n\\n a. Sufficiency of the Evidence\\n\\n The existence of federal jurisdiction over the place in\\nwhich the offense occurred is an element of the offenses\\ndefined at 18 U.S.C. § 113(a), which must be proved to the\\njury beyond a reasonable doubt. See United States v.\\nGaudin, 515 U.S. 506, 510 (1995). Read claims that the\\ngovernment’s failure to submit historical documents\\nestablishing FCI-Phoenix’s jurisdictional status means that\\nit failed to prove the jurisdictional component of the charged\\noffenses. But, while historical documents can be sufficient\\nto prove that land is subject to the “special maritime and\\nterritorial jurisdiction of the United States,” they are not\\nnecessary. Our cases addressing convictions under statutes\\nwhose jurisdictional component demands proof that a bank\\nis federally insured are instructive; we have stressed in that\\ncontext that while documentary evidence can be used to\\nestablish jurisdiction, “bank employees’ uncontradicted\\ntestimony of a bank’s insured status can sufficiently support\\nthe jury’s conclusion that this element was proven beyond a\\nreasonable doubt.” United States v. Corbin, 972 F.2d 271,\\n272 (9th Cir. 1992) (per curiam) (citing United States v.\\nCampbell, 616 F.2d 1151, 1153 (9th Cir. 1980). Likewise,\\nuncontradicted testimony from inmates or employees at a\\nfederal prison can establish the jurisdictional element of 18\\nU.S.C. § 113.\\n\\n The government elicited precisely this type of testimony\\nat trial. The victim testified he was an inmate “of the\\n\\f10 UNITED STATES V. READ\\n\\nPhoenix federal prison” at the time of the assault. Phillip\\nLopez, the officer who responded to the assault, testified he\\nworked at the Federal Bureau of Prisons’ male facility in\\nPhoenix, and was employed by the United States\\nDepartment of Justice, Federal Bureau of Prisons. None of\\nthe relevant testimony was objected to or contradicted at\\ntrial; nor, indeed, has Read ever alleged that FCI-Phoenix is\\nnot, in fact, subject to federal jurisdiction. Nor does Read\\ndeny that the assault took place at the prison. A reasonable\\njuror could conclude from these statements that FCI-Phoenix\\nwas under federal jurisdiction at the time Read allegedly\\ncommitted assault. See Jackson v. Virginia, 443 U.S. 307,\\n319 (1979) (noting that we must affirm a conviction if, “after\\nviewing the evidence in the light most favorable to the\\nprosecution, any rational trier of fact could have found the\\nessential elements of the crime beyond a reasonable\\ndoubt.”).\\n\\n b. Sufficiency of the Indictment\\n\\n Read also argues that the indictment should be dismissed\\nbecause it omitted the jurisdictional element of the two\\ncrimes charged. “When the sufficiency of the indictment is\\nchallenged after trial, it is only required that the necessary\\nfacts appear in any form or by fair construction can be found\\nwithin the terms of the indictment.” United States v.\\nVelasco-Medina, 305 F.3d 839, 846 (9th Cir. 2002)\\n(citations, brackets, and quotation marks omitted). “A\\ndefendant is not prejudiced where her counsel has notice of\\nthe omitted element and the jury is properly instructed\\nregarding the missing element.” United States v. Arnt, 474\\nF.3d 1159, 1162 (9th Cir. 2007).\\n\\n Here, Read’s counsel and the jury were adequately\\ninformed of the missing element. The indictment’s express\\nreference to 18 U.S.C. § 113(a) put Read’s counsel on notice\\n\\f UNITED STATES V. READ 11\\n\\nof the jurisdictional element. See Arnt, 474 F.3d at 1162\\n(counsel “had notice” of the element “from the statute itself,\\nspecifically cited in the indictment[]”). And, the jury was\\nproperly instructed that the government had to prove the\\nassaults took place in FCI-Phoenix. See United States v.\\nWarren, 984 F.2d 325, 327–28 (9th Cir. 1993). Thus, the\\nindictment was sufficient to support Read’s conviction.\\n\\nII. Whether the District Court Erred by Permitting\\n Counsel to Present an Insanity Defense\\n\\n Read claims the district court violated his Sixth\\nAmendment right to present a defense of his own choosing\\nby terminating self-representation and permitting counsel to\\nmake an insanity defense. Reviewing de novo, see United\\nStates v. Brown, 859 F.3d 730, 733 (9th Cir. 2017), we hold\\nthat a district court commits reversible error by permitting\\ndefense counsel to present a defense of insanity over a\\ncompetent defendant’s clear rejection of that defense.\\n\\n The trial judge here undoubtedly faced a difficult\\ndilemma: whether to permit a defendant, competent and\\nallowed self-representation but clearly mentally ill, to\\neschew a plausible defense of insanity in favor of one based\\nin delusion and certain to fail. Although several of our sister\\ncircuits and many state courts recognize a defendant’s right\\nto refuse a defense of insanity, 2 we today face a question of\\n\\n 2\\n These discussions of the insanity defense arise in a variety of\\nprocedural postures; some are dicta. See Petrovich v. Leonardo, 229\\nF.3d 384, 386–87 (2d Cir. 2000); United States v. Marble, 940 F.2d\\n1543, 1547 (D.C. Cir. 1991); Lowenfield v. Phelps, 817 F.2d 285, 292\\n(5th Cir. 1987); Foster v. Strickland, 707 F.2d 1339, 1343 & n.3 (11th\\nCir. 1983); Snider v. Cunningham, 292 F.2d 683, 685 (4th Cir. 1961);\\nState v. Fayle, 658 P.2d 218, 228–29 (Ariz. Ct. App. 1982); People v.\\nFrierson, 705 P.2d 396, 401–05 (Cal. 1985); People v. Redmond, 94 Cal.\\n\\f12 UNITED STATES V. READ\\n\\nfirst impression in this circuit. 3 The trial judge faced this\\nquestion without the benefit of McCoy, which was not\\ndecided until after Read filed his opening brief in this appeal.\\nWe hold that, in light of McCoy, Read’s Sixth Amendment\\nrights were violated when the trial judge permitted counsel\\nto present an insanity defense against Read’s clear objection.\\n\\n McCoy affirmed the defendant’s autonomy to determine\\nthe “objectives” of a defense, 138 S. Ct. at 1508, a right the\\nSupreme Court recognized in Faretta v. California, 422 U.S.\\n806 (1975). A represented defendant surrenders control over\\ntactical decisions at trial while retaining the right to be the\\n“master” of his or her own defense. See Faretta, 422 U.S. at\\n820. Counsel can choose “the objections to make, the\\nwitnesses to call, and the arguments to advance.” Gonzalez\\nv. United States, 553 U.S. 242, 249 (2008). But the\\n\\nRptr. 543, 548–49 (Cal. Ct. App. 1971); Jacobs v. Commonwealth, 870\\nS.W.2d 412, 418 (Ky. 1994), overruled on other grounds by St. Clair v.\\nCommonwealth, 451 S.W.3d 597 (Ky. 2014); State v. Lowenfield, 495\\nSo. 2d 1245, 1252 (La. 1985); Commonwealth v. Simpson, 689 N.E. 2d\\n824, 830–31 (Mass. Ct. App. 1998), rev’d on other grounds, 704 N.E.2d\\n1131 (1999); Treece v. State, 547 A.2d 1054, 1062 (Md. Ct. App. 1988);\\nState v. Gorthy, 145 A.3d 146, 157 (N.J. 2016); People v. Morton, 570\\nN.Y.S.2d 846 (N.Y. App. Div. 1991); People v. MacDowell, 508 N.Y.S.\\n2d 870 (N.Y. Sup. Ct. 1986); State v. Payne, 808 S.E. 2d 476, 486 (N.C.\\nCt. App. 2017); State v. Tenace, 700 N.E. 2d 899, 908 (Ohio Ct. App.\\n1997); State v. Peterson, 689 P.2d 985, 992 (Or. Ct. App. 1984); State v.\\nBean, 762 A.2d 1259, 1266–67 (Vt. 2000); State v. Jones, 664 P.2d 1216,\\n1219–21 (Wash. 1983); State v. Higa, 685 P.2d 1117, 1119–20 (Wash.\\nCt. App. 1984); State v. Felton, 329 N.W.2d 161, 174–75 (Wis. 1983);\\nMcLaren v. State, 407 P.3d 1200, 1212–16 (Wy. 2017); but see\\nHendricks v. People, 10 P.3d 1231, 1241–44 (Colo. 2000) (permitting\\ncounsel, with leave of court, to impose insanity defense on defendant\\nover defendant’s objection).\\n 3\\n We expressly declined to decide whether such a right exists in\\nUnited States v. Kaczynski, 239 F.3d 1108, 1118–19 (9th Cir. 2001).\\n\\f UNITED STATES V. READ 13\\n\\ndefendant retains “ultimate authority to make certain\\nfundamental decisions regarding the case, as to whether to\\nplead guilty, waive a jury, testify in his or her own behalf, or\\ntake an appeal[.]” Jones v. Barnes, 463 U.S. 745, 751\\n(1983). As the Court explained in McCoy, the latter category\\nof decisions “are not strategic choices about how best to\\nachieve a client’s objectives; they are choices about what the\\nclient’s objectives in fact are.” 138 S. Ct. at 1508–09.\\n\\n In McCoy, the Court held that the decision of whether to\\nadmit guilt, even in the face of overwhelming evidence, is\\none of the choices that must remain with the defendant.\\nMcCoy was charged with murdering his estranged wife’s\\nmother, stepfather, and son; his counsel concluded the only\\nway to avoid the death penalty would be to concede\\nMcCoy’s guilt at both the guilt and penalty phases of his\\ntrial. Id. at 1505–07. McCoy clearly instructed his counsel\\nnot to concede his guilt, preferring a defense—every bit as\\nbizarre as Read’s—that “the victims were killed by the local\\npolice and that he had been framed by a farflung conspiracy\\nof state and federal officials, reaching from Louisiana to\\nIdaho” and “that even his attorney and the trial judge had\\njoined the plot.” Id. at 1513 (Alito, J., dissenting). Counsel\\nnonetheless told the guilt and penalty phase juries that\\nMcCoy was guilty, and the jury returned three death\\nverdicts. Id. at 1506–07.\\n\\n The Supreme Court held that counsel’s actions violated\\nMcCoy’s Sixth Amendment “autonomy” right and\\nremanded the case for a new trial. Id. at 1510–11. The Court\\nnoted that, although a concession of guilt might have been\\nMcCoy’s best chance at avoiding the death penalty,\\n\\n the client may not share that objective. He\\n may wish to avoid, above all else, the\\n opprobrium that comes with admitting he\\n\\f14 UNITED STATES V. READ\\n\\n killed family members. Or he may hold life\\n in prison not worth living and prefer to risk\\n death for any hope, however small, of\\n exoneration. When a client expressly asserts\\n that the objective of “his defence” is to\\n maintain innocence of the charged criminal\\n acts, his lawyer must abide by that objective\\n and may not override it by conceding guilt.\\n\\nId. at 1508–09 (citations omitted).\\n\\n McCoy’s emphasis on the defendant’s autonomy\\nstrongly suggests that counsel cannot impose an insanity\\ndefense on a non-consenting defendant. An insanity defense\\nis tantamount to a concession of guilt. See McLaren v. State,\\n407 P.3d 1200, 1213 (Wyo. 2017). Moreover, a defense of\\ninsanity, like a concession of guilt, carries grave personal\\nconsequences that go beyond the sphere of trial tactics. A\\ndefendant may not wish to plead insane because of a firmly\\nheld “feeling that he was not mentally ill at the time of the\\ncrime.” Id. Just as conceding guilt might carry\\n“opprobrium” that a defendant might “wish to avoid, above\\nall else,” McCoy, 138 S. Ct. at 1508, “a defendant, with good\\nreason, may choose to avoid the stigma of insanity.”\\nFrendak v. United States, 408 A.2d 364, 377 (D.C. Ct. App.\\n1979). A defendant may also prefer a remote chance of\\nexoneration to the prospect of “indefinite commitment to a\\nstate institution.” Treece v. State, 547 A.2d 1054, 1060 (Md.\\n1988).\\n\\n The government’s arguments to the contrary are not\\npersuasive. First, the government claims that this case does\\nnot implicate the McCoy “objectives” because Read and his\\ncounsel “agreed on the fundamental objective of the defense:\\nto persuade the jury that [Read] was not capable of being\\nmentally responsible for the assault.” This argument goes\\n\\f UNITED STATES V. READ 15\\n\\ntoo far. While the best way to present the argument that “the\\ndevil made me do it” may well be through the “rubric” of an\\ninsanity defense, Read’s goal was not merely to persuade the\\njury, in the best way possible, that he was not responsible for\\nthe alleged assaults. To the contrary, he repeatedly\\nemphasized that an insanity defense was inconsistent with\\nhis goals.\\n\\n Second, the government argues there is no right to refuse\\nan insanity defense beyond the “core” McCoy right to\\nmaintain factual innocence. Because an insanity defense is\\ntantamount to an admission of guilt, the government\\nconcedes that imposing it on a defendant who wishes to\\nmaintain factual innocence might violate the right\\nannounced in McCoy. But, the government argues, when the\\ndefendant’s preferred defense is not one of factual\\ninnocence, no separate right to refuse an insanity defense\\napplies.\\n\\n This argument fails because pleading insanity has grave,\\npersonal implications that are separate from its functional\\nequivalence to a guilty plea. True, one reason that an\\ninsanity defense should not be imposed on a defendant is that\\nit can sometimes directly violate the McCoy right to maintain\\ninnocence. However, even where this concern is absent, the\\ndefendant’s choice to avoid contradicting his own deeply\\npersonal belief that he is sane, as well as to avoid the risk of\\nconfinement in a mental institution and the social stigma\\nassociated with an assertion or adjudication of insanity, are\\nstill present. These considerations go beyond mere trial\\ntactics and so must be left with the defendant.\\n\\n The district court’s violation of a defendant’s Sixth\\nAmendment right to choose his or her defense is a structural\\nerror, and the proper remedy is a new trial. See McCoy, 138\\nS. Ct. at 1511.\\n\\f16 UNITED STATES V. READ\\n\\nIII. Whether the District Court Erred by Appointing\\n Counsel\\n\\n Read also argues that the district court should not have\\nrevoked his pro se status under Indiana v. Edwards, 554 U.S.\\n164 (2008), on the basis of his “decidedly bizarre” behavior\\nand “nonsensical” arguments. We find that the district\\ncourt’s Edwards holding was not erroneous.\\n\\n Trial judges may “insist upon representation by counsel\\nfor those competent enough to stand trial . . . but who still\\nsuffer from severe mental illness to the point where they are\\nnot competent to conduct trial proceedings by themselves.”\\nId. at 178. The Court specified no single standard for such\\n“gray-area” cases because a trial judge “will often prove best\\nable to make more fine-tuned mental capacity decisions,\\ntailored to the individualized circumstances of a particular\\ndefendant.” Id.at 177.\\n\\n However, we have identified several considerations that\\nbear on the Edwards inquiry. One is the psychiatric\\nevidence. Reports evaluating whether a defendant meets the\\nlesser standard of competency to stand trial “are of limited\\nvalue,” United States v. Ferguson, 560 F.3d 1060, 1068 (9th\\nCir. 2009), but reports evaluating whether the defendant can\\nrepresent himself or herself are more helpful, see United\\nStates v. Thompson, 587 F.3d 1165, 1173 (9th Cir. 2009).\\nAnother issue is the defendant’s behavior. Behavior that is\\n“decidedly bizarre,” Ferguson, 560 F.3d at 1068, or\\n“disruptive or defiant,” United States v. Johnson, 610 F.3d\\n1138, 1144 (9th Cir. 2010), weighs in favor of appointing\\ncounsel. Conduct indicating the defendant is “acutely aware\\nof what was occurring” cuts the other way. Thompson, 587\\nF.3d at 1173. A defendant’s “bizarre and wholly ineffective\\nbehavior” while proceeding pro se indicates that counsel\\nshould be appointed. Ferguson, 560 F.3d at 1069.\\n\\f UNITED STATES V. READ 17\\n\\n Read’s claim that the district court applied the wrong\\nlegal standard fails. The district court explained that, in the\\ncontext of a mentally ill defendant, the Edwards standard is\\nmet when “the defendant’s behavior is ‘decidedly bizarre’\\nand his arguments in defense to the charges against him are\\nnonsensical.’” The court discussed the considerations raised\\nin Ferguson, Thompson, and Johnson: Read’s psychiatric\\nevaluations, behavior at trial, and ability to present an\\neffective defense.\\n\\n Nor did the district court abuse its discretion by\\nappointing counsel. The district court noted Read’s\\ndiagnosis of schizophrenia and his unusual beliefs. It also\\nnoted that Read’s behavior was “bizarre,” and that his\\nproposed defense would likely be “wholly ineffective.”\\nRead cites to Johnson for the proposition that a defendant\\nhas a right to espouse “unorthodox defenses . . . to the bitter\\nend.” See 610 F.3d at 1147. But there, we found “no\\nevidence that [the defendant] was suffering from any mental\\ndisorder.” Id. at 1146. Considering Read’s mental illness\\nand behavior, the district court did not abuse its discretion\\nby appointing counsel.\\n\\nIV. Whether Read’s Trial Violated the Speedy Trial Act\\n\\n Finally, Read claims the indictment must be dismissed\\nbecause his trial did not comply with the Speedy Trial Act.\\nWe find this claim waived.\\n\\n An indictment may be dismissed if trial does not\\ncommence within seventy days from the filing of the\\nindictment, not counting days properly excluded under the\\nAct. 18 U.S.C. §§ 3161–62. However, the “[f]ailure of the\\ndefendant to move for dismissal prior to trial . . . shall\\nconstitute a waiver of the right to dismissal under this\\nsection.” Id. § 3162(a)(2).\\n\\f18 UNITED STATES V. READ\\n\\n Read waived his Speedy Trial Act claim. The sole\\ninstance in which the issue was arguably raised occurred at\\na status conference during which the government sought an\\nadditional psychiatric evaluation. Read’s counsel said, “Mr.\\nRead has indicated that he would like to proceed to trial and\\nhe would prefer no further delays.” But, Read’s counsel\\ndeclined to object to the evaluation, noting that, because\\nRead would be in custody until 2020 on his robbery\\nsentence, there would be no prejudice in granting the motion.\\n\\n This objection did not constitute a motion to dismiss the\\nindictment. A defendant’s “passing reference” to the Speedy\\nTrial Act is inadequate to preserve the statutory claim. See\\nUnited States v. Brown, 761 F.2d 1272, 1276–77 (9th Cir.\\n1985). Read’s objection was even more cursory; it was\\ndirected more to the mental health evaluation than to delay,\\nand did not mention his Speedy Trial rights at all. Cf. Hall,\\n181 F.3d at 1060 (defendant preserved his claim with an in\\npropria persona motion to dismiss stating,\\n“Notwithstanding the actions of appointed counsel,\\ndefendant has never waived or given up any of the rights to\\na Speedy Trial . . .”). 4\\n\\n\\n\\n\\n 4\\n To the extent Read suggests his counsel reiterated his objection in\\na February 27, 2017 filing, his claim is meritless. That filing expressed\\nno objection at all to a subsequent delay, notwithstanding Read’s claim\\nthat the filing was “carefully worded” to imply Read’s objection.\\n\\f UNITED STATES V. READ 19\\n\\n CONCLUSION\\n\\n The district court committed structural error by\\npermitting counsel to present an insanity defense over\\nRead’s clear objections. 5, 6\\n\\n REVERSED and REMANDED.\\n\\n\\n\\n\\n 5\\n At this time, we decline the government’s invitation to limit a\\ndefendant’s right to refuse an insanity defense in cases where the\\ndefendant is mentally ill. But see Hendricks v. People, 10 P.3d 1231,\\n1241–44 (Colo. 2000) (on statutory grounds, permitting counsel to\\npresent an insanity defense over defendant’s clear objection where the\\ndefendant’s reasons for rejecting the defense did not pass a test of “basic\\nrationality.”).\\n\\n 6\\n All pending motions are denied as moot.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1291,"cells":{"id":{"kind":"number","value":4603418,"string":"4,603,418"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Michael McNeil v. Sherwood School District 88j"},"decision_date":{"kind":"string","value":"2019-03-14"},"docket_number":{"kind":"string","value":"17-35500"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Raymond C. Fisher and Consuelo M. Callahan, Circuit Judges, and Cathy Ann Bencivengo,* District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PER CURIAM\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nMICHAEL MCNEIL, guardian ad litem No. 17-35500\\nas to CLM; CLM, by and through\\nMichael McNeil guardian ad litem; D.C. No.\\nJULIE MCNEIL, individually, 3:15-cv-01098-\\n Plaintiffs-Appellants, SB\\n v.\\n OPINION\\nSHERWOOD SCHOOL DISTRICT 88J;\\nHEATHER H. CORDIE; KEN BELL;\\nBRIAN BAILEY; PETER MILLER; GARY\\nBENNETT,\\n Defendants-Appellees.\\n\\n Appeal from the United States District Court\\n for the District of Oregon\\n Michael H. Simon, District Judge, Presiding\\n\\n Argued and Submitted October 9, 2018\\n Portland, Oregon\\n\\n Filed March 14, 2019\\n\\n Before: Raymond C. Fisher and Consuelo M. Callahan,\\nCircuit Judges, and Cathy Ann Bencivengo,* District Judge.\\n\\n Per Curiam Opinion\\n\\n\\n *\\n The Honorable Cathy Ann Bencivengo, United States District Judge\\nfor the Southern District of California, sitting by designation.\\n\\f2 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\n SUMMARY**\\n\\n\\n Civil Rights\\n\\n The panel affirmed the district court’s summary judgment\\nin favor of a school district in an action brought pursuant to\\n42 U.S.C. § 1983 by student CLM and his parents alleging\\nthat the district violated plaintiffs’ First Amendment and\\nFourteenth Amendment substantive due process rights when\\nit expelled CLM for one year.\\n\\n CLM, then a high school sophomore at Sherwood High\\nSchool, created in his personal journal a hit list of students\\nthat “must die.” When his mother discovered the hit list, she\\ntold a therapist, who informed the police, who told the school\\ndistrict.\\n\\n The panel held that under the particular facts in this case,\\nincluding the nature of the hit list, CLM’s access to firearms,\\nand the close proximity of CLM’s home to the high school,\\nthe decision to discipline CLM for his off-campus speech did\\nnot violate his constitutional right to free speech. The panel\\nheld that when considering whether a school district may\\nconstitutionally regulate off-campus speech, courts must\\ndetermine, based on the totality of the circumstances, whether\\nthe speech bears a sufficient nexus to the school. The panel\\nstated that there is always a sufficient nexus between the\\nspeech and the school when the school district reasonably\\nconcludes that it faces a credible, identifiable threat of school\\nviolence. The panel further held that a student’s lack of intent\\n\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 3\\n\\nto convey his off-campus speech to any third party is relevant\\nto an evaluation of whether the speech constitutes a credible\\nthreat, but is not dispositive.\\n\\n The panel held that the claim brought by CLM’s parents\\nalleging substantive due process violations failed because\\ntheir fundamental right to choose CLM’s educational forum\\nwas not infringed by the School District’s discipline of CLM.\\n\\n\\n COUNSEL\\n\\nAdam S. Heder (argued) and Roger K. Harris, Harris Berne\\nChristensen LLP, Lake Oswego, Oregon, for Plaintiffs-\\nAppellants.\\n\\nBlake H. Fry (argued) and Peter R. Mersereau, Mersereau\\nShannon LLP, Portland, Oregon, for Defendants-Appellees.\\n\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n CLM, then a high school sophomore at Sherwood High\\nSchool (“Sherwood High”), created in his personal journal a\\nhit list of students that “must die.” When his mother\\ndiscovered the hit list, she told a therapist, who informed the\\npolice, who told Sherwood School District (“School\\nDistrict”). The School District expelled CLM for one year.\\nCLM sued the School District primarily on First Amendment\\ngrounds, and the district court held that the expulsion\\nwas constitutional. On appeal, CLM and his parents\\n(“Appellants”) assert that the School District cannot\\n\\f4 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nconstitutionally regulate student speech that the student\\nnever intended to communicate to any third party. We\\nconclude that under the particular facts in this case, including\\nthe nature of the hit list, CLM’s access to firearms, and the\\nclose proximity of CLM’s home to Sherwood High, the\\nSchool District did not violate any of Appellants’ asserted\\nconstitutional rights.\\n\\n I.\\n\\n On or about May 25, 2014, CLM created a “hit list” in his\\npersonal journal, naming 22 Sherwood High students and one\\nformer employee, and stating “I am God” and “All These\\nPeople Must Die.” On September 8, 2014, soon after CLM\\nstarted his junior year at Sherwood High, CLM’s mother,\\nMrs. McNeil, discovered the journal on CLM’s nightstand\\nwhile cleaning CLM’s room. Upon reviewing the journal’s\\ncontents, Mrs. McNeil came across the hit list and additional\\nentries containing graphic depictions of violence.\\nMrs. McNeil made copies of some of the journal entries,\\nincluding the hit list, and sought guidance from a therapist the\\nnext day regarding her findings. The therapist, alarmed by\\nthe entries and believing they triggered her duties as a\\nmandatory reporter, informed the Sherwood Police\\nDepartment (“Sherwood Police”) about the list. The therapist\\ntold Mrs. McNeil to call a local crisis hotline, which she did.\\nThe hotline’s social worker also contacted the Sherwood\\nPolice about CLM’s list.\\n\\n Later that day, the Sherwood Police searched the\\nMcNeils’ residence. Officers found and confiscated several\\nweapons, including a .22 caliber rifle and 525 rounds of\\nammunition belonging to CLM, but the officers found\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 5\\n\\nnothing “to indicate any planning had gone into following\\nthrough with the hit list.”\\n\\n Shortly after the search, CLM and his parents voluntarily\\nwent to the police station, where they provided the police\\nwith a copy of CLM’s hit list. After being read his Miranda\\nwarnings, CLM admitted he created the hit list and that\\n“sometimes he thinks killing people might relieve some of the\\nstress he feels,” but he also stated “he uses the journal to\\nvent” and that “he would never carry out” such thoughts.\\nCLM claimed he created the list about four months prior to its\\ndiscovery. The Sherwood Police determined that no criminal\\ncharges would be brought against CLM at that time.\\n\\n Meanwhile, the Sherwood Police informed the School\\nDistrict of CLM’s hit list, of the fact the police had seized\\nguns from his house, and that CLM’s journal contained\\nadditional entries that graphically depicted school violence.\\nSherwood High’s principal, Ken Bell, was aware that the\\nMcNeils lived “very close” to Sherwood High and assembled\\nan administrative team to address the issue.\\n\\n Consistent with the notification requirements of Oregon\\nRevised Statutes § 339.327(1), the School District’s policies\\nrequired school faculty to notify the parents of students found\\non a hit list within 12 hours of discovery. Within the 12-hour\\nspan, Bell’s team made the necessary calls, which did not\\nidentify CLM as the hit list’s author. Before the final call\\nwas made, the media began contacting the School District to\\ninquire about the list. The School District also learned that\\nCLM’s picture had been posted on social media accounts.\\nUpon realization of the widespread knowledge regarding\\nCLM’s hit list, Bell’s team sent out a recorded voice message\\nto all parents of students in the School District to notify them\\n\\f6 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nof the issue. The message stated that a student of Sherwood\\nHigh had authored a hit list containing 23 names, but that the\\nlist contained no specific threats and the home in which the\\nstudent resided was safe. The School District also issued a\\npress release containing similar information. Thereafter,\\nSherwood High received numerous calls and emails from\\nparents, media outlets, and the public requesting information\\nabout the hit list, CLM’s identity, and whether CLM posed a\\nthreat to others. Some parents demanded to meet with Bell,\\nand other parents had their children leave school early, miss\\nseveral days, or transfer out of the district. A student was\\ncaught on campus with a knife, and he alleged “he needed it\\nto protect himself in light of the hit list.”\\n\\n The School District suspended CLM pending an\\nexpulsion hearing. Bell recommended that CLM be expelled\\nfrom Sherwood High for one year because news of his hit list\\n“significantly disrupted the learning environment at school,”\\nwhich would only be increased by CLM’s return. Sherwood\\nHigh’s associate principal wrote a letter to the McNeils\\nstating that he recommended that CLM be expelled for one\\nyear for making “a threat of violence” that “has caused a\\ndistinct and substantial disruption to the school environment.”\\n\\n On September 22, 2014, an expulsion hearing was held\\nbefore Pete Miller, a hearings officer for the School District.\\nCLM was represented by an attorney at the hearing. Upon\\nreview of the testimony and evidence, Miller adopted the\\nprincipal’s recommendation for expulsion, largely based on\\n“the significant disruption” CLM’s list caused in the school\\nenvironment. After the hearing, the School District’s\\nassociate superintendent sent a decision letter to the McNeils,\\nexplaining that the list constituted “a threat of violence in\\nviolation of the Student Code of Conduct.” The letter\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 7\\n\\nexplained that CLM could not return to school that academic\\nyear, and that the McNeils had the right to appeal the decision\\nto the School District’s board. The McNeils did not appeal.\\n\\n Following his expulsion, CLM received alternative forms\\nof education through the School District, including online\\ncourses, in-person tutoring, and courses at a community\\ncollege. However, CLM had “no immediate or even daily\\naccess to instructors,” and the tutors were “inconsistent and\\nnot dependable,” as they missed appointments and in one\\ninstance failed to follow up for weeks. CLM ultimately had\\nto retake several classes.\\n\\n In June 2015, CLM and the McNeils filed a complaint in\\nthe U.S. District Court for the District of Oregon seeking\\ndamages and other relief under 42 U.S.C. § 1983. CLM\\nsought damages for the School District’s alleged violations of\\nthe Free Speech Clause of the First Amendment and of the\\nEqual Protection and Due Process Clauses of the Fourteenth\\nAmendment.1 The McNeils also brought a substantive due\\nprocess challenge to CLM’s expulsion. The complaint sought\\nexpungement of any documentation or reference to CLM’s\\nsuspension, expulsion, and the circumstances surrounding the\\ndisciplinary proceedings.\\n\\n The School District moved for summary judgment on all\\nclaims, while CLM and the McNeils moved for summary\\njudgment on their free speech claim and substantive due\\nprocess claim, respectively. The district court concluded that\\nthe School District could regulate CLM’s off-campus speech\\nfor three primary reasons: (1) the hit list had a sufficient\\n\\n 1\\n On appeal, CLM has withdrawn his equal protection and\\nprocedural due process claims.\\n\\f8 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nconnection to Sherwood High; (2) school officials could have\\nreasonably foreseen that the effects of the hit list would spill\\nover into the school environment; and (3) the facts in CLM’s\\ncase mirrored those in Wynar v. Douglas County School\\nDistrict, 728 F.3d 1062 (9th Cir. 2013). The district court\\nfurther held that the one-year expulsion did not violate\\nCLM’s First Amendment rights, reasoning that a student-\\nauthored hit list would cause a substantial disruption in any\\nschool community, particularly in light of Oregon’s statutory\\nnotification requirement.\\n\\n The district court also granted the School District\\nsummary judgment on the McNeils’ substantive due process\\nclaim. The district court reasoned that the claim failed\\nbecause it was derivative of CLM’s First Amendment claim\\nand because a school has the authority to discipline students\\non reasonable grounds. The district court determined that\\nalthough the McNeils had a liberty interest in an “initial\\n‘educational venue choice,’” their interest did not extend to\\ntemporary school discipline within that venue, which is an\\nissue “generally committed to the control of the state and\\nlocal authorities.”\\n\\n Appellants timely appealed. On appeal, CLM claims that\\nthe School District lacked authority under the First\\nAmendment to discipline CLM for his hit list. The McNeils\\nalso allege that the School District’s expulsion of CLM\\nviolated their substantive due process right “to be free from\\nstate interference with their choice of [CLM’s] educational\\nforum.”\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 9\\n\\n II.\\n\\n We review “the district court’s grant of summary\\njudgment de novo.” Szajer v. City of Los Angeles, 632 F.3d\\n607, 610 (9th Cir. 2011). The court views “evidence in the\\nlight most favorable to the nonmoving party,” to determine\\n“whether genuine issues of material fact exist.” George v.\\nEdholm, 752 F.3d 1206, 1214 (9th Cir. 2014) (quoting\\nBagdadi v. Nazar, 84 F.3d 1194, 1197 (9th Cir. 1996)). If a\\nrational juror “‘could resolve a genuine issue of material fact\\nin the nonmoving party’s favor,’ summary judgment is\\ninappropriate.” Id. (quoting Bravo v. City of Santa Maria,\\n665 F.3d 1076, 1083 (9th Cir. 2011)).\\n\\n III.\\n\\n “The first inquiry in any § 1983 suit . . . is whether the\\nplaintiff has been deprived of a right ‘secured by the\\nConstitution and laws.’” Baker v. McCollan, 443 U.S. 137,\\n140 (1979). Section 1983 does not create substantive rights,\\nbut rather provides the procedural mechanism for vindicating\\nfederal statutory or constitutional rights. Id. at 145 n.3.\\n\\n A. CLM’s First Amendment Claim\\n\\n The Supreme Court has held that public school students\\ndo not forfeit “their constitutional rights to freedom of speech\\nor expression at the schoolhouse gate.” Tinker v. Des Moines\\nIndep. Cnty. Sch. Dist., 393 U.S. 503, 506 (1969). However,\\ntheir rights are “not automatically coextensive with the rights\\nof adults in other settings.” Wynar, 728 F.3d at 1067 (quoting\\nBethel Sch. Dist. No. 403 v. Fraser, 478 U.S. 675, 682\\n(1986)). Additionally, although they enjoy greater freedom\\nto speak when they are off campus than when they are on\\n\\f10 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\ncampus, their off-campus speech is not necessarily beyond\\nthe reach of a school district’s regulatory authority. Our\\nreview of the School District’s treatment of CLM turns on\\ntwo inquiries: (1) whether the School District could\\npermissibly regulate CLM’s off-campus speech at all; and if\\nso, (2) whether the School District’s decision to expel CLM\\nviolated the First Amendment standard for school regulation\\nof speech set out in Tinker. See C.R. v. Eugene Sch. Dist. 4J,\\n835 F.3d 1142, 1148 (9th Cir. 2016).\\n\\n 1. The School District’s authority to regulate CLM’s\\n off-campus speech\\n\\n We have twice considered the authority of a school\\ndistrict to regulate off-campus speech. In Wynar, a high\\nschool student was expelled because he sent graphic instant\\nmessages to his friends from his home computer. 728 F.3d\\nat 1064–67. The messages expressed threats of school\\nviolence and indicated that the student had access to guns and\\nammunition. Id. at 1065–66. The messages were brought to\\nthe school’s attention by the student’s friends. Id. at 1066.\\nDespite the student’s lack of intent in bringing his speech to\\nthe school campus, we held that his expulsion was\\nconstitutional, reasoning that “when faced with an identifiable\\nthreat of school violence, schools may take disciplinary\\naction in response to off-campus speech . . . .” Id. at 1069.\\n\\n In C.R., a school district suspended a student for sexually\\nharassing two students shortly after school hours in a public\\npark near the school. 835 F.3d at 1145–47. The student\\nargued that the school’s regulation of his harassment was\\nunconstitutional because the harassment occurred in a public,\\noff-campus park. Id. at 1145–46. While we recognized that\\nthe harassment constituted off-campus speech, we held that\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 11\\n\\nthe suspension was constitutional. Id. at 1146. We reasoned\\nthat the speech was closely tied to the school, noting that all\\nof the individuals involved were students and that the incident\\noccurred in close proximity to the school and just minutes\\nafter the end of the school day. Id. at 1150–51. We also\\nnoted that school administrators could reasonably expect the\\neffects of the speech to disrupt the school environment. Id.\\nat 1151.\\n\\n Although our previous cases have not settled on a test for\\ndetermining when a school can constitutionally regulate off-\\ncampus speech, we have stressed the importance of flexibility\\nin dealing with the “myriad of circumstances” schools face.\\nWynar, 728 F.3d at 1069; see also C.R., 835 F.3d at 1150–51.\\nThese cases have noted two approaches: (1) requiring that\\noff-campus speech “have a sufficient ‘nexus’ to the school,”\\nWynar, 728 F.3d at 1068 (quoting Kowalski v. Berkeley Cty.\\nSchs., 652 F.3d 565, 573 (4th Cir. 2011)); and (2) requiring\\n“that it be ‘reasonably foreseeable that the speech will reach\\nthe school community,’” id. (quoting S.J.W. v. Lee’s Summit\\nR-7 Sch. Dist., 696 F.3d 771, 777 (8th Cir. 2012)).\\n\\n We now clarify that courts considering whether a school\\ndistrict may constitutionally regulate off-campus speech must\\ndetermine, based on the totality of the circumstances, whether\\nthe speech bears a sufficient nexus to the school. See C.R.,\\n835 F.3d at 1150–51; Wynar, 728 F.3d at 1069. This test is\\nflexible and fact-specific, but the relevant considerations will\\ninclude (1) the degree and likelihood of harm to the school\\ncaused or augured by the speech, see Wynar, 728 F.3d at\\n1069, (2) whether it was reasonably foreseeable that the\\nspeech would reach and impact the school, see C.R., 835 F.3d\\nat 1150–51; Wynar, 728 F.3d at 1069; Kowalski, 652 F.3d at\\n573, and (3) the relation between the content and context of\\n\\f12 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nthe speech and the school, see C.R., 835 F.3d at 1150–51;\\nWynar, 728 F.3d at 1069.2 There is always a sufficient nexus\\nbetween the speech and the school when the school district\\nreasonably concludes that it faces a credible, identifiable\\nthreat of school violence. See Wynar, 728 F.3d at 1069.\\n\\n CLM attempts to distinguish Wynar on the ground that he\\nhad no intent to communicate the contents of his speech to\\nanyone. That distinction cannot be dispositive. We have\\nrecognized repeatedly that the specter of school violence\\nplaces a weighty social responsibility on school districts to\\nensure that “warning signs” do not turn to tragedy. See\\nLaVine v. Blaine Sch. Dist., 257 F.3d 981, 987 (9th Cir.\\n2001). This responsibility does not mean schools may “expel\\nstudents just because they are ‘loners,’ wear black and play\\nvideo games.” Id. It does mean, however, that a student’s\\nintent, although relevant, does not necessarily define the\\nthreat of violence. We reaffirm our holding in Wynar that\\nregardless of the speaker’s intent or how speech comes to a\\nschool district’s attention, a school district may take\\ndisciplinary action in response to off-campus speech when it\\nreasonably determines that it faces an identifiable and\\ncredible threat of school violence. See Wynar, 728 F.3d at\\n1069.\\n\\n Of course, any information school officials have\\nregarding the intention of the speaker may bear on whether a\\nthreat is credible. The intent to keep a threat private,\\n\\n\\n 2\\n In analyzing foreseeability, we have considered both whether it\\nwas foreseeable to the speaker that the speech would reach the school, see\\nWynar, 728 F.3d at 1068, and whether it was foreseeable to school\\nadministrators that the speech would impact the school community, see\\nC.R., 835 F.3d at 1151.\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 13\\n\\nhowever, is not determinative—it would be absurd to suggest\\nthat secret threats or planning cannot give rise to reasonable\\nconcerns of school violence. See, e.g., Kristin McCauley,\\nArapahoe Cty. Sheriff, Investigative Report: Arapahoe\\nHigh School Case # CT13-44545, at 29–31 (2014),\\nhttps://mediaassets.thedenverchannel.com/document/2014/\\n10/10/AHSShootingReport_8903184_ver1.0.pdf?_ga=2.15\\n5919036.684175427.1543265365-980564543.1543265365\\n(describing password-protected diary entries containing\\nplanning for a fatal 2013 school shooting).\\n\\n Here, the School District reasonably determined CLM\\npresented a credible threat. The School District knew CLM\\nhad identified specific targets, had accentuated his hit list\\nwith the phrases “I am God” and “All These People Must\\nDie,” lived in a gun-owning home close to the school, and\\nhad had thoughts of suicide. The School District also knew\\nthe diary contained other graphic depictions of school\\nviolence. This evidence was sufficient to render the School\\nDistrict’s determination reasonable and to give it authority to\\nregulate CLM’s speech.\\n\\n CLM argues that, once he was in police custody, he\\nceased to pose a safety threat because he was being evaluated\\nin the care of competent authorities. This evaluation,\\nhowever, lasted only one day, and afterward CLM returned\\nhome with limited oversight and his parents retrieved the\\nfamily’s firearms. The police’s decision to release CLM\\ncannot prevent the school from addressing the threat of\\nviolence: a school may take action even where police or\\nmental health professionals have elected not to do so. See\\nLaVine, 257 F.3d at 990 (“It is true that the [police and a\\npsychiatrist] did not believe [the student at issue] should be\\ninvoluntarily committed, but the standard for involuntary\\n\\f14 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\ncommitment is not the same as that for school officials to take\\naction.”). Schools must be permitted to act preventatively,\\ntaking into account other students’ interest in a safe school\\nenvironment. Here, the school reasonably determined the risk\\nwas sufficient to take action.\\n\\n Our test does not allow a school to take disciplinary\\naction in response to just any perceived threat of school\\nviolence arising from off-campus speech. Although our\\nreview of a school’s determinations in this regard is\\ndeferential, “deference does not mean abdication; there are\\nsituations where school officials overstep their bounds and\\nviolate the Constitution.” Id. at 988.\\n\\n In Porter v. Ascension Parish School Board, 393 F.3d 608\\n(5th Cir. 2004), for example, Adam, a high school student,\\ndepicted a military attack on his school in his sketchpad. Id.\\nat 611. There were no students identified in the drawing and\\nAdam stored the sketchpad in his closet at home. Id. Two\\nyears later, Adam’s brother brought the sketchpad to the\\nschool. Id. Another student, with permission from Adam’s\\nbrother, flipped through the sketchpad and, upon coming\\nacross Adam’s drawing of the military attack, exclaimed,\\n“they’re going to blow up [the school].” Id. This brought the\\ndrawing to the attention of school authorities, who expelled\\nAdam for the drawing. Id. at 611–12. The Fifth Circuit held\\nthat Adam’s drawing was protected by the First Amendment\\nbecause it did not constitute “speech on campus or even\\nspeech directed at the campus.” Id. at 615. The Fifth Circuit\\nnoted that for Adam’s drawing to lose First Amendment\\nprotection, something more than “accidental and\\nunintentional exposure to public scrutiny must take place.”\\nId. at 618. The Fifth Circuit reasoned that because Adam’s\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 15\\n\\ndrawing was not intentionally communicated, it was not a\\ntrue threat and therefore not subject to school regulation. Id.\\n\\n We do not adopt the Fifth Circuit’s reasoning, but note\\nthat our approach supports the result in Porter and that there\\nare critical distinctions between the facts here and those in\\nPorter. There, the drawing was two years old, highly\\nfantastical, unspecific, and unaccompanied by other indicia\\nof a violent intent. See id. at 611. There was no evidence\\nAdam had ever had the capacity to carry out the actions\\ndepicted in the drawing. In contrast to this case, it was\\nunlikely that any reasonable administrator could have\\nbelieved the school faced a credible safety threat.\\n\\n Taken as a whole, the three considerations that guide\\napplication of the nexus test support the School District here.\\nFirst, it was reasonable for School District officials to\\nconclude that CLM presented a credible threat of severe harm\\nto the school community. This consideration establishes a\\nsufficient nexus between the speech and the school to permit\\nregulation. See Wynar, 728 F.3d at 1069.\\n\\n Second, once it learned of the hit list, the School District\\ncould reasonably foresee that news of the threat would reach\\nand impact the school and disrupt the school environment.\\nOregon’s notification requirement, Oregon Revised Statutes\\n§ 339.327(1), provides that “[a] superintendent of a school\\ndistrict . . . who has reasonable cause to believe that a person,\\nwhile in a school, is or has been in possession of a list that\\nthreatens harm to other persons” must notify the “parent or\\nguardian of any student whose name appears on the list as a\\ntarget of the harm.” The statute applies to enrolled students\\nwho are or have been in possession of a threatening list,\\nregardless of whether the list is on or off school property. It\\n\\f16 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nwas reasonably foreseeable that parents, once notified as\\nrequired by Oregon law, would inform their children about\\nthe hit list. It was also reasonably foreseeable that\\ninformation about the hit list would cause students to be\\nfearful for their safety. Invariably, these resulting\\nconsequences would create classroom distractions and\\nsubstantial disruptions within the Sherwood High community.\\nAlthough it was not foreseeable to CLM that his speech\\nwould reach the school, a lack of intent to share speech is of\\nminimal weight when, as here, the speech contains a credible\\nthreat of violence directed at the school. Someone planning\\na violent act does not need to advertise it.\\n\\n Third, the content of the speech involved the school. As\\nwe noted in Wynar and C.R., when all involved parties are\\nstudents, that fact “typically counsels in favor of finding that\\na student’s speech was susceptible to discipline.” C.R., 835\\nF.3d at 1150 (citing Wynar, 728 F.3d at 1069). Here, CLM’s\\nhit list contained the names of 22 students and one former\\nteacher of the School District, and thus, presented a particular\\nthreat to the school community. Ordinarily, schools may not\\ndiscipline students for the contents of their private, off-\\ncampus diary entries, any more than they can punish students\\nfor their private thoughts, but schools have a right, indeed an\\nobligation, to address a credible threat of violence involving\\nthe school community.\\n\\n In sum, we conclude that the School District could\\nregulate CLM’s off-campus speech without violating his First\\nAmendment rights. Although CLM may not have foreseen\\nhis speech reaching Sherwood High, the School District,\\nwhen informed of CLM’s hit list, reasonably determined that\\nit faced a credible, identifiable threat of school violence. The\\nspeech bore a sufficient nexus to the school. Accordingly, the\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 17\\n\\nSchool District could take disciplinary action consistent with\\nTinker.\\n\\n 2. The School District’s expulsion of CLM\\n\\n Although some disciplinary action was permissible, there\\nremains the issue of whether the School District’s expulsion\\nwas consistent with the First Amendment. “Under Tinker,\\nschools may restrict speech that ‘might reasonably [lead]\\nschool authorities to forecast substantial disruption of or\\nmaterial interference with school activities’ or that collides\\n‘with the rights of other students to be secure and to be let\\nalone.’” Wynar, 728 F.3d at 1070 (alteration in original)\\n(quoting Tinker, 393 U.S. at 508, 514). This analysis\\nconsiders not just the actions of the disciplined student, but\\nrather “all of the circumstances confronting the school\\nofficials that might reasonably portend disruption.” LaVine,\\n257 F.3d at 989. We have held that “the specter of a school\\nshooting qualifies under either prong of Tinker.” Wynar, 728\\nF.3d at 1070. School districts should be afforded deference\\nin determining what discipline is necessary to maintain safety\\nand order, see id. at 1072, but we have cautioned that\\ndiscipline can extend beyond the scope Tinker allows when\\nit takes on a purely punitive character, see LaVine, 257 F.3d\\nat 992.\\n\\n Here, taking disciplinary action was reasonable under\\nboth Tinker prongs. As discussed, CLM’s hit list identifying\\nspecific individuals constituted a credible threat of school\\nviolence that required the School District to act in furtherance\\nof the safety of its students. Oregon’s statutory notification\\nrequirement further confirms this proposition, as the School\\nDistrict could reasonably foresee that informing parents that\\ntheir children were identified on a student-authored hit list\\n\\f18 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nwould cause a substantial disruption within the Sherwood\\nHigh community. Moreover, CLM’s hit list led to a material\\ndisruption at Sherwood High by causing safety concerns\\namong both students and parents. The record evinces this\\ndisruption by the outcry from parents who demanded to speak\\nwith Sherwood High authorities and by the parents who\\ntransferred or threatened to transfer their child out of the\\nSchool District. Thus, in light of the nature of CLM’s hit list\\nand the widespread knowledge regarding the list in the\\nSherwood High community, it was reasonable for School\\nDistrict officials to forecast that CLM’s presence at the\\nschool would cause a substantial disruption.\\n\\n Moreover, CLM’s hit list also invaded the “rights of other\\nstudents to be secure and to be let alone.” Tinker, 393 U.S. at\\n508. As Wynar stated, a student “target[ing] specific students\\nby name” for a potential school shooting “represent[s] the\\nquintessential harm to the rights of other students to be\\nsecure.” 728 F.3d at 1072.\\n\\n On appeal, CLM does not challenge the length of his\\nexpulsion—he argues only that the School District did not\\nhave the authority to expel him at all. We thus are not called\\nupon to determine whether a one-year expulsion was\\nexcessive. At a certain point, discipline may lose its basis in\\nreasonable, ongoing concerns of campus safety, disruption, or\\ninterference with the rights of other students, and instead\\nbecome primarily a punitive, retrospective response to the\\nstudent’s speech. Such discipline would be in conflict with\\nTinker. See LaVine, 257 F.3d at 991–92. Here, however, we\\ndecide only that CLM has not shown that the decision to\\nexpel him violates his constitutional rights under Tinker. The\\nSchool District could reasonably conclude that CLM’s\\ncontinued presence at the school at the time of his expulsion\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 19\\n\\nwould have caused a “substantial disruption of, or material\\ninterference with school activities” and would have interfered\\n“with the rights of other students to be secure and to be let\\nalone.” Wynar, 728 F.3d at 1070 (quoting Tinker, 393 U.S.\\nat 508, 514).\\n\\n B. The McNeils’ Substantive Due Process Claim\\n\\n Finally, we address the McNeils’ substantive due process\\nclaim. The Fourteenth Amendment’s Substantive Due\\nProcess Clause “provides heightened protection against\\ngovernment interference with certain fundamental rights and\\nliberty interests,” such as a parent’s right to make decisions\\nregarding the “care, custody, and control of their children.”\\nTroxel v. Granville, 530 U.S. 57, 65 (2000) (quoting\\nWashington v. Glucksberg, 521 U.S. 702, 720 (1997)). One\\nsuch fundamental right is “the right of parents to be free from\\nstate interference with their choice of the educational forum\\nitself, a choice that ordinarily determines the type of\\neducation one’s child will receive.” Fields v. Palmdale Sch.\\nDist., 427 F.3d 1197, 1207 (9th Cir. 2005). However, once\\nparents determine their child’s educational forum, their\\nfundamental right to control the education is “substantially\\ndiminished.” Id. at 1206. Due process does not give parents\\nthe right to interfere with a public school’s operations\\nbecause issues such as school discipline, the content of\\nexaminations, and dress code are “issues of public education\\ngenerally ‘committed to the control of state and local\\nauthorities.’” Id. (quoting Blau v. Fort Thomas Pub. Sch.\\nDist., 401 F.3d 381, 395–96 (6th Cir. 2005)).\\n\\n Here, the School District did not deprive the McNeils of\\ntheir liberty interest in determining where CLM completes his\\nhigh school education or whether CLM attends public school.\\n\\f20 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nThe McNeils’ fundamental right allowed them to voluntarily\\nenroll CLM at Sherwood High. In doing so, they accepted\\nSherwood High’s curriculum, school policies, and reasonable\\ndisciplinary measures.\\n\\n IV.\\n\\n The number of reported tragic school shootings over the\\npast two decades emphasizes the need for school districts to\\nhave the authority to take disciplinary action when faced with\\na credible threat of school violence. See Wynar, 728 F.3d at\\n1069. Although a student’s expectation as to the circulation\\nof his speech may be relevant to an evaluation of whether the\\nspeech constitutes a credible threat, the student’s intent as to\\ncirculation does not condition the school’s regulation of\\nthreatening off-campus speech. As we noted in Wynar, “[w]e\\ncan only imagine what would have happened if the school\\nofficials, after learning of [the] writing, did nothing about it,”\\nand CLM did in fact come to Sherwood High with a firearm\\nand the intent to carry out his hit list. 728 F.3d at 1070\\n(alterations in original) (quoting Boim v. Fulton Cty. Sch.\\nDist., 494 F.3d 978, 984 (11th Cir. 2007)).\\n\\n We hold that the School District’s decision to discipline\\nCLM for his off-campus speech did not violate CLM’s\\nconstitutional right to free speech. A student’s lack of intent\\nto convey his off-campus speech to any third party is relevant\\nto an evaluation of whether the speech constitutes a credible\\nthreat, but is not dispositive. Rather, a court must consider all\\nthe relevant factual circumstances when determining whether\\na school’s regulation of a student’s off-campus speech is\\nconstitutional. Here, CLM’s lack of intent to communicate\\nthe contents of his hit list is offset by the fact that his speech\\ncreated a credible, identifiable threat of school violence.\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 21\\n\\n The McNeils’ substantive due process claim also fails\\nbecause their fundamental right to choose CLM’s educational\\nforum was not infringed by the School District’s discipline of\\nCLM.\\n\\n The district court’s judgment for Defendants is\\nAFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4380671/\", \"author_raw\": \"PER CURIAM\"}]}"},"judge1":{"kind":"string","value":"RAYMOND C FISHER"},"judge2":{"kind":"string","value":"CONSUELO M CALLAHAN"},"judge3":{"kind":"string","value":"CATHY ANN BENCIVENGO"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4603418/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Michael McNeil v. Sherwood School District 88j\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-14\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nMICHAEL MCNEIL, guardian ad litem No. 17-35500\nas to CLM; CLM, by and through\nMichael McNeil guardian ad litem; D.C. No.\nJULIE MCNEIL, individually, 3:15-cv-01098-\n Plaintiffs-Appellants, SB\n v.\n OPINION\nSHERWOOD SCHOOL DISTRICT 88J;\nHEATHER H. CORDIE; KEN BELL;\nBRIAN BAILEY; PETER MILLER; GARY\nBENNETT,\n Defendants-Appellees.\n\n Appeal from the United States District Court\n for the District of Oregon\n Michael H. Simon, District Judge, Presiding\n\n Argued and Submitted October 9, 2018\n Portland, Oregon\n\n Filed March 14, 2019\n\n Before: Raymond C. Fisher and Consuelo M. Callahan,\nCircuit Judges, and Cathy Ann Bencivengo,* District Judge.\n\n Per Curiam Opinion\n\n\n *\n The Honorable Cathy Ann Bencivengo, United States District Judge\nfor the Southern District of California, sitting by designation.\n\f2 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\n\n SUMMARY**\n\n\n Civil Rights\n\n The panel affirmed the district court’s summary judgment\nin favor of a school district in an action brought pursuant to\n42 U.S.C. § 1983 by student CLM and his parents alleging\nthat the district violated plaintiffs’ First Amendment and\nFourteenth Amendment substantive due process rights when\nit expelled CLM for one year.\n\n CLM, then a high school sophomore at Sherwood High\nSchool, created in his personal journal a hit list of students\nthat “must die.” When his mother discovered the hit list, she\ntold a therapist, who informed the police, who told the school\ndistrict.\n\n The panel held that under the particular facts in this case,\nincluding the nature of the hit list, CLM’s access to firearms,\nand the close proximity of CLM’s home to the high school,\nthe decision to discipline CLM for his off-campus speech did\nnot violate his constitutional right to free speech. The panel\nheld that when considering whether a school district may\nconstitutionally regulate off-campus speech, courts must\ndetermine, based on the totality of the circumstances, whether\nthe speech bears a sufficient nexus to the school. The panel\nstated that there is always a sufficient nexus between the\nspeech and the school when the school district reasonably\nconcludes that it faces a credible, identifiable threat of school\nviolence. The panel further held that a student’s lack of intent\n\n **\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 3\n\nto convey his off-campus speech to any third party is relevant\nto an evaluation of whether the speech constitutes a credible\nthreat, but is not dispositive.\n\n The panel held that the claim brought by CLM’s parents\nalleging substantive due process violations failed because\ntheir fundamental right to choose CLM’s educational forum\nwas not infringed by the School District’s discipline of CLM.\n\n\n COUNSEL\n\nAdam S. Heder (argued) and Roger K. Harris, Harris Berne\nChristensen LLP, Lake Oswego, Oregon, for Plaintiffs-\nAppellants.\n\nBlake H. Fry (argued) and Peter R. Mersereau, Mersereau\nShannon LLP, Portland, Oregon, for Defendants-Appellees.\n\n\n OPINION\n\nPER CURIAM:\n\n CLM, then a high school sophomore at Sherwood High\nSchool (“Sherwood High”), created in his personal journal a\nhit list of students that “must die.” When his mother\ndiscovered the hit list, she told a therapist, who informed the\npolice, who told Sherwood School District (“School\nDistrict”). The School District expelled CLM for one year.\nCLM sued the School District primarily on First Amendment\ngrounds, and the district court held that the expulsion\nwas constitutional. On appeal, CLM and his parents\n(“Appellants”) assert that the School District cannot\n\f4 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\n\nconstitutionally regulate student speech that the student\nnever intended to communicate to any third party. We\nconclude that under the particular facts in this case, including\nthe nature of the hit list, CLM’s access to firearms, and the\nclose proximity of CLM’s home to Sherwood High, the\nSchool District did not violate any of Appellants’ asserted\nconstitutional rights.\n\n I.\n\n On or about May 25, 2014, CLM created a “hit list” in his\npersonal journal, naming 22 Sherwood High students and one\nformer employee, and stating “I am God” and “All These\nPeople Must Die.” On September 8, 2014, soon after CLM\nstarted his junior year at Sherwood High, CLM’s mother,\nMrs. McNeil, discovered the journal on CLM’s nightstand\nwhile cleaning CLM’s room. Upon reviewing the journal’s\ncontents, Mrs. McNeil came across the hit list and additional\nentries containing graphic depictions of violence.\nMrs. McNeil made copies of some of the journal entries,\nincluding the hit list, and sought guidance from a therapist the\nnext day regarding her findings. The therapist, alarmed by\nthe entries and believing they triggered her duties as a\nmandatory reporter, informed the Sherwood Police\nDepartment (“Sherwood Police”) about the list. The therapist\ntold Mrs. McNeil to call a local crisis hotline, which she did.\nThe hotline’s social worker also contacted the Sherwood\nPolice about CLM’s list.\n\n Later that day, the Sherwood Police searched the\nMcNeils’ residence. Officers found and confiscated several\nweapons, including a .22 caliber rifle and 525 rounds of\nammunition belonging to CLM, but the officers found\n\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 5\n\nnothing “to indicate any planning had gone into following\nthrough with the hit list.”\n\n Shortly after the search, CLM and his parents voluntarily\nwent to the police station, where they provided the police\nwith a copy of CLM’s hit list. After being read his Miranda\nwarnings, CLM admitted he created the hit list and that\n“sometimes he thinks killing people might relieve some of the\nstress he feels,” but he also stated “he uses the journal to\nvent” and that “he would never carry out” such thoughts.\nCLM claimed he created the list about four months prior to its\ndiscovery. The Sherwood Police determined that no criminal\ncharges would be brought against CLM at that time.\n\n Meanwhile, the Sherwood Police informed the School\nDistrict of CLM’s hit list, of the fact the police had seized\nguns from his house, and that CLM’s journal contained\nadditional entries that graphically depicted school violence.\nSherwood High’s principal, Ken Bell, was aware that the\nMcNeils lived “very close” to Sherwood High and assembled\nan administrative team to address the issue.\n\n Consistent with the notification requirements of Oregon\nRevised Statutes § 339.327(1), the School District’s policies\nrequired school faculty to notify the parents of students found\non a hit list within 12 hours of discovery. Within the 12-hour\nspan, Bell’s team made the necessary calls, which did not\nidentify CLM as the hit list’s author. Before the final call\nwas made, the media began contacting the School District to\ninquire about the list. The School District also learned that\nCLM’s picture had been posted on social media accounts.\nUpon realization of the widespread knowledge regarding\nCLM’s hit list, Bell’s team sent out a recorded voice message\nto all parents of students in the School District to notify them\n\f6 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\n\nof the issue. The message stated that a student of Sherwood\nHigh had authored a hit list containing 23 names, but that the\nlist contained no specific threats and the home in which the\nstudent resided was safe. The School District also issued a\npress release containing similar information. Thereafter,\nSherwood High received numerous calls and emails from\nparents, media outlets, and the public requesting information\nabout the hit list, CLM’s identity, and whether CLM posed a\nthreat to others. Some parents demanded to meet with Bell,\nand other parents had their children leave school early, miss\nseveral days, or transfer out of the district. A student was\ncaught on campus with a knife, and he alleged “he needed it\nto protect himself in light of the hit list.”\n\n The School District suspended CLM pending an\nexpulsion hearing. Bell recommended that CLM be expelled\nfrom Sherwood High for one year because news of his hit list\n“significantly disrupted the learning environment at school,”\nwhich would only be increased by CLM’s return. Sherwood\nHigh’s associate principal wrote a letter to the McNeils\nstating that he recommended that CLM be expelled for one\nyear for making “a threat of violence” that “has caused a\ndistinct and substantial disruption to the school environment.”\n\n On September 22, 2014, an expulsion hearing was held\nbefore Pete Miller, a hearings officer for the School District.\nCLM was represented by an attorney at the hearing. Upon\nreview of the testimony and evidence, Miller adopted the\nprincipal’s recommendation for expulsion, largely based on\n“the significant disruption” CLM’s list caused in the school\nenvironment. After the hearing, the School District’s\nassociate superintendent sent a decision letter to the McNeils,\nexplaining that the list constituted “a threat of violence in\nviolation of the Student Code of Conduct.” The letter\n\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 7\n\nexplained that CLM could not return to school that academic\nyear, and that the McNeils had the right to appeal the decision\nto the School District’s board. The McNeils did not appeal.\n\n Following his expulsion, CLM received alternative forms\nof education through the School District, including online\ncourses, in-person tutoring, and courses at a community\ncollege. However, CLM had “no immediate or even daily\naccess to instructors,” and the tutors were “inconsistent and\nnot dependable,” as they missed appointments and in one\ninstance failed to follow up for weeks. CLM ultimately had\nto retake several classes.\n\n In June 2015, CLM and the McNeils filed a complaint in\nthe U.S. District Court for the District of Oregon seeking\ndamages and other relief under 42 U.S.C. § 1983. CLM\nsought damages for the School District’s alleged violations of\nthe Free Speech Clause of the First Amendment and of the\nEqual Protection and Due Process Clauses of the Fourteenth\nAmendment.1 The McNeils also brought a substantive due\nprocess challenge to CLM’s expulsion. The complaint sought\nexpungement of any documentation or reference to CLM’s\nsuspension, expulsion, and the circumstances surrounding the\ndisciplinary proceedings.\n\n The School District moved for summary judgment on all\nclaims, while CLM and the McNeils moved for summary\njudgment on their free speech claim and substantive due\nprocess claim, respectively. The district court concluded that\nthe School District could regulate CLM’s off-campus speech\nfor three primary reasons: (1) the hit list had a sufficient\n\n 1\n On appeal, CLM has withdrawn his equal protection and\nprocedural due process claims.\n\f8 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\n\nconnection to Sherwood High; (2) school officials could have\nreasonably foreseen that the effects of the hit list would spill\nover into the school environment; and (3) the facts in CLM’s\ncase mirrored those in Wynar v. Douglas County School\nDistrict, 728 F.3d 1062 (9th Cir. 2013). The district court\nfurther held that the one-year expulsion did not violate\nCLM’s First Amendment rights, reasoning that a student-\nauthored hit list would cause a substantial disruption in any\nschool community, particularly in light of Oregon’s statutory\nnotification requirement.\n\n The district court also granted the School District\nsummary judgment on the McNeils’ substantive due process\nclaim. The district court reasoned that the claim failed\nbecause it was derivative of CLM’s First Amendment claim\nand because a school has the authority to discipline students\non reasonable grounds. The district court determined that\nalthough the McNeils had a liberty interest in an “initial\n‘educational venue choice,’” their interest did not extend to\ntemporary school discipline within that venue, which is an\nissue “generally committed to the control of the state and\nlocal authorities.”\n\n Appellants timely appealed. On appeal, CLM claims that\nthe School District lacked authority under the First\nAmendment to discipline CLM for his hit list. The McNeils\nalso allege that the School District’s expulsion of CLM\nviolated their substantive due process right “to be free from\nstate interference with their choice of [CLM’s] educational\nforum.”\n\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 9\n\n II.\n\n We review “the district court’s grant of summary\njudgment de novo.” Szajer v. City of Los Angeles, 632 F.3d\n607, 610 (9th Cir. 2011). The court views “evidence in the\nlight most favorable to the nonmoving party,” to determine\n“whether genuine issues of material fact exist.” George v.\nEdholm, 752 F.3d 1206, 1214 (9th Cir. 2014) (quoting\nBagdadi v. Nazar, 84 F.3d 1194, 1197 (9th Cir. 1996)). If a\nrational juror “‘could resolve a genuine issue of material fact\nin the nonmoving party’s favor,’ summary judgment is\ninappropriate.” Id. (quoting Bravo v. City of Santa Maria,\n665 F.3d 1076, 1083 (9th Cir. 2011)).\n\n III.\n\n “The first inquiry in any § 1983 suit . . . is whether the\nplaintiff has been deprived of a right ‘secured by the\nConstitution and laws.’” Baker v. McCollan, 443 U.S. 137,\n140 (1979). Section 1983 does not create substantive rights,\nbut rather provides the procedural mechanism for vindicating\nfederal statutory or constitutional rights. Id. at 145 n.3.\n\n A. CLM’s First Amendment Claim\n\n The Supreme Court has held that public school students\ndo not forfeit “their constitutional rights to freedom of speech\nor expression at the schoolhouse gate.” Tinker v. Des Moines\nIndep. Cnty. Sch. Dist., 393 U.S. 503, 506 (1969). However,\ntheir rights are “not automatically coextensive with the rights\nof adults in other settings.” Wynar, 728 F.3d at 1067 (quoting\nBethel Sch. Dist. No. 403 v. Fraser, 478 U.S. 675, 682\n(1986)). Additionally, although they enjoy greater freedom\nto speak when they are off campus than when they are on\n\f10 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\n\ncampus, their off-campus speech is not necessarily beyond\nthe reach of a school district’s regulatory authority. Our\nreview of the School District’s treatment of CLM turns on\ntwo inquiries: (1) whether the School District could\npermissibly regulate CLM’s off-campus speech at all; and if\nso, (2) whether the School District’s decision to expel CLM\nviolated the First Amendment standard for school regulation\nof speech set out in Tinker. See C.R. v. Eugene Sch. Dist. 4J,\n835 F.3d 1142, 1148 (9th Cir. 2016).\n\n 1. The School District’s authority to regulate CLM’s\n off-campus speech\n\n We have twice considered the authority of a school\ndistrict to regulate off-campus speech. In Wynar, a high\nschool student was expelled because he sent graphic instant\nmessages to his friends from his home computer. 728 F.3d\nat 1064–67. The messages expressed threats of school\nviolence and indicated that the student had access to guns and\nammunition. Id. at 1065–66. The messages were brought to\nthe school’s attention by the student’s friends. Id. at 1066.\nDespite the student’s lack of intent in bringing his speech to\nthe school campus, we held that his expulsion was\nconstitutional, reasoning that “when faced with an identifiable\nthreat of school violence, schools may take disciplinary\naction in response to off-campus speech . . . .” Id. at 1069.\n\n In C.R., a school district suspended a student for sexually\nharassing two students shortly after school hours in a public\npark near the school. 835 F.3d at 1145–47. The student\nargued that the school’s regulation of his harassment was\nunconstitutional because the harassment occurred in a public,\noff-campus park. Id. at 1145–46. While we recognized that\nthe harassment constituted off-campus speech, we held that\n\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 11\n\nthe suspension was constitutional. Id. at 1146. We reasoned\nthat the speech was closely tied to the school, noting that all\nof the individuals involved were students and that the incident\noccurred in close proximity to the school and just minutes\nafter the end of the school day. Id. at 1150–51. We also\nnoted that school administrators could reasonably expect the\neffects of the speech to disrupt the school environment. Id.\nat 1151.\n\n Although our previous cases have not settled on a test for\ndetermining when a school can constitutionally regulate off-\ncampus speech, we have stressed the importance of flexibility\nin dealing with the “myriad of circumstances” schools face.\nWynar, 728 F.3d at 1069; see also C.R., 835 F.3d at 1150–51.\nThese cases have noted two approaches: (1) requiring that\noff-campus speech “have a sufficient ‘nexus’ to the school,”\nWynar, 728 F.3d at 1068 (quoting Kowalski v. Berkeley Cty.\nSchs., 652 F.3d 565, 573 (4th Cir. 2011)); and (2) requiring\n“that it be ‘reasonably foreseeable that the speech will reach\nthe school community,’” id. (quoting S.J.W. v. Lee’s Summit\nR-7 Sch. Dist., 696 F.3d 771, 777 (8th Cir. 2012)).\n\n We now clarify that courts considering whether a school\ndistrict may constitutionally regulate off-campus speech must\ndetermine, based on the totality of the circumstances, whether\nthe speech bears a sufficient nexus to the school. See C.R.,\n835 F.3d at 1150–51; Wynar, 728 F.3d at 1069. This test is\nflexible and fact-specific, but the relevant considerations will\ninclude (1) the degree and likelihood of harm to the school\ncaused or augured by the speech, see Wynar, 728 F.3d at\n1069, (2) whether it was reasonably foreseeable that the\nspeech would reach and impact the school, see C.R., 835 F.3d\nat 1150–51; Wynar, 728 F.3d at 1069; Kowalski, 652 F.3d at\n573, and (3) the relation between the content and context of\n\f12 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\n\nthe speech and the school, see C.R., 835 F.3d at 1150–51;\nWynar, 728 F.3d at 1069.2 There is always a sufficient nexus\nbetween the speech and the school when the school district\nreasonably concludes that it faces a credible, identifiable\nthreat of school violence. See Wynar, 728 F.3d at 1069.\n\n CLM attempts to distinguish Wynar on the ground that he\nhad no intent to communicate the contents of his speech to\nanyone. That distinction cannot be dispositive. We have\nrecognized repeatedly that the specter of school violence\nplaces a weighty social responsibility on school districts to\nensure that “warning signs” do not turn to tragedy. See\nLaVine v. Blaine Sch. Dist., 257 F.3d 981, 987 (9th Cir.\n2001). This responsibility does not mean schools may “expel\nstudents just because they are ‘loners,’ wear black and play\nvideo games.” Id. It does mean, however, that a student’s\nintent, although relevant, does not necessarily define the\nthreat of violence. We reaffirm our holding in Wynar that\nregardless of the speaker’s intent or how speech comes to a\nschool district’s attention, a school district may take\ndisciplinary action in response to off-campus speech when it\nreasonably determines that it faces an identifiable and\ncredible threat of school violence. See Wynar, 728 F.3d at\n1069.\n\n Of course, any information school officials have\nregarding the intention of the speaker may bear on whether a\nthreat is credible. The intent to keep a threat private,\n\n\n 2\n In analyzing foreseeability, we have considered both whether it\nwas foreseeable to the speaker that the speech would reach the school, see\nWynar, 728 F.3d at 1068, and whether it was foreseeable to school\nadministrators that the speech would impact the school community, see\nC.R., 835 F.3d at 1151.\n\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 13\n\nhowever, is not determinative—it would be absurd to suggest\nthat secret threats or planning cannot give rise to reasonable\nconcerns of school violence. See, e.g., Kristin McCauley,\nArapahoe Cty. Sheriff, Investigative Report: Arapahoe\nHigh School Case # CT13-44545, at 29–31 (2014),\nhttps://mediaassets.thedenverchannel.com/document/2014/\n10/10/AHSShootingReport_8903184_ver1.0.pdf?_ga=2.15\n5919036.684175427.1543265365-980564543.1543265365\n(describing password-protected diary entries containing\nplanning for a fatal 2013 school shooting).\n\n Here, the School District reasonably determined CLM\npresented a credible threat. The School District knew CLM\nhad identified specific targets, had accentuated his hit list\nwith the phrases “I am God” and “All These People Must\nDie,” lived in a gun-owning home close to the school, and\nhad had thoughts of suicide. The School District also knew\nthe diary contained other graphic depictions of school\nviolence. This evidence was sufficient to render the School\nDistrict’s determination reasonable and to give it authority to\nregulate CLM’s speech.\n\n CLM argues that, once he was in police custody, he\nceased to pose a safety threat because he was being evaluated\nin the care of competent authorities. This evaluation,\nhowever, lasted only one day, and afterward CLM returned\nhome with limited oversight and his parents retrieved the\nfamily’s firearms. The police’s decision to release CLM\ncannot prevent the school from addressing the threat of\nviolence: a school may take action even where police or\nmental health professionals have elected not to do so. See\nLaVine, 257 F.3d at 990 (“It is true that the [police and a\npsychiatrist] did not believe [the student at issue] should be\ninvoluntarily committed, but the standard for involuntary\n\f14 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\n\ncommitment is not the same as that for school officials to take\naction.”). Schools must be permitted to act preventatively,\ntaking into account other students’ interest in a safe school\nenvironment. Here, the school reasonably determined the risk\nwas sufficient to take action.\n\n Our test does not allow a school to take disciplinary\naction in response to just any perceived threat of school\nviolence arising from off-campus speech. Although our\nreview of a school’s determinations in this regard is\ndeferential, “deference does not mean abdication; there are\nsituations where school officials overstep their bounds and\nviolate the Constitution.” Id. at 988.\n\n In Porter v. Ascension Parish School Board, 393 F.3d 608\n(5th Cir. 2004), for example, Adam, a high school student,\ndepicted a military attack on his school in his sketchpad. Id.\nat 611. There were no students identified in the drawing and\nAdam stored the sketchpad in his closet at home. Id. Two\nyears later, Adam’s brother brought the sketchpad to the\nschool. Id. Another student, with permission from Adam’s\nbrother, flipped through the sketchpad and, upon coming\nacross Adam’s drawing of the military attack, exclaimed,\n“they’re going to blow up [the school].” Id. This brought the\ndrawing to the attention of school authorities, who expelled\nAdam for the drawing. Id. at 611–12. The Fifth Circuit held\nthat Adam’s drawing was protected by the First Amendment\nbecause it did not constitute “speech on campus or even\nspeech directed at the campus.” Id. at 615. The Fifth Circuit\nnoted that for Adam’s drawing to lose First Amendment\nprotection, something more than “accidental and\nunintentional exposure to public scrutiny must take place.”\nId. at 618. The Fifth Circuit reasoned that because Adam’s\n\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 15\n\ndrawing was not intentionally communicated, it was not a\ntrue threat and therefore not subject to school regulation. Id.\n\n We do not adopt the Fifth Circuit’s reasoning, but note\nthat our approach supports the result in Porter and that there\nare critical distinctions between the facts here and those in\nPorter. There, the drawing was two years old, highly\nfantastical, unspecific, and unaccompanied by other indicia\nof a violent intent. See id. at 611. There was no evidence\nAdam had ever had the capacity to carry out the actions\ndepicted in the drawing. In contrast to this case, it was\nunlikely that any reasonable administrator could have\nbelieved the school faced a credible safety threat.\n\n Taken as a whole, the three considerations that guide\napplication of the nexus test support the School District here.\nFirst, it was reasonable for School District officials to\nconclude that CLM presented a credible threat of severe harm\nto the school community. This consideration establishes a\nsufficient nexus between the speech and the school to permit\nregulation. See Wynar, 728 F.3d at 1069.\n\n Second, once it learned of the hit list, the School District\ncould reasonably foresee that news of the threat would reach\nand impact the school and disrupt the school environment.\nOregon’s notification requirement, Oregon Revised Statutes\n§ 339.327(1), provides that “[a] superintendent of a school\ndistrict . . . who has reasonable cause to believe that a person,\nwhile in a school, is or has been in possession of a list that\nthreatens harm to other persons” must notify the “parent or\nguardian of any student whose name appears on the list as a\ntarget of the harm.” The statute applies to enrolled students\nwho are or have been in possession of a threatening list,\nregardless of whether the list is on or off school property. It\n\f16 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\n\nwas reasonably foreseeable that parents, once notified as\nrequired by Oregon law, would inform their children about\nthe hit list. It was also reasonably foreseeable that\ninformation about the hit list would cause students to be\nfearful for their safety. Invariably, these resulting\nconsequences would create classroom distractions and\nsubstantial disruptions within the Sherwood High community.\nAlthough it was not foreseeable to CLM that his speech\nwould reach the school, a lack of intent to share speech is of\nminimal weight when, as here, the speech contains a credible\nthreat of violence directed at the school. Someone planning\na violent act does not need to advertise it.\n\n Third, the content of the speech involved the school. As\nwe noted in Wynar and C.R., when all involved parties are\nstudents, that fact “typically counsels in favor of finding that\na student’s speech was susceptible to discipline.” C.R., 835\nF.3d at 1150 (citing Wynar, 728 F.3d at 1069). Here, CLM’s\nhit list contained the names of 22 students and one former\nteacher of the School District, and thus, presented a particular\nthreat to the school community. Ordinarily, schools may not\ndiscipline students for the contents of their private, off-\ncampus diary entries, any more than they can punish students\nfor their private thoughts, but schools have a right, indeed an\nobligation, to address a credible threat of violence involving\nthe school community.\n\n In sum, we conclude that the School District could\nregulate CLM’s off-campus speech without violating his First\nAmendment rights. Although CLM may not have foreseen\nhis speech reaching Sherwood High, the School District,\nwhen informed of CLM’s hit list, reasonably determined that\nit faced a credible, identifiable threat of school violence. The\nspeech bore a sufficient nexus to the school. Accordingly, the\n\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 17\n\nSchool District could take disciplinary action consistent with\nTinker.\n\n 2. The School District’s expulsion of CLM\n\n Although some disciplinary action was permissible, there\nremains the issue of whether the School District’s expulsion\nwas consistent with the First Amendment. “Under Tinker,\nschools may restrict speech that ‘might reasonably [lead]\nschool authorities to forecast substantial disruption of or\nmaterial interference with school activities’ or that collides\n‘with the rights of other students to be secure and to be let\nalone.’” Wynar, 728 F.3d at 1070 (alteration in original)\n(quoting Tinker, 393 U.S. at 508, 514). This analysis\nconsiders not just the actions of the disciplined student, but\nrather “all of the circumstances confronting the school\nofficials that might reasonably portend disruption.” LaVine,\n257 F.3d at 989. We have held that “the specter of a school\nshooting qualifies under either prong of Tinker.” Wynar, 728\nF.3d at 1070. School districts should be afforded deference\nin determining what discipline is necessary to maintain safety\nand order, see id. at 1072, but we have cautioned that\ndiscipline can extend beyond the scope Tinker allows when\nit takes on a purely punitive character, see LaVine, 257 F.3d\nat 992.\n\n Here, taking disciplinary action was reasonable under\nboth Tinker prongs. As discussed, CLM’s hit list identifying\nspecific individuals constituted a credible threat of school\nviolence that required the School District to act in furtherance\nof the safety of its students. Oregon’s statutory notification\nrequirement further confirms this proposition, as the School\nDistrict could reasonably foresee that informing parents that\ntheir children were identified on a student-authored hit list\n\f18 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\n\nwould cause a substantial disruption within the Sherwood\nHigh community. Moreover, CLM’s hit list led to a material\ndisruption at Sherwood High by causing safety concerns\namong both students and parents. The record evinces this\ndisruption by the outcry from parents who demanded to speak\nwith Sherwood High authorities and by the parents who\ntransferred or threatened to transfer their child out of the\nSchool District. Thus, in light of the nature of CLM’s hit list\nand the widespread knowledge regarding the list in the\nSherwood High community, it was reasonable for School\nDistrict officials to forecast that CLM’s presence at the\nschool would cause a substantial disruption.\n\n Moreover, CLM’s hit list also invaded the “rights of other\nstudents to be secure and to be let alone.” Tinker, 393 U.S. at\n508. As Wynar stated, a student “target[ing] specific students\nby name” for a potential school shooting “represent[s] the\nquintessential harm to the rights of other students to be\nsecure.” 728 F.3d at 1072.\n\n On appeal, CLM does not challenge the length of his\nexpulsion—he argues only that the School District did not\nhave the authority to expel him at all. We thus are not called\nupon to determine whether a one-year expulsion was\nexcessive. At a certain point, discipline may lose its basis in\nreasonable, ongoing concerns of campus safety, disruption, or\ninterference with the rights of other students, and instead\nbecome primarily a punitive, retrospective response to the\nstudent’s speech. Such discipline would be in conflict with\nTinker. See LaVine, 257 F.3d at 991–92. Here, however, we\ndecide only that CLM has not shown that the decision to\nexpel him violates his constitutional rights under Tinker. The\nSchool District could reasonably conclude that CLM’s\ncontinued presence at the school at the time of his expulsion\n\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 19\n\nwould have caused a “substantial disruption of, or material\ninterference with school activities” and would have interfered\n“with the rights of other students to be secure and to be let\nalone.” Wynar, 728 F.3d at 1070 (quoting Tinker, 393 U.S.\nat 508, 514).\n\n B. The McNeils’ Substantive Due Process Claim\n\n Finally, we address the McNeils’ substantive due process\nclaim. The Fourteenth Amendment’s Substantive Due\nProcess Clause “provides heightened protection against\ngovernment interference with certain fundamental rights and\nliberty interests,” such as a parent’s right to make decisions\nregarding the “care, custody, and control of their children.”\nTroxel v. Granville, 530 U.S. 57, 65 (2000) (quoting\nWashington v. Glucksberg, 521 U.S. 702, 720 (1997)). One\nsuch fundamental right is “the right of parents to be free from\nstate interference with their choice of the educational forum\nitself, a choice that ordinarily determines the type of\neducation one’s child will receive.” Fields v. Palmdale Sch.\nDist., 427 F.3d 1197, 1207 (9th Cir. 2005). However, once\nparents determine their child’s educational forum, their\nfundamental right to control the education is “substantially\ndiminished.” Id. at 1206. Due process does not give parents\nthe right to interfere with a public school’s operations\nbecause issues such as school discipline, the content of\nexaminations, and dress code are “issues of public education\ngenerally ‘committed to the control of state and local\nauthorities.’” Id. (quoting Blau v. Fort Thomas Pub. Sch.\nDist., 401 F.3d 381, 395–96 (6th Cir. 2005)).\n\n Here, the School District did not deprive the McNeils of\ntheir liberty interest in determining where CLM completes his\nhigh school education or whether CLM attends public school.\n\f20 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\n\nThe McNeils’ fundamental right allowed them to voluntarily\nenroll CLM at Sherwood High. In doing so, they accepted\nSherwood High’s curriculum, school policies, and reasonable\ndisciplinary measures.\n\n IV.\n\n The number of reported tragic school shootings over the\npast two decades emphasizes the need for school districts to\nhave the authority to take disciplinary action when faced with\na credible threat of school violence. See Wynar, 728 F.3d at\n1069. Although a student’s expectation as to the circulation\nof his speech may be relevant to an evaluation of whether the\nspeech constitutes a credible threat, the student’s intent as to\ncirculation does not condition the school’s regulation of\nthreatening off-campus speech. As we noted in Wynar, “[w]e\ncan only imagine what would have happened if the school\nofficials, after learning of [the] writing, did nothing about it,”\nand CLM did in fact come to Sherwood High with a firearm\nand the intent to carry out his hit list. 728 F.3d at 1070\n(alterations in original) (quoting Boim v. Fulton Cty. Sch.\nDist., 494 F.3d 978, 984 (11th Cir. 2007)).\n\n We hold that the School District’s decision to discipline\nCLM for his off-campus speech did not violate CLM’s\nconstitutional right to free speech. A student’s lack of intent\nto convey his off-campus speech to any third party is relevant\nto an evaluation of whether the speech constitutes a credible\nthreat, but is not dispositive. Rather, a court must consider all\nthe relevant factual circumstances when determining whether\na school’s regulation of a student’s off-campus speech is\nconstitutional. Here, CLM’s lack of intent to communicate\nthe contents of his hit list is offset by the fact that his speech\ncreated a credible, identifiable threat of school violence.\n\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 21\n\n The McNeils’ substantive due process claim also fails\nbecause their fundamental right to choose CLM’s educational\nforum was not infringed by the School District’s discipline of\nCLM.\n\n The district court’s judgment for Defendants is\nAFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Michael McNeil v. Sherwood School District 88j\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-14\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nMICHAEL MCNEIL, guardian ad litem No. 17-35500\\nas to CLM; CLM, by and through\\nMichael McNeil guardian ad litem; D.C. No.\\nJULIE MCNEIL, individually, 3:15-cv-01098-\\n Plaintiffs-Appellants, SB\\n v.\\n OPINION\\nSHERWOOD SCHOOL DISTRICT 88J;\\nHEATHER H. CORDIE; KEN BELL;\\nBRIAN BAILEY; PETER MILLER; GARY\\nBENNETT,\\n Defendants-Appellees.\\n\\n Appeal from the United States District Court\\n for the District of Oregon\\n Michael H. Simon, District Judge, Presiding\\n\\n Argued and Submitted October 9, 2018\\n Portland, Oregon\\n\\n Filed March 14, 2019\\n\\n Before: Raymond C. Fisher and Consuelo M. Callahan,\\nCircuit Judges, and Cathy Ann Bencivengo,* District Judge.\\n\\n Per Curiam Opinion\\n\\n\\n *\\n The Honorable Cathy Ann Bencivengo, United States District Judge\\nfor the Southern District of California, sitting by designation.\\n\\f2 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\n SUMMARY**\\n\\n\\n Civil Rights\\n\\n The panel affirmed the district court’s summary judgment\\nin favor of a school district in an action brought pursuant to\\n42 U.S.C. § 1983 by student CLM and his parents alleging\\nthat the district violated plaintiffs’ First Amendment and\\nFourteenth Amendment substantive due process rights when\\nit expelled CLM for one year.\\n\\n CLM, then a high school sophomore at Sherwood High\\nSchool, created in his personal journal a hit list of students\\nthat “must die.” When his mother discovered the hit list, she\\ntold a therapist, who informed the police, who told the school\\ndistrict.\\n\\n The panel held that under the particular facts in this case,\\nincluding the nature of the hit list, CLM’s access to firearms,\\nand the close proximity of CLM’s home to the high school,\\nthe decision to discipline CLM for his off-campus speech did\\nnot violate his constitutional right to free speech. The panel\\nheld that when considering whether a school district may\\nconstitutionally regulate off-campus speech, courts must\\ndetermine, based on the totality of the circumstances, whether\\nthe speech bears a sufficient nexus to the school. The panel\\nstated that there is always a sufficient nexus between the\\nspeech and the school when the school district reasonably\\nconcludes that it faces a credible, identifiable threat of school\\nviolence. The panel further held that a student’s lack of intent\\n\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 3\\n\\nto convey his off-campus speech to any third party is relevant\\nto an evaluation of whether the speech constitutes a credible\\nthreat, but is not dispositive.\\n\\n The panel held that the claim brought by CLM’s parents\\nalleging substantive due process violations failed because\\ntheir fundamental right to choose CLM’s educational forum\\nwas not infringed by the School District’s discipline of CLM.\\n\\n\\n COUNSEL\\n\\nAdam S. Heder (argued) and Roger K. Harris, Harris Berne\\nChristensen LLP, Lake Oswego, Oregon, for Plaintiffs-\\nAppellants.\\n\\nBlake H. Fry (argued) and Peter R. Mersereau, Mersereau\\nShannon LLP, Portland, Oregon, for Defendants-Appellees.\\n\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n CLM, then a high school sophomore at Sherwood High\\nSchool (“Sherwood High”), created in his personal journal a\\nhit list of students that “must die.” When his mother\\ndiscovered the hit list, she told a therapist, who informed the\\npolice, who told Sherwood School District (“School\\nDistrict”). The School District expelled CLM for one year.\\nCLM sued the School District primarily on First Amendment\\ngrounds, and the district court held that the expulsion\\nwas constitutional. On appeal, CLM and his parents\\n(“Appellants”) assert that the School District cannot\\n\\f4 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nconstitutionally regulate student speech that the student\\nnever intended to communicate to any third party. We\\nconclude that under the particular facts in this case, including\\nthe nature of the hit list, CLM’s access to firearms, and the\\nclose proximity of CLM’s home to Sherwood High, the\\nSchool District did not violate any of Appellants’ asserted\\nconstitutional rights.\\n\\n I.\\n\\n On or about May 25, 2014, CLM created a “hit list” in his\\npersonal journal, naming 22 Sherwood High students and one\\nformer employee, and stating “I am God” and “All These\\nPeople Must Die.” On September 8, 2014, soon after CLM\\nstarted his junior year at Sherwood High, CLM’s mother,\\nMrs. McNeil, discovered the journal on CLM’s nightstand\\nwhile cleaning CLM’s room. Upon reviewing the journal’s\\ncontents, Mrs. McNeil came across the hit list and additional\\nentries containing graphic depictions of violence.\\nMrs. McNeil made copies of some of the journal entries,\\nincluding the hit list, and sought guidance from a therapist the\\nnext day regarding her findings. The therapist, alarmed by\\nthe entries and believing they triggered her duties as a\\nmandatory reporter, informed the Sherwood Police\\nDepartment (“Sherwood Police”) about the list. The therapist\\ntold Mrs. McNeil to call a local crisis hotline, which she did.\\nThe hotline’s social worker also contacted the Sherwood\\nPolice about CLM’s list.\\n\\n Later that day, the Sherwood Police searched the\\nMcNeils’ residence. Officers found and confiscated several\\nweapons, including a .22 caliber rifle and 525 rounds of\\nammunition belonging to CLM, but the officers found\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 5\\n\\nnothing “to indicate any planning had gone into following\\nthrough with the hit list.”\\n\\n Shortly after the search, CLM and his parents voluntarily\\nwent to the police station, where they provided the police\\nwith a copy of CLM’s hit list. After being read his Miranda\\nwarnings, CLM admitted he created the hit list and that\\n“sometimes he thinks killing people might relieve some of the\\nstress he feels,” but he also stated “he uses the journal to\\nvent” and that “he would never carry out” such thoughts.\\nCLM claimed he created the list about four months prior to its\\ndiscovery. The Sherwood Police determined that no criminal\\ncharges would be brought against CLM at that time.\\n\\n Meanwhile, the Sherwood Police informed the School\\nDistrict of CLM’s hit list, of the fact the police had seized\\nguns from his house, and that CLM’s journal contained\\nadditional entries that graphically depicted school violence.\\nSherwood High’s principal, Ken Bell, was aware that the\\nMcNeils lived “very close” to Sherwood High and assembled\\nan administrative team to address the issue.\\n\\n Consistent with the notification requirements of Oregon\\nRevised Statutes § 339.327(1), the School District’s policies\\nrequired school faculty to notify the parents of students found\\non a hit list within 12 hours of discovery. Within the 12-hour\\nspan, Bell’s team made the necessary calls, which did not\\nidentify CLM as the hit list’s author. Before the final call\\nwas made, the media began contacting the School District to\\ninquire about the list. The School District also learned that\\nCLM’s picture had been posted on social media accounts.\\nUpon realization of the widespread knowledge regarding\\nCLM’s hit list, Bell’s team sent out a recorded voice message\\nto all parents of students in the School District to notify them\\n\\f6 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nof the issue. The message stated that a student of Sherwood\\nHigh had authored a hit list containing 23 names, but that the\\nlist contained no specific threats and the home in which the\\nstudent resided was safe. The School District also issued a\\npress release containing similar information. Thereafter,\\nSherwood High received numerous calls and emails from\\nparents, media outlets, and the public requesting information\\nabout the hit list, CLM’s identity, and whether CLM posed a\\nthreat to others. Some parents demanded to meet with Bell,\\nand other parents had their children leave school early, miss\\nseveral days, or transfer out of the district. A student was\\ncaught on campus with a knife, and he alleged “he needed it\\nto protect himself in light of the hit list.”\\n\\n The School District suspended CLM pending an\\nexpulsion hearing. Bell recommended that CLM be expelled\\nfrom Sherwood High for one year because news of his hit list\\n“significantly disrupted the learning environment at school,”\\nwhich would only be increased by CLM’s return. Sherwood\\nHigh’s associate principal wrote a letter to the McNeils\\nstating that he recommended that CLM be expelled for one\\nyear for making “a threat of violence” that “has caused a\\ndistinct and substantial disruption to the school environment.”\\n\\n On September 22, 2014, an expulsion hearing was held\\nbefore Pete Miller, a hearings officer for the School District.\\nCLM was represented by an attorney at the hearing. Upon\\nreview of the testimony and evidence, Miller adopted the\\nprincipal’s recommendation for expulsion, largely based on\\n“the significant disruption” CLM’s list caused in the school\\nenvironment. After the hearing, the School District’s\\nassociate superintendent sent a decision letter to the McNeils,\\nexplaining that the list constituted “a threat of violence in\\nviolation of the Student Code of Conduct.” The letter\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 7\\n\\nexplained that CLM could not return to school that academic\\nyear, and that the McNeils had the right to appeal the decision\\nto the School District’s board. The McNeils did not appeal.\\n\\n Following his expulsion, CLM received alternative forms\\nof education through the School District, including online\\ncourses, in-person tutoring, and courses at a community\\ncollege. However, CLM had “no immediate or even daily\\naccess to instructors,” and the tutors were “inconsistent and\\nnot dependable,” as they missed appointments and in one\\ninstance failed to follow up for weeks. CLM ultimately had\\nto retake several classes.\\n\\n In June 2015, CLM and the McNeils filed a complaint in\\nthe U.S. District Court for the District of Oregon seeking\\ndamages and other relief under 42 U.S.C. § 1983. CLM\\nsought damages for the School District’s alleged violations of\\nthe Free Speech Clause of the First Amendment and of the\\nEqual Protection and Due Process Clauses of the Fourteenth\\nAmendment.1 The McNeils also brought a substantive due\\nprocess challenge to CLM’s expulsion. The complaint sought\\nexpungement of any documentation or reference to CLM’s\\nsuspension, expulsion, and the circumstances surrounding the\\ndisciplinary proceedings.\\n\\n The School District moved for summary judgment on all\\nclaims, while CLM and the McNeils moved for summary\\njudgment on their free speech claim and substantive due\\nprocess claim, respectively. The district court concluded that\\nthe School District could regulate CLM’s off-campus speech\\nfor three primary reasons: (1) the hit list had a sufficient\\n\\n 1\\n On appeal, CLM has withdrawn his equal protection and\\nprocedural due process claims.\\n\\f8 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nconnection to Sherwood High; (2) school officials could have\\nreasonably foreseen that the effects of the hit list would spill\\nover into the school environment; and (3) the facts in CLM’s\\ncase mirrored those in Wynar v. Douglas County School\\nDistrict, 728 F.3d 1062 (9th Cir. 2013). The district court\\nfurther held that the one-year expulsion did not violate\\nCLM’s First Amendment rights, reasoning that a student-\\nauthored hit list would cause a substantial disruption in any\\nschool community, particularly in light of Oregon’s statutory\\nnotification requirement.\\n\\n The district court also granted the School District\\nsummary judgment on the McNeils’ substantive due process\\nclaim. The district court reasoned that the claim failed\\nbecause it was derivative of CLM’s First Amendment claim\\nand because a school has the authority to discipline students\\non reasonable grounds. The district court determined that\\nalthough the McNeils had a liberty interest in an “initial\\n‘educational venue choice,’” their interest did not extend to\\ntemporary school discipline within that venue, which is an\\nissue “generally committed to the control of the state and\\nlocal authorities.”\\n\\n Appellants timely appealed. On appeal, CLM claims that\\nthe School District lacked authority under the First\\nAmendment to discipline CLM for his hit list. The McNeils\\nalso allege that the School District’s expulsion of CLM\\nviolated their substantive due process right “to be free from\\nstate interference with their choice of [CLM’s] educational\\nforum.”\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 9\\n\\n II.\\n\\n We review “the district court’s grant of summary\\njudgment de novo.” Szajer v. City of Los Angeles, 632 F.3d\\n607, 610 (9th Cir. 2011). The court views “evidence in the\\nlight most favorable to the nonmoving party,” to determine\\n“whether genuine issues of material fact exist.” George v.\\nEdholm, 752 F.3d 1206, 1214 (9th Cir. 2014) (quoting\\nBagdadi v. Nazar, 84 F.3d 1194, 1197 (9th Cir. 1996)). If a\\nrational juror “‘could resolve a genuine issue of material fact\\nin the nonmoving party’s favor,’ summary judgment is\\ninappropriate.” Id. (quoting Bravo v. City of Santa Maria,\\n665 F.3d 1076, 1083 (9th Cir. 2011)).\\n\\n III.\\n\\n “The first inquiry in any § 1983 suit . . . is whether the\\nplaintiff has been deprived of a right ‘secured by the\\nConstitution and laws.’” Baker v. McCollan, 443 U.S. 137,\\n140 (1979). Section 1983 does not create substantive rights,\\nbut rather provides the procedural mechanism for vindicating\\nfederal statutory or constitutional rights. Id. at 145 n.3.\\n\\n A. CLM’s First Amendment Claim\\n\\n The Supreme Court has held that public school students\\ndo not forfeit “their constitutional rights to freedom of speech\\nor expression at the schoolhouse gate.” Tinker v. Des Moines\\nIndep. Cnty. Sch. Dist., 393 U.S. 503, 506 (1969). However,\\ntheir rights are “not automatically coextensive with the rights\\nof adults in other settings.” Wynar, 728 F.3d at 1067 (quoting\\nBethel Sch. Dist. No. 403 v. Fraser, 478 U.S. 675, 682\\n(1986)). Additionally, although they enjoy greater freedom\\nto speak when they are off campus than when they are on\\n\\f10 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\ncampus, their off-campus speech is not necessarily beyond\\nthe reach of a school district’s regulatory authority. Our\\nreview of the School District’s treatment of CLM turns on\\ntwo inquiries: (1) whether the School District could\\npermissibly regulate CLM’s off-campus speech at all; and if\\nso, (2) whether the School District’s decision to expel CLM\\nviolated the First Amendment standard for school regulation\\nof speech set out in Tinker. See C.R. v. Eugene Sch. Dist. 4J,\\n835 F.3d 1142, 1148 (9th Cir. 2016).\\n\\n 1. The School District’s authority to regulate CLM’s\\n off-campus speech\\n\\n We have twice considered the authority of a school\\ndistrict to regulate off-campus speech. In Wynar, a high\\nschool student was expelled because he sent graphic instant\\nmessages to his friends from his home computer. 728 F.3d\\nat 1064–67. The messages expressed threats of school\\nviolence and indicated that the student had access to guns and\\nammunition. Id. at 1065–66. The messages were brought to\\nthe school’s attention by the student’s friends. Id. at 1066.\\nDespite the student’s lack of intent in bringing his speech to\\nthe school campus, we held that his expulsion was\\nconstitutional, reasoning that “when faced with an identifiable\\nthreat of school violence, schools may take disciplinary\\naction in response to off-campus speech . . . .” Id. at 1069.\\n\\n In C.R., a school district suspended a student for sexually\\nharassing two students shortly after school hours in a public\\npark near the school. 835 F.3d at 1145–47. The student\\nargued that the school’s regulation of his harassment was\\nunconstitutional because the harassment occurred in a public,\\noff-campus park. Id. at 1145–46. While we recognized that\\nthe harassment constituted off-campus speech, we held that\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 11\\n\\nthe suspension was constitutional. Id. at 1146. We reasoned\\nthat the speech was closely tied to the school, noting that all\\nof the individuals involved were students and that the incident\\noccurred in close proximity to the school and just minutes\\nafter the end of the school day. Id. at 1150–51. We also\\nnoted that school administrators could reasonably expect the\\neffects of the speech to disrupt the school environment. Id.\\nat 1151.\\n\\n Although our previous cases have not settled on a test for\\ndetermining when a school can constitutionally regulate off-\\ncampus speech, we have stressed the importance of flexibility\\nin dealing with the “myriad of circumstances” schools face.\\nWynar, 728 F.3d at 1069; see also C.R., 835 F.3d at 1150–51.\\nThese cases have noted two approaches: (1) requiring that\\noff-campus speech “have a sufficient ‘nexus’ to the school,”\\nWynar, 728 F.3d at 1068 (quoting Kowalski v. Berkeley Cty.\\nSchs., 652 F.3d 565, 573 (4th Cir. 2011)); and (2) requiring\\n“that it be ‘reasonably foreseeable that the speech will reach\\nthe school community,’” id. (quoting S.J.W. v. Lee’s Summit\\nR-7 Sch. Dist., 696 F.3d 771, 777 (8th Cir. 2012)).\\n\\n We now clarify that courts considering whether a school\\ndistrict may constitutionally regulate off-campus speech must\\ndetermine, based on the totality of the circumstances, whether\\nthe speech bears a sufficient nexus to the school. See C.R.,\\n835 F.3d at 1150–51; Wynar, 728 F.3d at 1069. This test is\\nflexible and fact-specific, but the relevant considerations will\\ninclude (1) the degree and likelihood of harm to the school\\ncaused or augured by the speech, see Wynar, 728 F.3d at\\n1069, (2) whether it was reasonably foreseeable that the\\nspeech would reach and impact the school, see C.R., 835 F.3d\\nat 1150–51; Wynar, 728 F.3d at 1069; Kowalski, 652 F.3d at\\n573, and (3) the relation between the content and context of\\n\\f12 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nthe speech and the school, see C.R., 835 F.3d at 1150–51;\\nWynar, 728 F.3d at 1069.2 There is always a sufficient nexus\\nbetween the speech and the school when the school district\\nreasonably concludes that it faces a credible, identifiable\\nthreat of school violence. See Wynar, 728 F.3d at 1069.\\n\\n CLM attempts to distinguish Wynar on the ground that he\\nhad no intent to communicate the contents of his speech to\\nanyone. That distinction cannot be dispositive. We have\\nrecognized repeatedly that the specter of school violence\\nplaces a weighty social responsibility on school districts to\\nensure that “warning signs” do not turn to tragedy. See\\nLaVine v. Blaine Sch. Dist., 257 F.3d 981, 987 (9th Cir.\\n2001). This responsibility does not mean schools may “expel\\nstudents just because they are ‘loners,’ wear black and play\\nvideo games.” Id. It does mean, however, that a student’s\\nintent, although relevant, does not necessarily define the\\nthreat of violence. We reaffirm our holding in Wynar that\\nregardless of the speaker’s intent or how speech comes to a\\nschool district’s attention, a school district may take\\ndisciplinary action in response to off-campus speech when it\\nreasonably determines that it faces an identifiable and\\ncredible threat of school violence. See Wynar, 728 F.3d at\\n1069.\\n\\n Of course, any information school officials have\\nregarding the intention of the speaker may bear on whether a\\nthreat is credible. The intent to keep a threat private,\\n\\n\\n 2\\n In analyzing foreseeability, we have considered both whether it\\nwas foreseeable to the speaker that the speech would reach the school, see\\nWynar, 728 F.3d at 1068, and whether it was foreseeable to school\\nadministrators that the speech would impact the school community, see\\nC.R., 835 F.3d at 1151.\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 13\\n\\nhowever, is not determinative—it would be absurd to suggest\\nthat secret threats or planning cannot give rise to reasonable\\nconcerns of school violence. See, e.g., Kristin McCauley,\\nArapahoe Cty. Sheriff, Investigative Report: Arapahoe\\nHigh School Case # CT13-44545, at 29–31 (2014),\\nhttps://mediaassets.thedenverchannel.com/document/2014/\\n10/10/AHSShootingReport_8903184_ver1.0.pdf?_ga=2.15\\n5919036.684175427.1543265365-980564543.1543265365\\n(describing password-protected diary entries containing\\nplanning for a fatal 2013 school shooting).\\n\\n Here, the School District reasonably determined CLM\\npresented a credible threat. The School District knew CLM\\nhad identified specific targets, had accentuated his hit list\\nwith the phrases “I am God” and “All These People Must\\nDie,” lived in a gun-owning home close to the school, and\\nhad had thoughts of suicide. The School District also knew\\nthe diary contained other graphic depictions of school\\nviolence. This evidence was sufficient to render the School\\nDistrict’s determination reasonable and to give it authority to\\nregulate CLM’s speech.\\n\\n CLM argues that, once he was in police custody, he\\nceased to pose a safety threat because he was being evaluated\\nin the care of competent authorities. This evaluation,\\nhowever, lasted only one day, and afterward CLM returned\\nhome with limited oversight and his parents retrieved the\\nfamily’s firearms. The police’s decision to release CLM\\ncannot prevent the school from addressing the threat of\\nviolence: a school may take action even where police or\\nmental health professionals have elected not to do so. See\\nLaVine, 257 F.3d at 990 (“It is true that the [police and a\\npsychiatrist] did not believe [the student at issue] should be\\ninvoluntarily committed, but the standard for involuntary\\n\\f14 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\ncommitment is not the same as that for school officials to take\\naction.”). Schools must be permitted to act preventatively,\\ntaking into account other students’ interest in a safe school\\nenvironment. Here, the school reasonably determined the risk\\nwas sufficient to take action.\\n\\n Our test does not allow a school to take disciplinary\\naction in response to just any perceived threat of school\\nviolence arising from off-campus speech. Although our\\nreview of a school’s determinations in this regard is\\ndeferential, “deference does not mean abdication; there are\\nsituations where school officials overstep their bounds and\\nviolate the Constitution.” Id. at 988.\\n\\n In Porter v. Ascension Parish School Board, 393 F.3d 608\\n(5th Cir. 2004), for example, Adam, a high school student,\\ndepicted a military attack on his school in his sketchpad. Id.\\nat 611. There were no students identified in the drawing and\\nAdam stored the sketchpad in his closet at home. Id. Two\\nyears later, Adam’s brother brought the sketchpad to the\\nschool. Id. Another student, with permission from Adam’s\\nbrother, flipped through the sketchpad and, upon coming\\nacross Adam’s drawing of the military attack, exclaimed,\\n“they’re going to blow up [the school].” Id. This brought the\\ndrawing to the attention of school authorities, who expelled\\nAdam for the drawing. Id. at 611–12. The Fifth Circuit held\\nthat Adam’s drawing was protected by the First Amendment\\nbecause it did not constitute “speech on campus or even\\nspeech directed at the campus.” Id. at 615. The Fifth Circuit\\nnoted that for Adam’s drawing to lose First Amendment\\nprotection, something more than “accidental and\\nunintentional exposure to public scrutiny must take place.”\\nId. at 618. The Fifth Circuit reasoned that because Adam’s\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 15\\n\\ndrawing was not intentionally communicated, it was not a\\ntrue threat and therefore not subject to school regulation. Id.\\n\\n We do not adopt the Fifth Circuit’s reasoning, but note\\nthat our approach supports the result in Porter and that there\\nare critical distinctions between the facts here and those in\\nPorter. There, the drawing was two years old, highly\\nfantastical, unspecific, and unaccompanied by other indicia\\nof a violent intent. See id. at 611. There was no evidence\\nAdam had ever had the capacity to carry out the actions\\ndepicted in the drawing. In contrast to this case, it was\\nunlikely that any reasonable administrator could have\\nbelieved the school faced a credible safety threat.\\n\\n Taken as a whole, the three considerations that guide\\napplication of the nexus test support the School District here.\\nFirst, it was reasonable for School District officials to\\nconclude that CLM presented a credible threat of severe harm\\nto the school community. This consideration establishes a\\nsufficient nexus between the speech and the school to permit\\nregulation. See Wynar, 728 F.3d at 1069.\\n\\n Second, once it learned of the hit list, the School District\\ncould reasonably foresee that news of the threat would reach\\nand impact the school and disrupt the school environment.\\nOregon’s notification requirement, Oregon Revised Statutes\\n§ 339.327(1), provides that “[a] superintendent of a school\\ndistrict . . . who has reasonable cause to believe that a person,\\nwhile in a school, is or has been in possession of a list that\\nthreatens harm to other persons” must notify the “parent or\\nguardian of any student whose name appears on the list as a\\ntarget of the harm.” The statute applies to enrolled students\\nwho are or have been in possession of a threatening list,\\nregardless of whether the list is on or off school property. It\\n\\f16 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nwas reasonably foreseeable that parents, once notified as\\nrequired by Oregon law, would inform their children about\\nthe hit list. It was also reasonably foreseeable that\\ninformation about the hit list would cause students to be\\nfearful for their safety. Invariably, these resulting\\nconsequences would create classroom distractions and\\nsubstantial disruptions within the Sherwood High community.\\nAlthough it was not foreseeable to CLM that his speech\\nwould reach the school, a lack of intent to share speech is of\\nminimal weight when, as here, the speech contains a credible\\nthreat of violence directed at the school. Someone planning\\na violent act does not need to advertise it.\\n\\n Third, the content of the speech involved the school. As\\nwe noted in Wynar and C.R., when all involved parties are\\nstudents, that fact “typically counsels in favor of finding that\\na student’s speech was susceptible to discipline.” C.R., 835\\nF.3d at 1150 (citing Wynar, 728 F.3d at 1069). Here, CLM’s\\nhit list contained the names of 22 students and one former\\nteacher of the School District, and thus, presented a particular\\nthreat to the school community. Ordinarily, schools may not\\ndiscipline students for the contents of their private, off-\\ncampus diary entries, any more than they can punish students\\nfor their private thoughts, but schools have a right, indeed an\\nobligation, to address a credible threat of violence involving\\nthe school community.\\n\\n In sum, we conclude that the School District could\\nregulate CLM’s off-campus speech without violating his First\\nAmendment rights. Although CLM may not have foreseen\\nhis speech reaching Sherwood High, the School District,\\nwhen informed of CLM’s hit list, reasonably determined that\\nit faced a credible, identifiable threat of school violence. The\\nspeech bore a sufficient nexus to the school. Accordingly, the\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 17\\n\\nSchool District could take disciplinary action consistent with\\nTinker.\\n\\n 2. The School District’s expulsion of CLM\\n\\n Although some disciplinary action was permissible, there\\nremains the issue of whether the School District’s expulsion\\nwas consistent with the First Amendment. “Under Tinker,\\nschools may restrict speech that ‘might reasonably [lead]\\nschool authorities to forecast substantial disruption of or\\nmaterial interference with school activities’ or that collides\\n‘with the rights of other students to be secure and to be let\\nalone.’” Wynar, 728 F.3d at 1070 (alteration in original)\\n(quoting Tinker, 393 U.S. at 508, 514). This analysis\\nconsiders not just the actions of the disciplined student, but\\nrather “all of the circumstances confronting the school\\nofficials that might reasonably portend disruption.” LaVine,\\n257 F.3d at 989. We have held that “the specter of a school\\nshooting qualifies under either prong of Tinker.” Wynar, 728\\nF.3d at 1070. School districts should be afforded deference\\nin determining what discipline is necessary to maintain safety\\nand order, see id. at 1072, but we have cautioned that\\ndiscipline can extend beyond the scope Tinker allows when\\nit takes on a purely punitive character, see LaVine, 257 F.3d\\nat 992.\\n\\n Here, taking disciplinary action was reasonable under\\nboth Tinker prongs. As discussed, CLM’s hit list identifying\\nspecific individuals constituted a credible threat of school\\nviolence that required the School District to act in furtherance\\nof the safety of its students. Oregon’s statutory notification\\nrequirement further confirms this proposition, as the School\\nDistrict could reasonably foresee that informing parents that\\ntheir children were identified on a student-authored hit list\\n\\f18 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nwould cause a substantial disruption within the Sherwood\\nHigh community. Moreover, CLM’s hit list led to a material\\ndisruption at Sherwood High by causing safety concerns\\namong both students and parents. The record evinces this\\ndisruption by the outcry from parents who demanded to speak\\nwith Sherwood High authorities and by the parents who\\ntransferred or threatened to transfer their child out of the\\nSchool District. Thus, in light of the nature of CLM’s hit list\\nand the widespread knowledge regarding the list in the\\nSherwood High community, it was reasonable for School\\nDistrict officials to forecast that CLM’s presence at the\\nschool would cause a substantial disruption.\\n\\n Moreover, CLM’s hit list also invaded the “rights of other\\nstudents to be secure and to be let alone.” Tinker, 393 U.S. at\\n508. As Wynar stated, a student “target[ing] specific students\\nby name” for a potential school shooting “represent[s] the\\nquintessential harm to the rights of other students to be\\nsecure.” 728 F.3d at 1072.\\n\\n On appeal, CLM does not challenge the length of his\\nexpulsion—he argues only that the School District did not\\nhave the authority to expel him at all. We thus are not called\\nupon to determine whether a one-year expulsion was\\nexcessive. At a certain point, discipline may lose its basis in\\nreasonable, ongoing concerns of campus safety, disruption, or\\ninterference with the rights of other students, and instead\\nbecome primarily a punitive, retrospective response to the\\nstudent’s speech. Such discipline would be in conflict with\\nTinker. See LaVine, 257 F.3d at 991–92. Here, however, we\\ndecide only that CLM has not shown that the decision to\\nexpel him violates his constitutional rights under Tinker. The\\nSchool District could reasonably conclude that CLM’s\\ncontinued presence at the school at the time of his expulsion\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 19\\n\\nwould have caused a “substantial disruption of, or material\\ninterference with school activities” and would have interfered\\n“with the rights of other students to be secure and to be let\\nalone.” Wynar, 728 F.3d at 1070 (quoting Tinker, 393 U.S.\\nat 508, 514).\\n\\n B. The McNeils’ Substantive Due Process Claim\\n\\n Finally, we address the McNeils’ substantive due process\\nclaim. The Fourteenth Amendment’s Substantive Due\\nProcess Clause “provides heightened protection against\\ngovernment interference with certain fundamental rights and\\nliberty interests,” such as a parent’s right to make decisions\\nregarding the “care, custody, and control of their children.”\\nTroxel v. Granville, 530 U.S. 57, 65 (2000) (quoting\\nWashington v. Glucksberg, 521 U.S. 702, 720 (1997)). One\\nsuch fundamental right is “the right of parents to be free from\\nstate interference with their choice of the educational forum\\nitself, a choice that ordinarily determines the type of\\neducation one’s child will receive.” Fields v. Palmdale Sch.\\nDist., 427 F.3d 1197, 1207 (9th Cir. 2005). However, once\\nparents determine their child’s educational forum, their\\nfundamental right to control the education is “substantially\\ndiminished.” Id. at 1206. Due process does not give parents\\nthe right to interfere with a public school’s operations\\nbecause issues such as school discipline, the content of\\nexaminations, and dress code are “issues of public education\\ngenerally ‘committed to the control of state and local\\nauthorities.’” Id. (quoting Blau v. Fort Thomas Pub. Sch.\\nDist., 401 F.3d 381, 395–96 (6th Cir. 2005)).\\n\\n Here, the School District did not deprive the McNeils of\\ntheir liberty interest in determining where CLM completes his\\nhigh school education or whether CLM attends public school.\\n\\f20 MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J\\n\\nThe McNeils’ fundamental right allowed them to voluntarily\\nenroll CLM at Sherwood High. In doing so, they accepted\\nSherwood High’s curriculum, school policies, and reasonable\\ndisciplinary measures.\\n\\n IV.\\n\\n The number of reported tragic school shootings over the\\npast two decades emphasizes the need for school districts to\\nhave the authority to take disciplinary action when faced with\\na credible threat of school violence. See Wynar, 728 F.3d at\\n1069. Although a student’s expectation as to the circulation\\nof his speech may be relevant to an evaluation of whether the\\nspeech constitutes a credible threat, the student’s intent as to\\ncirculation does not condition the school’s regulation of\\nthreatening off-campus speech. As we noted in Wynar, “[w]e\\ncan only imagine what would have happened if the school\\nofficials, after learning of [the] writing, did nothing about it,”\\nand CLM did in fact come to Sherwood High with a firearm\\nand the intent to carry out his hit list. 728 F.3d at 1070\\n(alterations in original) (quoting Boim v. Fulton Cty. Sch.\\nDist., 494 F.3d 978, 984 (11th Cir. 2007)).\\n\\n We hold that the School District’s decision to discipline\\nCLM for his off-campus speech did not violate CLM’s\\nconstitutional right to free speech. A student’s lack of intent\\nto convey his off-campus speech to any third party is relevant\\nto an evaluation of whether the speech constitutes a credible\\nthreat, but is not dispositive. Rather, a court must consider all\\nthe relevant factual circumstances when determining whether\\na school’s regulation of a student’s off-campus speech is\\nconstitutional. Here, CLM’s lack of intent to communicate\\nthe contents of his hit list is offset by the fact that his speech\\ncreated a credible, identifiable threat of school violence.\\n\\f MCNEIL V. SHERWOOD SCHOOL DISTRICT 88J 21\\n\\n The McNeils’ substantive due process claim also fails\\nbecause their fundamental right to choose CLM’s educational\\nforum was not infringed by the School District’s discipline of\\nCLM.\\n\\n The district court’s judgment for Defendants is\\nAFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1292,"cells":{"id":{"kind":"number","value":4603419,"string":"4,603,419"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Martin Valdez, Jr. v. W. Montgomery"},"decision_date":{"kind":"string","value":"2019-03-14"},"docket_number":{"kind":"string","value":"16-56845"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Ronald M. Gould, Jacqueline H. Nguyen, and John B. Owens, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"GOULD, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nMARTIN LEYVA VALDEZ, JR., No. 16-56845\\n Petitioner-Appellant,\\n D.C. No.\\n v. 5:16-cv-00567-\\n VAP-DTB\\nW. L. MONTGOMERY, Acting Warden,\\n Respondent-Appellee. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Virginia A. Phillips, Chief District Judge, Presiding\\n\\n Submitted February 5, 2019 *\\n Pasadena, California\\n\\n Filed March 14, 2019\\n\\n Before: Ronald M. Gould, Jacqueline H. Nguyen, and\\n John B. Owens, Circuit Judges.\\n\\n Opinion by Judge Gould\\n\\n\\n\\n\\n *\\n The panel unanimously concludes this case is suitable for decision\\nwithout oral argument. See Fed. R. App. P. 34(a)(2).\\n\\f2 VALDEZ V. MONTGOMERY\\n\\n SUMMARY **\\n\\n\\n Habeas Corpus\\n\\n The panel affirmed the district court’s dismissal of\\nCalifornia state prisoner Martin Leyva Valdez, Jr.’s federal\\nhabeas petition as untimely under the Antiterrorism and\\nEffective Death Penalty Act.\\n\\n The parties agreed that the petition was untimely unless\\nthe statute of limitations was tolled from May 15, 2014—\\nwhen the California Superior Court denied Valdez’s first\\nstate habeas petition—until April 29, 2015—when Valdez\\nfiled his second state habeas petition in the California Court\\nof Appeal.\\n\\n Because the question of whether Valdez’s second state\\nhabeas petition was timely filed in the Court of Appeal is an\\nentirely distinct issue from whether his habeas petition in the\\nSuperior Court was timely filed, the panel held that the “look\\nthrough” doctrine cannot answer whether the second state\\nhabeas petition was timely.\\n\\n The panel held that Valdez is not entitled to statutory\\ntolling. Because Valdez filed his second state habeas\\npetition before the California Supreme Court decided People\\nv. Elizalde, 351 P.3d 1010 (Cal. 2015), the panel rejected his\\ncontention that he can establish good cause for the delay by\\nwaiting until Elizalde was decided. The panel likewise\\nrejected Valdez’s contention that the size of the state-court\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f VALDEZ V. MONTGOMERY 3\\n\\nrecord and complexity of the case renders his delay\\nreasonable and establishes good cause, where Valdez offered\\nno explanation for why he could timely file his first petition\\nbut not his second.\\n\\n The panel concluded that the district court did not err by\\nnot ordering the State to respond and lodge the state-court\\nrecord.\\n\\n\\n COUNSEL\\n\\nStephanie M. Adraktas, Berkeley, California, for Petitioner-\\nAppellant.\\n\\nXavier Becerra, Attorney General of California; Julie L.\\nGarland, Senior Assistant Attorney General; Robin\\nUrbanski, Supervising Deputy Attorney General; Sharon L.\\nRhodes, Deputy Attorney General; Vincent P. LaPietra,\\nDeputy Attorney General; Office of the California Attorney\\nGeneral, San Diego, California; for Respondent-Appellee.\\n\\n\\n\\n OPINION\\n\\nGOULD, Circuit Judge:\\n\\n We once again consider whether a California-state\\nprisoner is entitled to statutory tolling under the\\nAntiterrorism and Effective Death Penalty Act of 1996\\n(“AEDPA”). Because we hold that Petitioner-Appellant\\nMartin Valdez is not, we affirm the district court’s dismissal\\nof Valdez’s federal habeas petition as untimely.\\n\\f4 VALDEZ V. MONTGOMERY\\n\\n I\\n\\n Over the course of two jury trials, Valdez was convicted\\nof murder, attempted murder, assault with a firearm, and\\nrobbery. People v. Valdez, No. E053309, 2013 WL\\n1770856, at *1 (Cal. Ct. App., Apr. 25, 2013) (unpublished).\\nThe trial court sentenced Valdez to life without the\\npossibility of parole, plus seventy years to life, plus nine\\nyears. Id. Valdez appealed his conviction to the California\\nCourt of Appeal, which affirmed. Id. at *2. The California\\nSupreme Court then denied Valdez’s petition for review on\\nJuly 31, 2013.\\n\\n Valdez filed his first state habeas petition in California\\nSuperior Court on April 10, 2014. The court denied that\\npetition on May 15, 2014. Almost one year later, in April\\n2015, Valdez filed his second state petition in the California\\nCourt of Appeal, asserting the same claims. 1 The court\\ndenied that petition without explanation. Valdez then filed\\nhis third state petition in the California Supreme Court on\\nJune 10, 2015, again raising the same claims. The court\\ndenied that petition without explanation. 2\\n\\n\\n\\n 1\\n California has a unique postconviction review system. Rather than\\nappealing adverse decisions, prisoners must file a new, original habeas\\npetition at each court level. In practice, however, California’s system\\noperates like a normal appellate system, and the Supreme Court and this\\ncourt treat it as analogous to a normal appellate system. See Evans v.\\nChavis, 546 U.S. 189, 192–93 (2006); Carey v. Saffold, 536 U.S. 214,\\n221–25 (2002); Curiel v. Miller, 830 F.3d 864, 870 n.3 (9th Cir. 2016)\\n(en banc).\\n 2\\n Valdez also filed a second round of state habeas petitions. Those\\npetitions are irrelevant to this appeal, except as briefly discussed below.\\n\\f VALDEZ V. MONTGOMERY 5\\n\\n Valdez constructively filed his current federal petition\\nfor writ of habeas corpus in the district court on March 1,\\n2016, raising the same claims he had raised in the state\\nproceedings. After an initial review, the district court\\nordered Valdez to show cause why his petition should not be\\ndismissed as untimely. Valdez responded that he is entitled\\nto tolling because he was waiting for the California Supreme\\nCourt to decide People v. Elizalde, 351 P.3d 1010 (Cal.\\n2015), a case highly relevant to one of Valdez’s claims. 3\\nValdez also argued that tolling applies because his case is\\ncomplex: it involved “two trials, over 6,000 pages of\\ntranscripts, and . . . the prosecution sought the death\\npenalty.”\\n\\n The Magistrate Judge was not persuaded and\\nrecommended that the district court dismiss Valdez’s\\npetition as untimely. Valdez objected to the magistrate’s\\nfindings and recommendations, but the district court adopted\\nthem and dismissed Valdez’s petition.\\n\\n Valdez filed a timely notice of appeal. We granted\\nValdez a Certificate of Appealability and appointed counsel.\\nOn appeal, Valdez contends that he is entitled to statutory\\ntolling. 4 In the alternative, he contends that we should\\n\\n\\n\\n\\n 3\\n The State concedes Elizalde is relevant.\\n\\n 4\\n The district court held that Valdez is not entitled to equitable\\ntolling. Valdez does not challenge that holding on appeal. Any\\nchallenge is therefore waived. See, e.g., Bohmker v. Oregon, 903 F.3d\\n1029, 1040 n.6 (9th Cir. 2018).\\n\\f6 VALDEZ V. MONTGOMERY\\n\\nremand the case to the district court for further factual\\ndevelopment. 5\\n\\n II\\n\\n Because Valdez’s conviction became final and he filed\\nhis federal habeas petition after the enactment of AEDPA,\\nAEDPA’s one-year statute of limitations applies. See\\nCampbell v. Henry, 614 F.3d 1056, 1058 (9th Cir. 2010).\\nWe review de novo the district court’s dismissal of Valdez’s\\nfederal habeas petition as untimely. Id.\\n\\n III\\n\\n AEDPA “affords a state prisoner one year from the end\\nof the direct review process in state court to apply in federal\\ncourt for a writ of habeas corpus . . . .” Campbell, 614 F.3d\\nat 1058. AEDPA’s statute of limitations is tolled, however,\\nwhile a “properly filed” state habeas petition is pending in\\nstate court. 28 U.S.C. § 2244(d)(2). “A state habeas petition\\nis ‘pending’ as long as the ordinary state collateral review\\nprocess continues.” Trigueros v. Adams, 658 F.3d 983, 988\\n(9th Cir. 2011) (citing Carey v. Saffold, 536 U.S. 214, 219–\\n20 (2002)).\\n\\n We have previously held that if a California prisoner\\ntimely files his or her initial state habeas petition, AEDPA’s\\nstatute of limitations is tolled while the state court considers\\nthe petition. See, e.g., Velasquez v. Kirkland, 639 F.3d 964,\\n967 (9th Cir. 2011). In California, a state habeas petition is\\n\\n 5\\n Valdez also contends that we should hold this case in abeyance\\npending the California Supreme Court’s response to the question we\\ncertified in Robinson v. Lewis, 795 F.3d 926 (9th Cir. 2015). We find it\\nunnecessary to do so. We can, and do, resolve this appeal based on\\ncurrent precedent.\\n\\f VALDEZ V. MONTGOMERY 7\\n\\n“timely if filed within a ‘reasonable time.’” Evans v. Chavis,\\n546 U.S. 189, 192 (2006) (quoting In re Harris, 855 P.2d\\n391, 398 n.7 (Cal. 1993)). If the state court denies that\\npetition, then as long as the prisoner timely files another\\npetition in a higher court, AEDPA’s statute of limitations is\\ntolled for “the days between (1) the time the lower state court\\nreached an adverse decision, and (2) the day [the prisoner]\\nfiled a petition in the higher state court.” Id. at 193; see also\\nSaffold, 536 U.S. at 221–25. This is often referred to as\\n“gap” tolling.\\n\\n But if a California prisoner does not timely file his or her\\nfirst state habeas petition, then the prisoner is not entitled to\\ntolling. See, e.g., Bonner v. Carey, 425 F.3d 1145, 1148–49\\n(9th Cir. 2005), amended by 439 F.3d 993 (9th Cir. 2006).\\nSimilarly, if a prisoner timely files his or her first state\\nhabeas petition but does not timely file a second petition,\\nthen the prisoner is not entitled to tolling for the period\\nfollowing the denial of the first petition. See, e.g.,\\nVelasquez, 639 F.3d at 968 (prisoner was not entitled to\\nstatutory tolling because he waited 91 days after denial of his\\nfirst state habeas petition to file his second, without\\nexplanation); Chaffer v. Prosper, 592 F.3d 1046, 1048 (9th\\nCir. 2010) (per curiam) (prisoner was not entitled to statutory\\ntolling because he waited 115 days after denial of his first\\nstate habeas petition to file his second, without explanation).\\n\\n The parties agree that unless the statute of limitations\\nwas tolled here from May 15, 2014—when the California\\nSuperior Court denied Valdez’s first state habeas petition—\\nuntil April 29, 2015—when Valdez filed his second state\\nhabeas petition in the California Court of Appeal—Valdez’s\\nfederal habeas petition was untimely. Valdez “bears the\\nburden of proving that the statute of limitation was tolled.”\\nBanjo v. Ayers, 614 F.3d 964, 967 (9th Cir. 2010). He makes\\n\\f8 VALDEZ V. MONTGOMERY\\n\\ntwo contentions as to why he is entitled to tolling. We\\naddress, and reject, both.\\n\\n A\\n\\n Valdez first contends that, because the California\\nSuperior Court held that his first state habeas petition was\\ntimely, and the California Court of Appeal denied his second\\nstate habeas petition without explanation, 6 under the “look\\nthrough” doctrine, we should presume that the Court of\\nAppeal adopted the Superior Court’s reasoning and held that\\nValdez’s second state habeas petition was timely.\\n\\n The “look through” doctrine provides that “[w]hen at\\nleast one state court has rendered a reasoned decision, but\\nthe last state court to reject a prisoner’s claim issues an order\\n‘whose text or accompanying opinion does not disclose the\\nreason for the judgment,’ we ‘look through’ the mute\\ndecision and presume the higher court agreed with and\\nadopted the reasons given by the lower court.” Curiel v.\\nMiller, 830 F.3d 864, 870 (9th Cir. 2016) (en banc) (quoting\\nYlst v. Nunnemaker, 501 U.S. 797, 802–06 (1991)). That\\ndoctrine has universally been applied in cases where the\\ncourt rendering a reasoned decision and a later court making\\na summary determination were facing precisely the same\\nissue. See, e.g., Ylst, 501 U.S. at 803 (“Where there has been\\none reasoned state judgment rejecting a federal claim, later\\nunexplained orders upholding that judgment or rejecting the\\nsame claim rest upon the same ground.”); Bonner, 425 F.3d\\nat 1148–49, 1148 n.13 (looking through unexplained\\ndecisions of the California Court of Appeal and California\\nSupreme Court to a California Superior Court decision\\n\\n 6\\n Its decision states: “The petition for writ of habeas corpus is\\nDENIED.”\\n\\f VALDEZ V. MONTGOMERY 9\\n\\nholding that a petition was untimely to conclude that the\\npetition was untimely and the prisoner was not entitled to\\nstatutory tolling); Casey v. Moore, 386 F.3d 896, 918 n.23\\n(2004) (“We cannot ‘look through’ to see what the state\\nappeals court did on the merits of Casey’s case, because the\\nmerits of the federal issue were not raised until Casey\\nappealed.”).\\n\\n For that reason, the “look through” doctrine does not\\nprovide a basis for holding that Valdez’s second state habeas\\npetition was timely. Whether Valdez’s second state habeas\\npetition was timely filed in the Court of Appeal is a different\\nand entirely distinct issue from whether his habeas petition\\nin the Superior Court was timely filed. Cf. Kernan v.\\nHinojosa, 136 S. Ct. 1603, 1606 (2016) (per curiam) (“look\\nthrough” doctrine inapplicable where “[i]mproper venue\\ncould not possibly have been a ground for the high court’s\\nsummary denial of Hinojosa’s claim”). Specifically, the\\nquestion here is whether Valdez timely filed his second state\\nhabeas petition in the Court of Appeal “within a ‘reasonable\\ntime’” following the Superior Court’s “adverse\\ndetermination.” Chavis, 546 U.S. at 191, 192–93 (quoting\\nIn re Harris, 855 P.2d at 398 n.7). The question is not\\nwhether Valdez’s first state habeas petition was timely—a\\nquestion the “look through” doctrine could answer. See\\nBonner, 425 F.3d at 1148–49, 1148 n.13. Because the “look\\nthrough” doctrine cannot answer whether Valdez’s second\\nstate habeas petition was timely, we address that question\\nourselves.\\n\\n B\\n\\n If a California court has held that a state habeas petition\\nwas timely or untimely, we are bound by that decision. See\\nRobinson v. Lewis, 795 F.3d 926, 929 (9th Cir. 2015) (“If a\\nCalifornia court states it has dismissed a state habeas petition\\n\\f10 VALDEZ V. MONTGOMERY\\n\\nbecause the petition was untimely, ‘that would be the end of\\nthe matter.’” (quoting Saffold, 536 U.S. at 226)); Trigueros,\\n658 F.3d at 990 (“[W]e rely on the California Supreme\\nCourt’s orders practice explained in Robbins and conclude\\nthat it considered Trigueros’s petition timely because the\\nCalifornia Supreme Court had the timeliness question before\\nit, and did not cite to cases involving a timeliness procedural\\nbar.”). Here, the California Court of Appeal dismissed\\nValdez’s second state habeas petition without explanation.\\nThat decision does not provide a basis for concluding that\\nthe petition was timely. See Chavis, 546 U.S. at 197 (“If the\\nappearance of the words ‘on the merits’ does not\\nautomatically warrant a holding that the filing was timely,\\nthe absence of those words could not automatically warrant\\na holding that the filing was timely.” (emphasis in original));\\nCuriel, 830 F.3d at 871 (“The Supreme Court has\\nadmonished us in the past not to assume that a California\\ncourt found a state habeas petition to be timely from the\\ncourt’s silence on the question.”).\\n\\n Instead, we “must . . . examine the delay . . . and\\ndetermine what the state courts would have held in respect\\nto timeliness.” Chavis, 546 U.S. at 198. The question is\\nwhether Valdez filed his second state habeas petition “within\\na ‘reasonable time.’” Id. at 192 (quoting In re Harris, 855\\nP.2d at 398 n.7). To answer that question, the U.S. Supreme\\nCourt has instructed that we reference the “‘short period[s]\\nof time,’ 30 to 60 days, that most States provide for filing an\\nappeal.” Id. at 201 (alteration in original) (quoting Saffold,\\n536 U.S. at 219). Heeding that instruction, we have\\n“indicated that the Supreme Court’s 60-day limit is the\\n‘benchmark’ from which we will not depart without a\\nshowing of good cause,” and we have held that unexplained\\ndelays of 81, 101, and 115 days are unreasonable. Robinson,\\n795 F.3d at 930–31.\\n\\f VALDEZ V. MONTGOMERY 11\\n\\n Here, the California Superior Court denied Valdez’s first\\nstate habeas petition on May 15, 2014. Valdez then filed his\\nsecond state habeas petition on April 29, 2015. Because\\nValdez waited almost one year to file his second state\\npetition (the gap was about two weeks short of a year), it was\\nuntimely under our decisions unless Valdez can establish\\ngood cause. See Robinson, 795 F.3d at 929 (“California\\ncourts allow a longer delay if the petitioner demonstrates\\ngood cause.” (citing In re Robbins, 959 P.2d 311, 317 (Cal.\\n1998)).\\n\\n Valdez presses two arguments on this point. First, he\\ncontends that he waited to file his second state habeas\\npetition until the California Supreme Court decided Elizalde,\\nwhich establishes good cause. We reject that contention.\\nValdez filed his second state habeas petition before the\\nCalifornia Supreme Court decided Elizalde, “so waiting\\nuntil [Elizalde] was decided does not explain the delay.”\\nWaldrip v. Hall, 548 F.3d 729, 737 (9th Cir. 2008). 7 In re\\nLucero, 132 Cal. Rptr. 3d 499, 503–04 (Ct. App. 2011)—on\\nwhich Valdez relies—is inapposite. Unlike the prisoner\\nthere, Valdez did not wait to file his second state habeas\\npetition until after a new decision; he filed it before.\\n\\n In the alternative, Valdez claims that his delay was\\nreasonable because of the size of the state-court record and\\ncomplexity of the case. This contention is likewise\\ninsufficient to establish good cause. Valdez timely filed his\\nfirst state habeas petition. That petition raised the same\\nclaims he then raised in his second state petition. He has\\n\\n 7\\n Valdez makes much of the fact that he filed a second round of\\nhabeas petitions based on Elizalde, asserting that doing so demonstrates\\nhis diligence. But that assertion does not explain why Valdez filed his\\nsecond state habeas petition before Elizalde was decided.\\n\\f12 VALDEZ V. MONTGOMERY\\n\\noffered no explanation for why he could timely file his first\\npetition but not his second. See Velasquez, 639 F.3d at 968\\n(“[E]ach of Velasquez’s habeas petitions is nearly identical\\nto the petition that came before it. It is not reasonable that\\nVelasquez’s counsel would need excess time essentially to\\nre-file an already-written brief.”).\\n\\n In summary, the district court correctly held that Valdez\\nis not entitled to statutory tolling.\\n\\n IV\\n\\n We next address Valdez’s contention that the district\\ncourt should not have dismissed his federal habeas petition\\nwithout requiring the State to respond and lodge the state-\\ncourt record. Valdez contends that the district court could\\nnot determine whether his state habeas petitions were timely\\nfiled absent the state-court record.\\n\\n A district court may summarily dismiss a federal habeas\\ncorpus petition sua sponte if “it plainly appears from the\\npetition and any attached exhibits that the petitioner is not\\nentitled to relief . . . .” Rule 4 of the Rules Governing Section\\n2254 Cases. A district court should do so, however, only\\nafter “provid[ing] the petitioner with adequate notice and an\\nopportunity to respond.” Herbst v. Cook, 260 F.3d 1039,\\n1043 (9th Cir. 2001). Moreover, because “federal habeas\\ncourts” have a duty to “independently [review] the basis for\\nthe state court’s decision,” a district court must “obtain and\\nreview the relevant portions of the state court record,” or\\nhold an evidentiary hearing, as necessary to discharge its\\nduty. Nasby v. McDaniel, 853 F.3d 1049, 1053 (9th Cir.\\n2017).\\n\\n Valdez, in his federal habeas petition, listed the date the\\nCalifornia Supreme Court denied his petition for review on\\n\\f VALDEZ V. MONTGOMERY 13\\n\\ndirect appeal, the dates he filed his state habeas petitions, and\\nthe dates the state courts denied those petitions. The district\\ncourt, after ordering Valdez to show cause and thereby\\nproviding Valdez “adequate notice and an opportunity to\\nrespond,” Herbst, 260 F.3d at 1043, had Valdez’s\\nexplanations for why he delayed in filing his second state\\nhabeas petition. Finally, the district court had the California\\nCourt of Appeal’s decision dismissing Valdez’s second state\\npetition because Valdez attached that decision as an exhibit\\nto his objections to the magistrate judge’s findings and\\nrecommendations. As demonstrated in Section III, the\\nforegoing information is all that is necessary to conclude that\\nValdez’s federal habeas petition was untimely. The district\\ncourt did not err by not ordering the State to respond and\\nlodge the state-court record. 8\\n\\n V\\n\\n Because we conclude that Valdez is not entitled to\\nstatutory tolling for the period following the California\\nSuperior Court’s denial of his first state habeas petition,\\nValdez’s federal habeas petition is untimely, and we affirm\\nthe district court’s dismissal of his petition.\\n\\n AFFIRMED.\\n\\n\\n\\n\\n 8\\n Valdez asks us to take judicial notice of various state-court\\ndocuments because they purportedly show why the district court needed\\nto order the State to lodge the state-court record. Because we can, and\\ndo, affirm the district court’s decision on the record before it, we DENY\\nValdez’s motion for judicial notice as MOOT.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4380672/\", \"author_raw\": \"GOULD, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"RONALD M GOULD"},"judge2":{"kind":"string","value":"JACQUELINE H NGUYEN"},"judge3":{"kind":"string","value":"JOHN B OWENS"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4603419/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Martin Valdez, Jr. v. W. Montgomery\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-14\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nMARTIN LEYVA VALDEZ, JR., No. 16-56845\n Petitioner-Appellant,\n D.C. No.\n v. 5:16-cv-00567-\n VAP-DTB\nW. L. MONTGOMERY, Acting Warden,\n Respondent-Appellee. OPINION\n\n\n\n Appeal from the United States District Court\n for the Central District of California\n Virginia A. Phillips, Chief District Judge, Presiding\n\n Submitted February 5, 2019 *\n Pasadena, California\n\n Filed March 14, 2019\n\n Before: Ronald M. Gould, Jacqueline H. Nguyen, and\n John B. Owens, Circuit Judges.\n\n Opinion by Judge Gould\n\n\n\n\n *\n The panel unanimously concludes this case is suitable for decision\nwithout oral argument. See Fed. R. App. P. 34(a)(2).\n\f2 VALDEZ V. MONTGOMERY\n\n SUMMARY **\n\n\n Habeas Corpus\n\n The panel affirmed the district court’s dismissal of\nCalifornia state prisoner Martin Leyva Valdez, Jr.’s federal\nhabeas petition as untimely under the Antiterrorism and\nEffective Death Penalty Act.\n\n The parties agreed that the petition was untimely unless\nthe statute of limitations was tolled from May 15, 2014—\nwhen the California Superior Court denied Valdez’s first\nstate habeas petition—until April 29, 2015—when Valdez\nfiled his second state habeas petition in the California Court\nof Appeal.\n\n Because the question of whether Valdez’s second state\nhabeas petition was timely filed in the Court of Appeal is an\nentirely distinct issue from whether his habeas petition in the\nSuperior Court was timely filed, the panel held that the “look\nthrough” doctrine cannot answer whether the second state\nhabeas petition was timely.\n\n The panel held that Valdez is not entitled to statutory\ntolling. Because Valdez filed his second state habeas\npetition before the California Supreme Court decided People\nv. Elizalde, 351 P.3d 1010 (Cal. 2015), the panel rejected his\ncontention that he can establish good cause for the delay by\nwaiting until Elizalde was decided. The panel likewise\nrejected Valdez’s contention that the size of the state-court\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f VALDEZ V. MONTGOMERY 3\n\nrecord and complexity of the case renders his delay\nreasonable and establishes good cause, where Valdez offered\nno explanation for why he could timely file his first petition\nbut not his second.\n\n The panel concluded that the district court did not err by\nnot ordering the State to respond and lodge the state-court\nrecord.\n\n\n COUNSEL\n\nStephanie M. Adraktas, Berkeley, California, for Petitioner-\nAppellant.\n\nXavier Becerra, Attorney General of California; Julie L.\nGarland, Senior Assistant Attorney General; Robin\nUrbanski, Supervising Deputy Attorney General; Sharon L.\nRhodes, Deputy Attorney General; Vincent P. LaPietra,\nDeputy Attorney General; Office of the California Attorney\nGeneral, San Diego, California; for Respondent-Appellee.\n\n\n\n OPINION\n\nGOULD, Circuit Judge:\n\n We once again consider whether a California-state\nprisoner is entitled to statutory tolling under the\nAntiterrorism and Effective Death Penalty Act of 1996\n(“AEDPA”). Because we hold that Petitioner-Appellant\nMartin Valdez is not, we affirm the district court’s dismissal\nof Valdez’s federal habeas petition as untimely.\n\f4 VALDEZ V. MONTGOMERY\n\n I\n\n Over the course of two jury trials, Valdez was convicted\nof murder, attempted murder, assault with a firearm, and\nrobbery. People v. Valdez, No. E053309, 2013 WL\n1770856, at *1 (Cal. Ct. App., Apr. 25, 2013) (unpublished).\nThe trial court sentenced Valdez to life without the\npossibility of parole, plus seventy years to life, plus nine\nyears. Id. Valdez appealed his conviction to the California\nCourt of Appeal, which affirmed. Id. at *2. The California\nSupreme Court then denied Valdez’s petition for review on\nJuly 31, 2013.\n\n Valdez filed his first state habeas petition in California\nSuperior Court on April 10, 2014. The court denied that\npetition on May 15, 2014. Almost one year later, in April\n2015, Valdez filed his second state petition in the California\nCourt of Appeal, asserting the same claims. 1 The court\ndenied that petition without explanation. Valdez then filed\nhis third state petition in the California Supreme Court on\nJune 10, 2015, again raising the same claims. The court\ndenied that petition without explanation. 2\n\n\n\n 1\n California has a unique postconviction review system. Rather than\nappealing adverse decisions, prisoners must file a new, original habeas\npetition at each court level. In practice, however, California’s system\noperates like a normal appellate system, and the Supreme Court and this\ncourt treat it as analogous to a normal appellate system. See Evans v.\nChavis, 546 U.S. 189, 192–93 (2006); Carey v. Saffold, 536 U.S. 214,\n221–25 (2002); Curiel v. Miller, 830 F.3d 864, 870 n.3 (9th Cir. 2016)\n(en banc).\n 2\n Valdez also filed a second round of state habeas petitions. Those\npetitions are irrelevant to this appeal, except as briefly discussed below.\n\f VALDEZ V. MONTGOMERY 5\n\n Valdez constructively filed his current federal petition\nfor writ of habeas corpus in the district court on March 1,\n2016, raising the same claims he had raised in the state\nproceedings. After an initial review, the district court\nordered Valdez to show cause why his petition should not be\ndismissed as untimely. Valdez responded that he is entitled\nto tolling because he was waiting for the California Supreme\nCourt to decide People v. Elizalde, 351 P.3d 1010 (Cal.\n2015), a case highly relevant to one of Valdez’s claims. 3\nValdez also argued that tolling applies because his case is\ncomplex: it involved “two trials, over 6,000 pages of\ntranscripts, and . . . the prosecution sought the death\npenalty.”\n\n The Magistrate Judge was not persuaded and\nrecommended that the district court dismiss Valdez’s\npetition as untimely. Valdez objected to the magistrate’s\nfindings and recommendations, but the district court adopted\nthem and dismissed Valdez’s petition.\n\n Valdez filed a timely notice of appeal. We granted\nValdez a Certificate of Appealability and appointed counsel.\nOn appeal, Valdez contends that he is entitled to statutory\ntolling. 4 In the alternative, he contends that we should\n\n\n\n\n 3\n The State concedes Elizalde is relevant.\n\n 4\n The district court held that Valdez is not entitled to equitable\ntolling. Valdez does not challenge that holding on appeal. Any\nchallenge is therefore waived. See, e.g., Bohmker v. Oregon, 903 F.3d\n1029, 1040 n.6 (9th Cir. 2018).\n\f6 VALDEZ V. MONTGOMERY\n\nremand the case to the district court for further factual\ndevelopment. 5\n\n II\n\n Because Valdez’s conviction became final and he filed\nhis federal habeas petition after the enactment of AEDPA,\nAEDPA’s one-year statute of limitations applies. See\nCampbell v. Henry, 614 F.3d 1056, 1058 (9th Cir. 2010).\nWe review de novo the district court’s dismissal of Valdez’s\nfederal habeas petition as untimely. Id.\n\n III\n\n AEDPA “affords a state prisoner one year from the end\nof the direct review process in state court to apply in federal\ncourt for a writ of habeas corpus . . . .” Campbell, 614 F.3d\nat 1058. AEDPA’s statute of limitations is tolled, however,\nwhile a “properly filed” state habeas petition is pending in\nstate court. 28 U.S.C. § 2244(d)(2). “A state habeas petition\nis ‘pending’ as long as the ordinary state collateral review\nprocess continues.” Trigueros v. Adams, 658 F.3d 983, 988\n(9th Cir. 2011) (citing Carey v. Saffold, 536 U.S. 214, 219–\n20 (2002)).\n\n We have previously held that if a California prisoner\ntimely files his or her initial state habeas petition, AEDPA’s\nstatute of limitations is tolled while the state court considers\nthe petition. See, e.g., Velasquez v. Kirkland, 639 F.3d 964,\n967 (9th Cir. 2011). In California, a state habeas petition is\n\n 5\n Valdez also contends that we should hold this case in abeyance\npending the California Supreme Court’s response to the question we\ncertified in Robinson v. Lewis, 795 F.3d 926 (9th Cir. 2015). We find it\nunnecessary to do so. We can, and do, resolve this appeal based on\ncurrent precedent.\n\f VALDEZ V. MONTGOMERY 7\n\n“timely if filed within a ‘reasonable time.’” Evans v. Chavis,\n546 U.S. 189, 192 (2006) (quoting In re Harris, 855 P.2d\n391, 398 n.7 (Cal. 1993)). If the state court denies that\npetition, then as long as the prisoner timely files another\npetition in a higher court, AEDPA’s statute of limitations is\ntolled for “the days between (1) the time the lower state court\nreached an adverse decision, and (2) the day [the prisoner]\nfiled a petition in the higher state court.” Id. at 193; see also\nSaffold, 536 U.S. at 221–25. This is often referred to as\n“gap” tolling.\n\n But if a California prisoner does not timely file his or her\nfirst state habeas petition, then the prisoner is not entitled to\ntolling. See, e.g., Bonner v. Carey, 425 F.3d 1145, 1148–49\n(9th Cir. 2005), amended by 439 F.3d 993 (9th Cir. 2006).\nSimilarly, if a prisoner timely files his or her first state\nhabeas petition but does not timely file a second petition,\nthen the prisoner is not entitled to tolling for the period\nfollowing the denial of the first petition. See, e.g.,\nVelasquez, 639 F.3d at 968 (prisoner was not entitled to\nstatutory tolling because he waited 91 days after denial of his\nfirst state habeas petition to file his second, without\nexplanation); Chaffer v. Prosper, 592 F.3d 1046, 1048 (9th\nCir. 2010) (per curiam) (prisoner was not entitled to statutory\ntolling because he waited 115 days after denial of his first\nstate habeas petition to file his second, without explanation).\n\n The parties agree that unless the statute of limitations\nwas tolled here from May 15, 2014—when the California\nSuperior Court denied Valdez’s first state habeas petition—\nuntil April 29, 2015—when Valdez filed his second state\nhabeas petition in the California Court of Appeal—Valdez’s\nfederal habeas petition was untimely. Valdez “bears the\nburden of proving that the statute of limitation was tolled.”\nBanjo v. Ayers, 614 F.3d 964, 967 (9th Cir. 2010). He makes\n\f8 VALDEZ V. MONTGOMERY\n\ntwo contentions as to why he is entitled to tolling. We\naddress, and reject, both.\n\n A\n\n Valdez first contends that, because the California\nSuperior Court held that his first state habeas petition was\ntimely, and the California Court of Appeal denied his second\nstate habeas petition without explanation, 6 under the “look\nthrough” doctrine, we should presume that the Court of\nAppeal adopted the Superior Court’s reasoning and held that\nValdez’s second state habeas petition was timely.\n\n The “look through” doctrine provides that “[w]hen at\nleast one state court has rendered a reasoned decision, but\nthe last state court to reject a prisoner’s claim issues an order\n‘whose text or accompanying opinion does not disclose the\nreason for the judgment,’ we ‘look through’ the mute\ndecision and presume the higher court agreed with and\nadopted the reasons given by the lower court.” Curiel v.\nMiller, 830 F.3d 864, 870 (9th Cir. 2016) (en banc) (quoting\nYlst v. Nunnemaker, 501 U.S. 797, 802–06 (1991)). That\ndoctrine has universally been applied in cases where the\ncourt rendering a reasoned decision and a later court making\na summary determination were facing precisely the same\nissue. See, e.g., Ylst, 501 U.S. at 803 (“Where there has been\none reasoned state judgment rejecting a federal claim, later\nunexplained orders upholding that judgment or rejecting the\nsame claim rest upon the same ground.”); Bonner, 425 F.3d\nat 1148–49, 1148 n.13 (looking through unexplained\ndecisions of the California Court of Appeal and California\nSupreme Court to a California Superior Court decision\n\n 6\n Its decision states: “The petition for writ of habeas corpus is\nDENIED.”\n\f VALDEZ V. MONTGOMERY 9\n\nholding that a petition was untimely to conclude that the\npetition was untimely and the prisoner was not entitled to\nstatutory tolling); Casey v. Moore, 386 F.3d 896, 918 n.23\n(2004) (“We cannot ‘look through’ to see what the state\nappeals court did on the merits of Casey’s case, because the\nmerits of the federal issue were not raised until Casey\nappealed.”).\n\n For that reason, the “look through” doctrine does not\nprovide a basis for holding that Valdez’s second state habeas\npetition was timely. Whether Valdez’s second state habeas\npetition was timely filed in the Court of Appeal is a different\nand entirely distinct issue from whether his habeas petition\nin the Superior Court was timely filed. Cf. Kernan v.\nHinojosa, 136 S. Ct. 1603, 1606 (2016) (per curiam) (“look\nthrough” doctrine inapplicable where “[i]mproper venue\ncould not possibly have been a ground for the high court’s\nsummary denial of Hinojosa’s claim”). Specifically, the\nquestion here is whether Valdez timely filed his second state\nhabeas petition in the Court of Appeal “within a ‘reasonable\ntime’” following the Superior Court’s “adverse\ndetermination.” Chavis, 546 U.S. at 191, 192–93 (quoting\nIn re Harris, 855 P.2d at 398 n.7). The question is not\nwhether Valdez’s first state habeas petition was timely—a\nquestion the “look through” doctrine could answer. See\nBonner, 425 F.3d at 1148–49, 1148 n.13. Because the “look\nthrough” doctrine cannot answer whether Valdez’s second\nstate habeas petition was timely, we address that question\nourselves.\n\n B\n\n If a California court has held that a state habeas petition\nwas timely or untimely, we are bound by that decision. See\nRobinson v. Lewis, 795 F.3d 926, 929 (9th Cir. 2015) (“If a\nCalifornia court states it has dismissed a state habeas petition\n\f10 VALDEZ V. MONTGOMERY\n\nbecause the petition was untimely, ‘that would be the end of\nthe matter.’” (quoting Saffold, 536 U.S. at 226)); Trigueros,\n658 F.3d at 990 (“[W]e rely on the California Supreme\nCourt’s orders practice explained in Robbins and conclude\nthat it considered Trigueros’s petition timely because the\nCalifornia Supreme Court had the timeliness question before\nit, and did not cite to cases involving a timeliness procedural\nbar.”). Here, the California Court of Appeal dismissed\nValdez’s second state habeas petition without explanation.\nThat decision does not provide a basis for concluding that\nthe petition was timely. See Chavis, 546 U.S. at 197 (“If the\nappearance of the words ‘on the merits’ does not\nautomatically warrant a holding that the filing was timely,\nthe absence of those words could not automatically warrant\na holding that the filing was timely.” (emphasis in original));\nCuriel, 830 F.3d at 871 (“The Supreme Court has\nadmonished us in the past not to assume that a California\ncourt found a state habeas petition to be timely from the\ncourt’s silence on the question.”).\n\n Instead, we “must . . . examine the delay . . . and\ndetermine what the state courts would have held in respect\nto timeliness.” Chavis, 546 U.S. at 198. The question is\nwhether Valdez filed his second state habeas petition “within\na ‘reasonable time.’” Id. at 192 (quoting In re Harris, 855\nP.2d at 398 n.7). To answer that question, the U.S. Supreme\nCourt has instructed that we reference the “‘short period[s]\nof time,’ 30 to 60 days, that most States provide for filing an\nappeal.” Id. at 201 (alteration in original) (quoting Saffold,\n536 U.S. at 219). Heeding that instruction, we have\n“indicated that the Supreme Court’s 60-day limit is the\n‘benchmark’ from which we will not depart without a\nshowing of good cause,” and we have held that unexplained\ndelays of 81, 101, and 115 days are unreasonable. Robinson,\n795 F.3d at 930–31.\n\f VALDEZ V. MONTGOMERY 11\n\n Here, the California Superior Court denied Valdez’s first\nstate habeas petition on May 15, 2014. Valdez then filed his\nsecond state habeas petition on April 29, 2015. Because\nValdez waited almost one year to file his second state\npetition (the gap was about two weeks short of a year), it was\nuntimely under our decisions unless Valdez can establish\ngood cause. See Robinson, 795 F.3d at 929 (“California\ncourts allow a longer delay if the petitioner demonstrates\ngood cause.” (citing In re Robbins, 959 P.2d 311, 317 (Cal.\n1998)).\n\n Valdez presses two arguments on this point. First, he\ncontends that he waited to file his second state habeas\npetition until the California Supreme Court decided Elizalde,\nwhich establishes good cause. We reject that contention.\nValdez filed his second state habeas petition before the\nCalifornia Supreme Court decided Elizalde, “so waiting\nuntil [Elizalde] was decided does not explain the delay.”\nWaldrip v. Hall, 548 F.3d 729, 737 (9th Cir. 2008). 7 In re\nLucero, 132 Cal. Rptr. 3d 499, 503–04 (Ct. App. 2011)—on\nwhich Valdez relies—is inapposite. Unlike the prisoner\nthere, Valdez did not wait to file his second state habeas\npetition until after a new decision; he filed it before.\n\n In the alternative, Valdez claims that his delay was\nreasonable because of the size of the state-court record and\ncomplexity of the case. This contention is likewise\ninsufficient to establish good cause. Valdez timely filed his\nfirst state habeas petition. That petition raised the same\nclaims he then raised in his second state petition. He has\n\n 7\n Valdez makes much of the fact that he filed a second round of\nhabeas petitions based on Elizalde, asserting that doing so demonstrates\nhis diligence. But that assertion does not explain why Valdez filed his\nsecond state habeas petition before Elizalde was decided.\n\f12 VALDEZ V. MONTGOMERY\n\noffered no explanation for why he could timely file his first\npetition but not his second. See Velasquez, 639 F.3d at 968\n(“[E]ach of Velasquez’s habeas petitions is nearly identical\nto the petition that came before it. It is not reasonable that\nVelasquez’s counsel would need excess time essentially to\nre-file an already-written brief.”).\n\n In summary, the district court correctly held that Valdez\nis not entitled to statutory tolling.\n\n IV\n\n We next address Valdez’s contention that the district\ncourt should not have dismissed his federal habeas petition\nwithout requiring the State to respond and lodge the state-\ncourt record. Valdez contends that the district court could\nnot determine whether his state habeas petitions were timely\nfiled absent the state-court record.\n\n A district court may summarily dismiss a federal habeas\ncorpus petition sua sponte if “it plainly appears from the\npetition and any attached exhibits that the petitioner is not\nentitled to relief . . . .” Rule 4 of the Rules Governing Section\n2254 Cases. A district court should do so, however, only\nafter “provid[ing] the petitioner with adequate notice and an\nopportunity to respond.” Herbst v. Cook, 260 F.3d 1039,\n1043 (9th Cir. 2001). Moreover, because “federal habeas\ncourts” have a duty to “independently [review] the basis for\nthe state court’s decision,” a district court must “obtain and\nreview the relevant portions of the state court record,” or\nhold an evidentiary hearing, as necessary to discharge its\nduty. Nasby v. McDaniel, 853 F.3d 1049, 1053 (9th Cir.\n2017).\n\n Valdez, in his federal habeas petition, listed the date the\nCalifornia Supreme Court denied his petition for review on\n\f VALDEZ V. MONTGOMERY 13\n\ndirect appeal, the dates he filed his state habeas petitions, and\nthe dates the state courts denied those petitions. The district\ncourt, after ordering Valdez to show cause and thereby\nproviding Valdez “adequate notice and an opportunity to\nrespond,” Herbst, 260 F.3d at 1043, had Valdez’s\nexplanations for why he delayed in filing his second state\nhabeas petition. Finally, the district court had the California\nCourt of Appeal’s decision dismissing Valdez’s second state\npetition because Valdez attached that decision as an exhibit\nto his objections to the magistrate judge’s findings and\nrecommendations. As demonstrated in Section III, the\nforegoing information is all that is necessary to conclude that\nValdez’s federal habeas petition was untimely. The district\ncourt did not err by not ordering the State to respond and\nlodge the state-court record. 8\n\n V\n\n Because we conclude that Valdez is not entitled to\nstatutory tolling for the period following the California\nSuperior Court’s denial of his first state habeas petition,\nValdez’s federal habeas petition is untimely, and we affirm\nthe district court’s dismissal of his petition.\n\n AFFIRMED.\n\n\n\n\n 8\n Valdez asks us to take judicial notice of various state-court\ndocuments because they purportedly show why the district court needed\nto order the State to lodge the state-court record. Because we can, and\ndo, affirm the district court’s decision on the record before it, we DENY\nValdez’s motion for judicial notice as MOOT.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Martin Valdez, Jr. v. W. Montgomery\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-14\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nMARTIN LEYVA VALDEZ, JR., No. 16-56845\\n Petitioner-Appellant,\\n D.C. No.\\n v. 5:16-cv-00567-\\n VAP-DTB\\nW. L. MONTGOMERY, Acting Warden,\\n Respondent-Appellee. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Virginia A. Phillips, Chief District Judge, Presiding\\n\\n Submitted February 5, 2019 *\\n Pasadena, California\\n\\n Filed March 14, 2019\\n\\n Before: Ronald M. Gould, Jacqueline H. Nguyen, and\\n John B. Owens, Circuit Judges.\\n\\n Opinion by Judge Gould\\n\\n\\n\\n\\n *\\n The panel unanimously concludes this case is suitable for decision\\nwithout oral argument. See Fed. R. App. P. 34(a)(2).\\n\\f2 VALDEZ V. MONTGOMERY\\n\\n SUMMARY **\\n\\n\\n Habeas Corpus\\n\\n The panel affirmed the district court’s dismissal of\\nCalifornia state prisoner Martin Leyva Valdez, Jr.’s federal\\nhabeas petition as untimely under the Antiterrorism and\\nEffective Death Penalty Act.\\n\\n The parties agreed that the petition was untimely unless\\nthe statute of limitations was tolled from May 15, 2014—\\nwhen the California Superior Court denied Valdez’s first\\nstate habeas petition—until April 29, 2015—when Valdez\\nfiled his second state habeas petition in the California Court\\nof Appeal.\\n\\n Because the question of whether Valdez’s second state\\nhabeas petition was timely filed in the Court of Appeal is an\\nentirely distinct issue from whether his habeas petition in the\\nSuperior Court was timely filed, the panel held that the “look\\nthrough” doctrine cannot answer whether the second state\\nhabeas petition was timely.\\n\\n The panel held that Valdez is not entitled to statutory\\ntolling. Because Valdez filed his second state habeas\\npetition before the California Supreme Court decided People\\nv. Elizalde, 351 P.3d 1010 (Cal. 2015), the panel rejected his\\ncontention that he can establish good cause for the delay by\\nwaiting until Elizalde was decided. The panel likewise\\nrejected Valdez’s contention that the size of the state-court\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f VALDEZ V. MONTGOMERY 3\\n\\nrecord and complexity of the case renders his delay\\nreasonable and establishes good cause, where Valdez offered\\nno explanation for why he could timely file his first petition\\nbut not his second.\\n\\n The panel concluded that the district court did not err by\\nnot ordering the State to respond and lodge the state-court\\nrecord.\\n\\n\\n COUNSEL\\n\\nStephanie M. Adraktas, Berkeley, California, for Petitioner-\\nAppellant.\\n\\nXavier Becerra, Attorney General of California; Julie L.\\nGarland, Senior Assistant Attorney General; Robin\\nUrbanski, Supervising Deputy Attorney General; Sharon L.\\nRhodes, Deputy Attorney General; Vincent P. LaPietra,\\nDeputy Attorney General; Office of the California Attorney\\nGeneral, San Diego, California; for Respondent-Appellee.\\n\\n\\n\\n OPINION\\n\\nGOULD, Circuit Judge:\\n\\n We once again consider whether a California-state\\nprisoner is entitled to statutory tolling under the\\nAntiterrorism and Effective Death Penalty Act of 1996\\n(“AEDPA”). Because we hold that Petitioner-Appellant\\nMartin Valdez is not, we affirm the district court’s dismissal\\nof Valdez’s federal habeas petition as untimely.\\n\\f4 VALDEZ V. MONTGOMERY\\n\\n I\\n\\n Over the course of two jury trials, Valdez was convicted\\nof murder, attempted murder, assault with a firearm, and\\nrobbery. People v. Valdez, No. E053309, 2013 WL\\n1770856, at *1 (Cal. Ct. App., Apr. 25, 2013) (unpublished).\\nThe trial court sentenced Valdez to life without the\\npossibility of parole, plus seventy years to life, plus nine\\nyears. Id. Valdez appealed his conviction to the California\\nCourt of Appeal, which affirmed. Id. at *2. The California\\nSupreme Court then denied Valdez’s petition for review on\\nJuly 31, 2013.\\n\\n Valdez filed his first state habeas petition in California\\nSuperior Court on April 10, 2014. The court denied that\\npetition on May 15, 2014. Almost one year later, in April\\n2015, Valdez filed his second state petition in the California\\nCourt of Appeal, asserting the same claims. 1 The court\\ndenied that petition without explanation. Valdez then filed\\nhis third state petition in the California Supreme Court on\\nJune 10, 2015, again raising the same claims. The court\\ndenied that petition without explanation. 2\\n\\n\\n\\n 1\\n California has a unique postconviction review system. Rather than\\nappealing adverse decisions, prisoners must file a new, original habeas\\npetition at each court level. In practice, however, California’s system\\noperates like a normal appellate system, and the Supreme Court and this\\ncourt treat it as analogous to a normal appellate system. See Evans v.\\nChavis, 546 U.S. 189, 192–93 (2006); Carey v. Saffold, 536 U.S. 214,\\n221–25 (2002); Curiel v. Miller, 830 F.3d 864, 870 n.3 (9th Cir. 2016)\\n(en banc).\\n 2\\n Valdez also filed a second round of state habeas petitions. Those\\npetitions are irrelevant to this appeal, except as briefly discussed below.\\n\\f VALDEZ V. MONTGOMERY 5\\n\\n Valdez constructively filed his current federal petition\\nfor writ of habeas corpus in the district court on March 1,\\n2016, raising the same claims he had raised in the state\\nproceedings. After an initial review, the district court\\nordered Valdez to show cause why his petition should not be\\ndismissed as untimely. Valdez responded that he is entitled\\nto tolling because he was waiting for the California Supreme\\nCourt to decide People v. Elizalde, 351 P.3d 1010 (Cal.\\n2015), a case highly relevant to one of Valdez’s claims. 3\\nValdez also argued that tolling applies because his case is\\ncomplex: it involved “two trials, over 6,000 pages of\\ntranscripts, and . . . the prosecution sought the death\\npenalty.”\\n\\n The Magistrate Judge was not persuaded and\\nrecommended that the district court dismiss Valdez’s\\npetition as untimely. Valdez objected to the magistrate’s\\nfindings and recommendations, but the district court adopted\\nthem and dismissed Valdez’s petition.\\n\\n Valdez filed a timely notice of appeal. We granted\\nValdez a Certificate of Appealability and appointed counsel.\\nOn appeal, Valdez contends that he is entitled to statutory\\ntolling. 4 In the alternative, he contends that we should\\n\\n\\n\\n\\n 3\\n The State concedes Elizalde is relevant.\\n\\n 4\\n The district court held that Valdez is not entitled to equitable\\ntolling. Valdez does not challenge that holding on appeal. Any\\nchallenge is therefore waived. See, e.g., Bohmker v. Oregon, 903 F.3d\\n1029, 1040 n.6 (9th Cir. 2018).\\n\\f6 VALDEZ V. MONTGOMERY\\n\\nremand the case to the district court for further factual\\ndevelopment. 5\\n\\n II\\n\\n Because Valdez’s conviction became final and he filed\\nhis federal habeas petition after the enactment of AEDPA,\\nAEDPA’s one-year statute of limitations applies. See\\nCampbell v. Henry, 614 F.3d 1056, 1058 (9th Cir. 2010).\\nWe review de novo the district court’s dismissal of Valdez’s\\nfederal habeas petition as untimely. Id.\\n\\n III\\n\\n AEDPA “affords a state prisoner one year from the end\\nof the direct review process in state court to apply in federal\\ncourt for a writ of habeas corpus . . . .” Campbell, 614 F.3d\\nat 1058. AEDPA’s statute of limitations is tolled, however,\\nwhile a “properly filed” state habeas petition is pending in\\nstate court. 28 U.S.C. § 2244(d)(2). “A state habeas petition\\nis ‘pending’ as long as the ordinary state collateral review\\nprocess continues.” Trigueros v. Adams, 658 F.3d 983, 988\\n(9th Cir. 2011) (citing Carey v. Saffold, 536 U.S. 214, 219–\\n20 (2002)).\\n\\n We have previously held that if a California prisoner\\ntimely files his or her initial state habeas petition, AEDPA’s\\nstatute of limitations is tolled while the state court considers\\nthe petition. See, e.g., Velasquez v. Kirkland, 639 F.3d 964,\\n967 (9th Cir. 2011). In California, a state habeas petition is\\n\\n 5\\n Valdez also contends that we should hold this case in abeyance\\npending the California Supreme Court’s response to the question we\\ncertified in Robinson v. Lewis, 795 F.3d 926 (9th Cir. 2015). We find it\\nunnecessary to do so. We can, and do, resolve this appeal based on\\ncurrent precedent.\\n\\f VALDEZ V. MONTGOMERY 7\\n\\n“timely if filed within a ‘reasonable time.’” Evans v. Chavis,\\n546 U.S. 189, 192 (2006) (quoting In re Harris, 855 P.2d\\n391, 398 n.7 (Cal. 1993)). If the state court denies that\\npetition, then as long as the prisoner timely files another\\npetition in a higher court, AEDPA’s statute of limitations is\\ntolled for “the days between (1) the time the lower state court\\nreached an adverse decision, and (2) the day [the prisoner]\\nfiled a petition in the higher state court.” Id. at 193; see also\\nSaffold, 536 U.S. at 221–25. This is often referred to as\\n“gap” tolling.\\n\\n But if a California prisoner does not timely file his or her\\nfirst state habeas petition, then the prisoner is not entitled to\\ntolling. See, e.g., Bonner v. Carey, 425 F.3d 1145, 1148–49\\n(9th Cir. 2005), amended by 439 F.3d 993 (9th Cir. 2006).\\nSimilarly, if a prisoner timely files his or her first state\\nhabeas petition but does not timely file a second petition,\\nthen the prisoner is not entitled to tolling for the period\\nfollowing the denial of the first petition. See, e.g.,\\nVelasquez, 639 F.3d at 968 (prisoner was not entitled to\\nstatutory tolling because he waited 91 days after denial of his\\nfirst state habeas petition to file his second, without\\nexplanation); Chaffer v. Prosper, 592 F.3d 1046, 1048 (9th\\nCir. 2010) (per curiam) (prisoner was not entitled to statutory\\ntolling because he waited 115 days after denial of his first\\nstate habeas petition to file his second, without explanation).\\n\\n The parties agree that unless the statute of limitations\\nwas tolled here from May 15, 2014—when the California\\nSuperior Court denied Valdez’s first state habeas petition—\\nuntil April 29, 2015—when Valdez filed his second state\\nhabeas petition in the California Court of Appeal—Valdez’s\\nfederal habeas petition was untimely. Valdez “bears the\\nburden of proving that the statute of limitation was tolled.”\\nBanjo v. Ayers, 614 F.3d 964, 967 (9th Cir. 2010). He makes\\n\\f8 VALDEZ V. MONTGOMERY\\n\\ntwo contentions as to why he is entitled to tolling. We\\naddress, and reject, both.\\n\\n A\\n\\n Valdez first contends that, because the California\\nSuperior Court held that his first state habeas petition was\\ntimely, and the California Court of Appeal denied his second\\nstate habeas petition without explanation, 6 under the “look\\nthrough” doctrine, we should presume that the Court of\\nAppeal adopted the Superior Court’s reasoning and held that\\nValdez’s second state habeas petition was timely.\\n\\n The “look through” doctrine provides that “[w]hen at\\nleast one state court has rendered a reasoned decision, but\\nthe last state court to reject a prisoner’s claim issues an order\\n‘whose text or accompanying opinion does not disclose the\\nreason for the judgment,’ we ‘look through’ the mute\\ndecision and presume the higher court agreed with and\\nadopted the reasons given by the lower court.” Curiel v.\\nMiller, 830 F.3d 864, 870 (9th Cir. 2016) (en banc) (quoting\\nYlst v. Nunnemaker, 501 U.S. 797, 802–06 (1991)). That\\ndoctrine has universally been applied in cases where the\\ncourt rendering a reasoned decision and a later court making\\na summary determination were facing precisely the same\\nissue. See, e.g., Ylst, 501 U.S. at 803 (“Where there has been\\none reasoned state judgment rejecting a federal claim, later\\nunexplained orders upholding that judgment or rejecting the\\nsame claim rest upon the same ground.”); Bonner, 425 F.3d\\nat 1148–49, 1148 n.13 (looking through unexplained\\ndecisions of the California Court of Appeal and California\\nSupreme Court to a California Superior Court decision\\n\\n 6\\n Its decision states: “The petition for writ of habeas corpus is\\nDENIED.”\\n\\f VALDEZ V. MONTGOMERY 9\\n\\nholding that a petition was untimely to conclude that the\\npetition was untimely and the prisoner was not entitled to\\nstatutory tolling); Casey v. Moore, 386 F.3d 896, 918 n.23\\n(2004) (“We cannot ‘look through’ to see what the state\\nappeals court did on the merits of Casey’s case, because the\\nmerits of the federal issue were not raised until Casey\\nappealed.”).\\n\\n For that reason, the “look through” doctrine does not\\nprovide a basis for holding that Valdez’s second state habeas\\npetition was timely. Whether Valdez’s second state habeas\\npetition was timely filed in the Court of Appeal is a different\\nand entirely distinct issue from whether his habeas petition\\nin the Superior Court was timely filed. Cf. Kernan v.\\nHinojosa, 136 S. Ct. 1603, 1606 (2016) (per curiam) (“look\\nthrough” doctrine inapplicable where “[i]mproper venue\\ncould not possibly have been a ground for the high court’s\\nsummary denial of Hinojosa’s claim”). Specifically, the\\nquestion here is whether Valdez timely filed his second state\\nhabeas petition in the Court of Appeal “within a ‘reasonable\\ntime’” following the Superior Court’s “adverse\\ndetermination.” Chavis, 546 U.S. at 191, 192–93 (quoting\\nIn re Harris, 855 P.2d at 398 n.7). The question is not\\nwhether Valdez’s first state habeas petition was timely—a\\nquestion the “look through” doctrine could answer. See\\nBonner, 425 F.3d at 1148–49, 1148 n.13. Because the “look\\nthrough” doctrine cannot answer whether Valdez’s second\\nstate habeas petition was timely, we address that question\\nourselves.\\n\\n B\\n\\n If a California court has held that a state habeas petition\\nwas timely or untimely, we are bound by that decision. See\\nRobinson v. Lewis, 795 F.3d 926, 929 (9th Cir. 2015) (“If a\\nCalifornia court states it has dismissed a state habeas petition\\n\\f10 VALDEZ V. MONTGOMERY\\n\\nbecause the petition was untimely, ‘that would be the end of\\nthe matter.’” (quoting Saffold, 536 U.S. at 226)); Trigueros,\\n658 F.3d at 990 (“[W]e rely on the California Supreme\\nCourt’s orders practice explained in Robbins and conclude\\nthat it considered Trigueros’s petition timely because the\\nCalifornia Supreme Court had the timeliness question before\\nit, and did not cite to cases involving a timeliness procedural\\nbar.”). Here, the California Court of Appeal dismissed\\nValdez’s second state habeas petition without explanation.\\nThat decision does not provide a basis for concluding that\\nthe petition was timely. See Chavis, 546 U.S. at 197 (“If the\\nappearance of the words ‘on the merits’ does not\\nautomatically warrant a holding that the filing was timely,\\nthe absence of those words could not automatically warrant\\na holding that the filing was timely.” (emphasis in original));\\nCuriel, 830 F.3d at 871 (“The Supreme Court has\\nadmonished us in the past not to assume that a California\\ncourt found a state habeas petition to be timely from the\\ncourt’s silence on the question.”).\\n\\n Instead, we “must . . . examine the delay . . . and\\ndetermine what the state courts would have held in respect\\nto timeliness.” Chavis, 546 U.S. at 198. The question is\\nwhether Valdez filed his second state habeas petition “within\\na ‘reasonable time.’” Id. at 192 (quoting In re Harris, 855\\nP.2d at 398 n.7). To answer that question, the U.S. Supreme\\nCourt has instructed that we reference the “‘short period[s]\\nof time,’ 30 to 60 days, that most States provide for filing an\\nappeal.” Id. at 201 (alteration in original) (quoting Saffold,\\n536 U.S. at 219). Heeding that instruction, we have\\n“indicated that the Supreme Court’s 60-day limit is the\\n‘benchmark’ from which we will not depart without a\\nshowing of good cause,” and we have held that unexplained\\ndelays of 81, 101, and 115 days are unreasonable. Robinson,\\n795 F.3d at 930–31.\\n\\f VALDEZ V. MONTGOMERY 11\\n\\n Here, the California Superior Court denied Valdez’s first\\nstate habeas petition on May 15, 2014. Valdez then filed his\\nsecond state habeas petition on April 29, 2015. Because\\nValdez waited almost one year to file his second state\\npetition (the gap was about two weeks short of a year), it was\\nuntimely under our decisions unless Valdez can establish\\ngood cause. See Robinson, 795 F.3d at 929 (“California\\ncourts allow a longer delay if the petitioner demonstrates\\ngood cause.” (citing In re Robbins, 959 P.2d 311, 317 (Cal.\\n1998)).\\n\\n Valdez presses two arguments on this point. First, he\\ncontends that he waited to file his second state habeas\\npetition until the California Supreme Court decided Elizalde,\\nwhich establishes good cause. We reject that contention.\\nValdez filed his second state habeas petition before the\\nCalifornia Supreme Court decided Elizalde, “so waiting\\nuntil [Elizalde] was decided does not explain the delay.”\\nWaldrip v. Hall, 548 F.3d 729, 737 (9th Cir. 2008). 7 In re\\nLucero, 132 Cal. Rptr. 3d 499, 503–04 (Ct. App. 2011)—on\\nwhich Valdez relies—is inapposite. Unlike the prisoner\\nthere, Valdez did not wait to file his second state habeas\\npetition until after a new decision; he filed it before.\\n\\n In the alternative, Valdez claims that his delay was\\nreasonable because of the size of the state-court record and\\ncomplexity of the case. This contention is likewise\\ninsufficient to establish good cause. Valdez timely filed his\\nfirst state habeas petition. That petition raised the same\\nclaims he then raised in his second state petition. He has\\n\\n 7\\n Valdez makes much of the fact that he filed a second round of\\nhabeas petitions based on Elizalde, asserting that doing so demonstrates\\nhis diligence. But that assertion does not explain why Valdez filed his\\nsecond state habeas petition before Elizalde was decided.\\n\\f12 VALDEZ V. MONTGOMERY\\n\\noffered no explanation for why he could timely file his first\\npetition but not his second. See Velasquez, 639 F.3d at 968\\n(“[E]ach of Velasquez’s habeas petitions is nearly identical\\nto the petition that came before it. It is not reasonable that\\nVelasquez’s counsel would need excess time essentially to\\nre-file an already-written brief.”).\\n\\n In summary, the district court correctly held that Valdez\\nis not entitled to statutory tolling.\\n\\n IV\\n\\n We next address Valdez’s contention that the district\\ncourt should not have dismissed his federal habeas petition\\nwithout requiring the State to respond and lodge the state-\\ncourt record. Valdez contends that the district court could\\nnot determine whether his state habeas petitions were timely\\nfiled absent the state-court record.\\n\\n A district court may summarily dismiss a federal habeas\\ncorpus petition sua sponte if “it plainly appears from the\\npetition and any attached exhibits that the petitioner is not\\nentitled to relief . . . .” Rule 4 of the Rules Governing Section\\n2254 Cases. A district court should do so, however, only\\nafter “provid[ing] the petitioner with adequate notice and an\\nopportunity to respond.” Herbst v. Cook, 260 F.3d 1039,\\n1043 (9th Cir. 2001). Moreover, because “federal habeas\\ncourts” have a duty to “independently [review] the basis for\\nthe state court’s decision,” a district court must “obtain and\\nreview the relevant portions of the state court record,” or\\nhold an evidentiary hearing, as necessary to discharge its\\nduty. Nasby v. McDaniel, 853 F.3d 1049, 1053 (9th Cir.\\n2017).\\n\\n Valdez, in his federal habeas petition, listed the date the\\nCalifornia Supreme Court denied his petition for review on\\n\\f VALDEZ V. MONTGOMERY 13\\n\\ndirect appeal, the dates he filed his state habeas petitions, and\\nthe dates the state courts denied those petitions. The district\\ncourt, after ordering Valdez to show cause and thereby\\nproviding Valdez “adequate notice and an opportunity to\\nrespond,” Herbst, 260 F.3d at 1043, had Valdez’s\\nexplanations for why he delayed in filing his second state\\nhabeas petition. Finally, the district court had the California\\nCourt of Appeal’s decision dismissing Valdez’s second state\\npetition because Valdez attached that decision as an exhibit\\nto his objections to the magistrate judge’s findings and\\nrecommendations. As demonstrated in Section III, the\\nforegoing information is all that is necessary to conclude that\\nValdez’s federal habeas petition was untimely. The district\\ncourt did not err by not ordering the State to respond and\\nlodge the state-court record. 8\\n\\n V\\n\\n Because we conclude that Valdez is not entitled to\\nstatutory tolling for the period following the California\\nSuperior Court’s denial of his first state habeas petition,\\nValdez’s federal habeas petition is untimely, and we affirm\\nthe district court’s dismissal of his petition.\\n\\n AFFIRMED.\\n\\n\\n\\n\\n 8\\n Valdez asks us to take judicial notice of various state-court\\ndocuments because they purportedly show why the district court needed\\nto order the State to lodge the state-court record. Because we can, and\\ndo, affirm the district court’s decision on the record before it, we DENY\\nValdez’s motion for judicial notice as MOOT.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1293,"cells":{"id":{"kind":"number","value":4603414,"string":"4,603,414"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Vht, Inc. v. Zillow Group, Inc."},"decision_date":{"kind":"string","value":"2019-03-15"},"docket_number":{"kind":"string","value":"17-35587"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: M. Margaret McKeown, William A. Fletcher, and Ronald M. Gould, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"McKEOWN, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nVHT, INC., a Delaware corporation, Nos. 17-35587\\n Plaintiff-Appellee/ 17-35588\\n Cross-Appellant,\\n\\n v. D.C. No.\\n 2:15-cv-01096-\\nZILLOW GROUP, INC., a Washington JLR\\ncorporation; ZILLOW, INC., a\\nWashington corporation,\\n Defendants-Appellants/ OPINION\\n Cross-Appellees.\\n\\n Appeal from the United States District Court\\n for the Western District of Washington\\n James L. Robart, Senior District Judge, Presiding\\n\\n Argued and Submitted August 28, 2018\\n Seattle, Washington\\n\\n Filed March 15, 2019\\n\\nBefore: M. Margaret McKeown, William A. Fletcher, and\\n Ronald M. Gould, Circuit Judges.\\n\\n Opinion by Judge McKeown\\n\\f2 VHT, INC. V. ZILLOW GROUP\\n\\n SUMMARY *\\n\\n\\n Copyright Law\\n\\n The panel affirmed in part and reversed in part the\\ndistrict court’s judgment after a jury trial and remanded in a\\ncopyright infringement action brought by VHT, Inc., a real\\nestate photography studio, against Zillow Group, Inc., and\\nZillow, Inc., an online real estate marketplace.\\n\\n VHT alleged that Zillow’s use of photos on the “Listing\\nPlatform” and “Digs” parts of its website exceeded the scope\\nof VHT’s licenses to brokers, agents, and listing services that\\nprovided those photos to Zillow. The district court granted\\npartial summary judgment on a limited set of claims. The\\njury found in favor of VHT on most remaining claims,\\nawarding over $8.27 million in damages. The district court\\npartially granted Zillow’s post-trial motion for judgment\\nnotwithstanding the verdict, reversing in part the jury verdict\\nand reducing total damages to approximately $4 million.\\n\\n The panel affirmed the district court’s summary\\njudgment in favor of Zillow on direct infringement of the\\nListing Platform photos. The panel held that VHT failed to\\nestablish that Zillow engaged in volitional conduct by\\nexercising control over the content of the Listing Platform.\\n\\n With respect to direct liability on the Digs photos, the\\npanel affirmed the district court’s grant in favor of Zillow of\\njudgment notwithstanding the verdict on 22,109 non-\\ndisplayed photos and 2,093 displayed but not searchable\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f VHT, INC. V. ZILLOW GROUP 3\\n\\nphotos. The panel held that VHT did not present substantial\\nevidence that Zillow, through the Digs platform, directly\\ninfringed its display, reproduction, or adaption rights.\\n\\n The panel upheld summary judgment in favor of VHT\\non 3,921 displayed, searchable Digs photos. The panel held\\nthat fair use did not absolve Zillow of liability because\\nZillow’s tagging of the photos for searchable functionality\\nwas not a transformative fair use.\\n\\n The panel affirmed the district court’s grant in favor of\\nZillow of judgment notwithstanding the verdict on\\nsecondary liability, both contributory and vicarious, on the\\nDigs photos.\\n\\n As to damages, the panel remanded consideration of the\\nissue whether VHT’s photos used on Digs were part of a\\ncompilation or were individual photos.\\n\\n The panel reversed the district court’s denial of judgment\\nnotwithstanding the verdict on the issue of willfulness and\\nvacated the jury’s finding on willfulness. The panel\\nconcluded that substantial evidence did not show that Zillow\\nwas “actually aware” of its infringing activity, nor that\\nZillow recklessly disregarded or willfully blinded itself to its\\ninfringement.\\n\\n\\n COUNSEL\\n\\nIan B. Crosby (argued), Edgar G. Sargent, Genevieve Vose\\nWallace, and Jenna G. Farleigh, Susman Godfrey LLP,\\nSeattle, Washington, for Defendants-Appellants/Cross-\\nAppellees.\\n\\f4 VHT, INC. V. ZILLOW GROUP\\n\\nStephen M. Rummage (argued), Marcia B. Paul, Jonathan\\nM. Lloyd, James E. Howard, and Max B. Hensley, Davis\\nWright Tremaine LLP, Seattle, Washington, for Plaintiff-\\nAppellee/Cross-Appellant.\\n\\nKeith Kupferschmid and Terry Hart, Copyright Alliance,\\nWashington, D.C.; Eleanor M. Lackman and Lindsay W.\\nBowen, Cowan, DeBaets, Abrahams & Sheppard LLP, New\\nYork, New York; for Amicus Curiae Copyright Alliance.\\n\\nAlicia Calzada, Alicia Wagner Calzada, PLLC, San\\nAntonio, Texas, for Amici Curiae The American Society of\\nMedia Photographers, Inc., Digital Media Licensing\\nAssociation, Inc., Graphic Artist Guild, Inc., and National\\nPress Photographers Association, Inc.\\n\\nThomas G. Hentoff and Chanakya A. Sethi, Williams &\\nConnolly LLP, Washington D. C., for Amici Curiae Disney\\nEnterprises, Inc., Twentieth Century Fox Film Corporation,\\nand Recording Industry Association of America, Inc.\\n\\nMitchell L. Stoltz, Electronic Frontier Foundation, San\\nFrancisco, California, for Amicus Curiae Electronic Frontier\\nFoundation.\\n\\nBrian M. Willen, Wilson Sonsini Goodrich & Rosati, P.C.,\\nNew York, New York; Ryan T. O’Hollaren, Wilson Sonsini\\nGoodrich & Rosati, P.C., Palo Alto, California; for Amici\\nCuriae Internet Association and Computer &\\nCommunications Industry Association.\\n\\f VHT, INC. V. ZILLOW GROUP 5\\n\\n OPINION\\n\\nMcKEOWN, Circuit Judge:\\n\\n Zillow, an online real estate marketplace, has become a\\npopular website for homeowners and others to check\\nestimated valuations of their property, look for houses and\\ncondominiums for sale and rent, and see photographs of a\\nwide range of properties. Thousands of those copyrighted\\nphotos come from VHT, the largest professional real estate\\nphotography studio in the country.\\n\\n The copyright claims on appeal concern Zillow’s use of\\nVHT’s photos on two parts of Zillow’s website: the “Listing\\nPlatform” and “Digs.” The Listing Platform is the core of\\nthe website, featuring photos and information about real\\nestate properties, both on and off the market. Zillow claims\\nthat the site includes “most homes in America.” Digs\\nfeatures photos of artfully-designed rooms in some of those\\nproperties and is geared toward home improvement and\\nremodeling. Zillow tags photos on the Listing Platform so\\nthat Digs users can search the database by various criteria,\\nlike room type, style, cost, and color.\\n\\n Real estate brokers, listing services, and agents hire VHT\\nto take professional photos of new listings for marketing\\npurposes. A VHT photographer takes the photos and sends\\nthem to the company’s studio for touch-up, where they are\\nsaved to VHT’s electronic photo database, and then\\ndelivered to the client for use under license. Each license\\nagreement between VHT and its clients differs slightly, but\\neach contract generally grants the requesting client the right\\nto use the photos in the sale or marketing of the featured\\nproperty. Zillow receives these photos and other data in\\nfeeds from various real estate-related sources.\\n\\f6 VHT, INC. V. ZILLOW GROUP\\n\\n In 2015, VHT sued Zillow Group, Inc., and Zillow, Inc.,\\n(collectively “Zillow”) for copyright infringement, alleging\\nthat Zillow’s use of photos on the Listing Platform and Digs\\nexceeded the scope of VHT’s licenses to brokers, agents, and\\nlisting services who provided those photos to Zillow. The\\ndistrict court granted partial summary judgment on a limited\\nset of claims, while other claims advanced to trial. The jury\\nfound in favor of VHT on most remaining claims, awarding\\nover $8.27 million in damages. The district court partially\\ngranted Zillow’s post-trial motion for judgment\\nnotwithstanding the verdict, reversing in part the jury verdict\\nand reducing total damages to approximately $4 million.\\n\\n The parties cross-appealed issues stemming from partial\\nsummary judgment, the jury verdict, and judgment\\nnotwithstanding the verdict. We affirm in part and reverse\\nin part. 1\\n\\n To simplify and make sense of the various claims, this\\nopinion does not split out the appeal and cross-appeal as was\\ndone in the briefing to the court. Instead, the opinion\\nseparately addresses liability for each of the categories of\\nphotos at issue, followed by a discussion of damages. In\\nview of the multiple theories of liability and categories of\\nphotos, following is an overview of the opinion.\\n\\n\\n\\n\\n 1\\n In connection with these proceedings, we received amicus curiae\\nbriefs from a broad array of interested parties, including nonprofit groups\\nand associations representing a diverse set of industry, technology, and\\nartistic interests. The briefs were helpful to our understanding of the\\nimplications of this case from various points of view. We thank amici\\nfor their participation.\\n\\f VHT, INC. V. ZILLOW GROUP 7\\n\\n I. Direct Infringement\\n A. Direct Infringement—Listing Platform Photos\\n B. Direct Infringement—Digs Photos\\n 1. Jury Verdict—Direct Infringement\\n 2. Summary Judgment—Fair Use re Searchable\\n Photos\\n a. Background on Fair Use\\n b. Evolution of Search Engine Cases\\n c. Application of Fair Use Principles\\n II. Secondary Infringement—Digs\\n A. Contributory Liability\\n B. Vicarious Liability\\n III. Damages\\n A. Compilation\\n B. Willfulness\\n IV. Conclusion\\n\\n ANALYSIS\\n\\n The heart of this dispute is Zillow’s copyright liability\\nfor use of VHT photos. VHT argues that Zillow directly\\ninfringed its copyrighted photos, both those on the Listing\\nPlatform and Digs. VHT also argues that Zillow indirectly\\ninfringed through use of the photos on Digs. These claims\\npertain to different images, focus on different features of\\nZillow’s website, and have different procedural postures, so\\nwe consider the various categories of photos separately.\\n\\n I. Direct Infringement\\n\\n VHT’s key claim is that Zillow is directly liable for\\ninfringing VHT’s copyright on photos that were posted on\\nthe Listing Platform and Digs. To prevail on a claim of\\ndirect copyright infringement, VHT must establish\\n“ownership of the allegedly infringed material” and that\\nZillow “violate[d] at least one exclusive right granted to”\\n\\f8 VHT, INC. V. ZILLOW GROUP\\n\\nVHT under 17 U.S.C. § 106. A&M Records, Inc. v. Napster,\\nInc., 239 F.3d 1004, 1013 (9th Cir. 2001). It is undisputed\\nthat VHT is the copyright holder of the allegedly infringed\\nphotos and therefore has the exclusive right to reproduce,\\nadapt, and display them. 2 17 U.S.C. § 106.\\n\\n VHT must also establish causation, which is commonly\\nreferred to as the “volitional-conduct requirement.” See\\nPerfect 10, Inc. v. Giganews, Inc., 847 F.3d 657, 666 (9th\\nCir. 2017). As we set out in Giganews—decided on the first\\nday of the VHT/Zillow trial and the closest circuit precedent\\non point—“volition in this context does not really mean an\\nact of willing or choosing or an act of deciding”; rather, “it\\nsimply stands for the unremarkable proposition that\\nproximate causation historically underlines copyright\\ninfringement liability no less than other torts.” Id. (internal\\ncitations omitted). Stated differently, “direct liability must\\nbe premised on conduct that can reasonably be described as\\nthe direct cause of the infringement.” Id. (citation omitted).\\nThis prerequisite takes on greater importance in cases\\ninvolving automated systems, like the Zillow website.\\n\\n In addressing this concept, Justice Scalia noted that\\n“[e]very Court of Appeals to have considered an automated-\\nservice provider’s direct liability for copyright infringement\\nhas adopted [the volitional-conduct] rule.” Am. Broad. Cos.,\\nInc. v. Aereo, Inc., 573 U.S. 431, 453 (2014) (Scalia, J.,\\ndissenting). 3 He went on to explain that while “most direct-\\n\\n 2\\n VHT does not appeal the district court’s finding that there was\\ninsufficient evidence that Zillow violated VHT’s distribution rights.\\n 3\\n Although the majority opinion in Aereo does not reference the\\nvolitional-conduct requirement, Justice Scalia’s dissent offers instructive\\nbackground on the doctrine. In Giganews, we embraced the principle\\n\\f VHT, INC. V. ZILLOW GROUP 9\\n\\ninfringement cases” do not present this issue, “it comes right\\nto the fore when a direct-infringement claim is lodged\\nagainst a defendant who does nothing more than operate an\\nautomated, user-controlled system. . . . Most of the time that\\nissue will come down to who selects the copyrighted\\ncontent: the defendant or its customers.” Id. at 454–55\\n(internal citations omitted).\\n\\n Giganews, Aereo, and out-of-circuit precedent counsel\\nthat direct copyright liability for website owners arises when\\nthey are actively involved in the infringement. “‘[T]he\\ndistinction between active and passive participation’” in the\\nalleged infringement is “‘central’” to the legal analysis.\\nGiganews, 847 F.3d at 667 (quoting Fox Broad. Co. v. Dish\\nNetwork LLC, 160 F. Supp. 3d 1139, 1160 (C.D. Cal. 2015)).\\n\\n That “direct” infringement requires “active”\\ninvolvement is hardly surprising, given the correlation\\nbetween the words “active” and “direct.” As the Fourth\\nCircuit held, “[t]here must be actual infringing conduct with\\na nexus sufficiently close and causal to the illegal copying\\nthat one could conclude that the machine owner himself\\ntrespassed on the exclusive domain of the copyright owner.”\\nCoStar Grp., Inc. v. LoopNet, Inc., 373 F.3d 544, 550 (4th\\nCir. 2004). By contrast, activities that fall on the other side\\nof the line, such as “‘automatic copying, storage, and\\ntransmission of copyrighted materials, when instigated by\\nothers, do[] not render an [Internet service provider] strictly\\nliable for copyright infringement[.]’” Giganews, 847 F.3d at\\n670 (quoting CoStar, 373 F.3d at 555).\\n\\n\\n\\nand held that it is “consistent with the Aereo majority opinion,” which\\nleft the requirement “intact.” 847 F.3d at 666–67.\\n\\f10 VHT, INC. V. ZILLOW GROUP\\n\\n In other words, to demonstrate volitional conduct, a party\\nlike VHT must provide some “evidence showing [the alleged\\ninfringer] exercised control (other than by general operation\\nof [its website]); selected any material for upload, download,\\ntransmission, or storage; or instigated any copying, storage,\\nor distribution” of its photos. Id. at 666, 670. VHT failed to\\nsatisfy that burden with respect to either the photos on the\\nListing Platform or on Digs.\\n\\n A. Direct Infringement—Listing Platform\\n Photos\\n\\n VHT asserted that Zillow directly infringed the photos\\ndisplayed on the Listing Platform after a real estate property\\nwas sold because VHT’s license agreements only authorized\\nuse of those photos in relation to the sale of the property.\\nThis claim, involving 54,257 non-searchable photos, was\\nresolved on summary judgment. The Listing Platform is the\\ncore of Zillow’s online real estate marketplace. It features\\nphotos and information about properties, which Zillow\\nreceives through digital feeds from real estate agents,\\nbrokerages, and multiple listing services, among others\\n(collectively “feed providers”).\\n\\n Zillow has agreements with its feed providers granting it\\nan express license to use, copy, distribute, publicly display,\\nand create derivative works from the feed data on its\\nwebsites. Feed providers represent that they “ha[ve] all\\nnecessary rights and authority to enter into” the agreements,\\nand that “Zillow’s exercise of the rights granted [t]hereunder\\nwill not violate the intellectual property rights, or any other\\nrights of any third party.”\\n\\n These agreements provide Zillow with either\\n“evergreen” or “deciduous” rights in the photos provided\\nthrough the feeds. An evergreen right permits use of a photo\\n\\f VHT, INC. V. ZILLOW GROUP 11\\n\\nwithout any time restriction, “on and in connection with the\\noperation, marketing and promotion of the web sites and\\nother properties, owned, operated or powered by Zillow or\\nits authorized licensees.” By contrast, a deciduous right is\\ntemporally limited: Zillow may use the photo when the real-\\nestate listing for its corresponding property is active, but\\nonce the listing is removed (for example, when the property\\nsells), the photo must be taken down from Zillow’s websites.\\nTo treat each photo consistently with its deciduous or\\nevergreen designation, Zillow developed automated\\n“trumping” rules to determine which photos to display on the\\nListing Platform.\\n\\n VHT argues that Zillow “designed its system to . . .\\ncause[] the reproduction, display, and adaptation of VHT\\nphotographs post-sale on the Listing Platform,” and “chose\\nto simply ignore VHT’s notices that post-sale use was\\nbeyond the scope of VHT’s licenses.” The district court\\ngranted summary judgment to Zillow, concluding that it did\\nnot engage in volitional conduct and therefore did not\\ndirectly infringe VHT’s copyrights in 54,257 photos by\\ndisplaying them on the Listing Platform after a real estate\\nproperty was sold.\\n\\n On de novo review, we agree with the district court’s\\nanalysis and affirm. Kelly v. Arriba Soft Corp., 336 F.3d\\n811, 817 (9th Cir. 2003). Although the district court did not\\nhave the benefit of Giganews at the time of summary\\njudgment, its careful reasoning was prescient in invoking the\\nsame principles.\\n\\n Zillow did not engage in volitional conduct necessary to\\nsupport a finding of direct liability. The content of the\\nListing Platform is populated with data submitted by third-\\nparty sources that attested to the permissible use of that data,\\n\\f12 VHT, INC. V. ZILLOW GROUP\\n\\nand Zillow’s system for managing photos on the Listing\\nPlatform was constructed in a copyright-protective way.\\nThe feed providers themselves select and upload every\\nphoto, along with the evergreen or deciduous designations,\\nthat wind up on the Listing Platform. As a result, the photos\\non the Listing Platform were not “selected” by Zillow. See\\nGiganews, 847 F.3d at 670. Nor did Zillow “exercise[]\\ncontrol” over these photos beyond the “general operation of\\n[its website].” Id. Zillow required feed providers to certify\\nthe extent of their rights to use each photo. Consistent with\\nthese designations, Zillow’s system classified each photo as\\ndeciduous or evergreen and programmed its automated\\nsystems to treat each photo consistently with that scope of\\nuse certified to by the third party.\\n\\n Further, when multiple versions of the same photo were\\nsubmitted through the various feeds, Zillow invoked its\\ncopyright-protective “trumping” rules. For example, one\\nrule might prefer a photo provided by an agent over one\\nprovided by a multiple listing service, and another might\\nprefer a local broker to an international one. Zillow used a\\nrule that gave preference to photos with evergreen rights\\nover photos with deciduous rights in the same image. As the\\ndistrict court recognized, “trumping” is a reasonable way to\\ndesign a system to manage multiple versions of the same\\nphoto when the authorized use varies across versions. These\\nrules, along with other features of the system, facilitate\\nkeeping the photos with evergreen rights on the website and\\nremoving the photos with deciduous rights once a property\\nhas sold. Thus, Zillow actively designed its system to avoid\\nand eliminate copyright infringement.\\n\\n Notably, VHT’s argument is primarily cast in terms of\\nZillow facilitating or enabling infringement by VHT’s\\nclients that are Zillow’s feed providers. But this type of\\n\\f VHT, INC. V. ZILLOW GROUP 13\\n\\nclaim more properly falls in the category of secondary\\ninfringement, a claim not advanced by VHT with respect to\\nthe Listing Platform photos.\\n\\n VHT also asserts that Zillow failed to remove photos\\nonce it received notice that infringing content was on the\\nListing Platform, a conscious choice that amounts to\\nvolitional conduct on Zillow’s part. This claim is unavailing\\nbecause, once VHT put Zillow on notice of claimed\\ninfringement, Zillow took affirmative action to address the\\nclaims. Additionally, VHT’s assertion that it “repeatedly\\nnotified Zillow that it was infringing” is unsupported in the\\nrecord.\\n\\n In July 2014, VHT sent Zillow a takedown notice letter\\nwith a list of thousands of allegedly infringing photos by\\nresidential street address (but not by web address). Zillow\\npromptly requested all executed license agreements between\\nVHT and the feed providers who had provided photos to\\nZillow, as well as license agreements between VHT and its\\nphotographers, so that Zillow could evaluate whether VHT\\npossessed exclusive rights to the photos on the Listing\\nPlatform. VHT responded with an unsigned form contract,\\nwhich it stated was used with many feed providers, but\\nwhich was not tied to any specific photos on Zillow’s\\nwebsite. Zillow again reiterated its need to see the specific\\ncontracts governing the contested photos. Instead of\\nresponding with the contracts, VHT filed suit. Zillow’s\\nreasonable response to VHT’s single formal inquiry\\n(supplemented in a follow-on email) can hardly be\\ncharacterized as rising to the level of volitional conduct or\\nturning a blind eye.\\n\\n In sum, VHT failed to “provide[] . . . evidence showing\\n[Zillow] exercised control (other than by general operation\\n\\f14 VHT, INC. V. ZILLOW GROUP\\n\\nof [its website]); selected any material for upload, download,\\ntransmission, or storage; or instigated any copying, storage,\\nor distribution” of these photos. See Giganews, 847 F.3d at\\n670; see also CoStar, 373 F.3d at 555. Thus, we affirm the\\ndistrict court and conclude Zillow did not directly infringe\\nVHT’s copyrights in photos displayed on the Listing\\nPlatform post-sale.\\n\\n B. Direct Infringement—Digs Photos\\n\\n VHT also claimed that Zillow directly infringed\\nthousands of photos used on Digs. The jury concluded that\\nZillow directly infringed 28,125 photos and rejected its fair\\nuse defense. Following trial, the district court granted in\\nsubstantial part Zillow’s motion for judgment\\nnotwithstanding the verdict on the ground that insufficient\\nevidence supported Zillow’s direct infringement of 22,109\\nphotos that were not displayed on Digs and 2,093 4 photos\\nthat were displayed but not searchable on Digs.\\n\\n By contrast, the court upheld the jury’s determination\\nthat Zillow directly infringed a set of 3,921 images that were\\nselected and tagged by Zillow moderators for searchable\\nfunctionality and displayed on Digs. Zillow does not appeal\\nthis ruling. However, Zillow argues that fair use insulates it\\nfrom liability as to this subset of photos. The jury was\\ninstructed not to consider this legal theory as to these photos\\nbecause the district court had determined pretrial that, as a\\nmatter of law, the searchability function did not constitute\\nfair use. It is that summary judgment ruling that Zillow\\nchallenges on appeal. Because we agree with the district\\n\\n 4\\n The displayed but non-searchable set includes 2,094 photos. The\\ndistrict court affirmed the jury verdict with respect to one of those\\nphotos, which Zillow also distributed via email.\\n\\f VHT, INC. V. ZILLOW GROUP 15\\n\\ncourt that the fair use defense does not absolve Zillow of\\ndirect liability for these searchable photos, this portion of the\\njury verdict remains intact.\\n\\n The following chart clarifies the status of the Digs photos\\nrelevant to the direct infringement claims.\\n\\n Photos Jury Verdict Post-Trial Party\\n Determination Bringing\\n Appeal\\n22,109 not Direct Overturned jury VHT\\ndisplayed 5 infringement verdict\\n2,094 Direct Overturned jury VHT\\ndisplayed, not infringement verdict\\nsearchable (except 1 email\\n photo)\\n3,921 Direct Upheld jury Zillow:\\ndisplayed, infringement verdict direct\\nsearchable infringement\\n not appealed;\\n appeals only\\n summary\\n judgment on\\n no fair use\\n1 blog post Direct Upheld jury Not appealed\\nphoto (not on infringement verdict\\nDigs)\\n\\n\\n\\n\\n 5\\n Searchable photos numbered 1,694; 20,415 photos were not\\nsearchable.\\n\\f16 VHT, INC. V. ZILLOW GROUP\\n\\n 1. Jury Verdict—Direct Infringement\\n\\n VHT’s claim that Zillow directly infringed photos on\\nDigs went to the jury. The jury verdict form was framed in\\ngeneral terms, asking only whether “VHT has proven its\\ndirect copyright infringement claim as to one or more of the\\nVHT Photos[.]” The jury answered “yes,” and was asked to\\nspecify how many VHT photos were directly infringed. The\\njury answered “28,125”—in other words, all of them.\\nHowever, the jury was not asked to specify which copyright\\nrights—display, reproduction, or adaption—were\\n 6\\ninfringed. Following trial, Zillow moved for judgment\\nnotwithstanding the verdict or for a new trial. The task fell\\nto the district court to examine the evidence as to each right.\\n\\n On de novo review, “we apply the same standard used\\nby the district court in evaluating the jury’s verdict” and\\nuphold the verdict unless “the evidence permits only one\\nreasonable conclusion, and that conclusion is contrary to the\\njury’s verdict.” Wallace v. City of San Diego, 479 F.3d 616,\\n624 (9th Cir. 2007); Fed. R. Civ. P. 50(a). Specifically, we\\n“ask[] whether the verdict is supported by substantial\\nevidence,” “which is evidence adequate to support the jury’s\\nconclusion, even if it is also possible to draw a contrary\\nconclusion.” Unicolors, Inc. v. Urban Outfitters, Inc., 853\\nF.3d 980, 984, 991 (9th Cir. 2017) (quotation and citation\\nomitted). Given the sanctity of the jury process, we\\nundertake this review with special care and reluctance to\\noverturn a verdict. However, because the verdict here did\\nnot meet this standard, we affirm the district court’s grant of\\nZillow’s motion for judgment notwithstanding the verdict\\n\\n\\n 6\\n As noted earlier, the distribution right was not contested and is not\\nan issue on appeal.\\n\\f VHT, INC. V. ZILLOW GROUP 17\\n\\nwith respect to direct infringement of 22,109 non-displayed\\nphotos and 2,093 displayed but not searchable photos.\\n\\n We first consider display rights. As background for our\\nanalysis, it is useful to consider the direct infringement by\\nZillow that the district court upheld and that Zillow did not\\nappeal. The district court found substantial evidence that\\nZillow directly infringed VHT’s display rights in 3,921\\nphotos displayed on Digs that Zillow moderators selected\\nand tagged for searchable functionality. Based on testimony,\\ncharts, and statistics, the court found that “the jury could\\nhave reasonably concluded that users accessed those images\\nthrough Digs’s search function.” The court went on to\\nreason that “the jury could have reasonably concluded that\\nZillow’s moderation efforts, which rendered those images\\nsearchable, proximately caused the copying.” Put\\ndifferently, active conduct by Zillow met the volitional-\\nconduct requirement for direct infringement. Zillow does\\nnot appeal this ruling upholding the jury’s verdict as to the\\n3,921 displayed, searchable photos.\\n\\n On appeal, VHT attempts to shoehorn an additional\\n1,694 photos into this category. This effort falls flat both as\\na factual and legal matter because substantial evidence does\\nnot support direct infringement of VHT’s display rights in\\nthe 1,694 searchable images that were not displayed.\\n\\n VHT posits that the jury could have found that these\\nphotos were displayed because of circumstantial evidence\\nand because Zillow failed to record whether they were\\ndisplayed. Not so. This argument is foreclosed by the\\nparties’ stipulated spreadsheet that categorized each photo.\\nThe column labeled “DISPLAYED” included an entry for\\n“Y” (yes) or “N” (no). The jury was instructed to “treat\\nevery fact on this spreadsheet as proven,” so VHT cannot\\n\\f18 VHT, INC. V. ZILLOW GROUP\\n\\nrecast the facts retroactively and now claim that 1,694\\nsearchable images stipulated as “N[OT] DISPLAYED”\\nwere in fact displayed or made available for display. Up is\\nnot down.\\n\\n VHT’s contention that the jury could have reasonably\\ninferred that Zillow made “available for public display” all\\n22,109 “N[OT] DISPLAYED” images similarly fails. This\\ntheory presumes that the Copyright Act’s display right\\nencompasses an exclusive right to “make available for\\ndisplay,” a position neither supported by the statute nor\\nembraced by this court. See Perfect 10, Inc. v. Amazon.com,\\nInc., 508 F.3d 1146, 1160 (9th Cir. 2007) (“[B]ased on the\\nplain language of the statute, a person displays a\\nphotographic image by using a computer to fill a computer\\nscreen with a copy of the photographic image fixed in the\\ncomputer’s memory.”); 17 U.S.C. § 101 (“To ‘display’ a\\nwork means to show a copy of it, either directly or by means\\nof a film, slide, television image, or any other device or\\nprocess or, in the case of a motion picture or other\\naudiovisual work, to show individual images\\nnonsequentially.”). To be sure, the Copyright Office notes\\nthat the outer limits of the public display right have yet to be\\ndefined. U.S. COPYRIGHT OFFICE, THE MAKING AVAILABLE\\nRIGHT IN THE UNITED STATES 47–51 (Feb. 2016),\\nhttps://www.copyright.gov/docs/making_available/making-\\navailable-right.pdf.\\n\\n What is most important here, though, is that VHT’s\\nargument comes too late. The jury was never instructed on\\nthe “made available” theory, nor did VHT raise this issue in\\nits proposed jury instructions or in objections to the final\\ninstructions. See Sinclair v. Long Island R.R., 985 F.2d 74,\\n78 (2d Cir. 1993) (“the verdict . . . cannot be sustained on a\\ntheory that was never presented to the jury”); Ramona Equip.\\n\\f VHT, INC. V. ZILLOW GROUP 19\\n\\nRental, Inc. ex rel. U.S. v. Carolina Cas. Ins. Co., 755 F.3d\\n1063, 1070 (9th Cir. 2014) (an argument raised for first time\\nin a post-trial motion is waived). For these reasons,\\nsubstantial evidence did not support a finding of direct\\ninfringement of VHT’s public display rights in these 22,109\\nphotos.\\n\\n Zillow also did not violate VHT’s display rights in 2,093\\ndisplayed, non-searchable photos. These are photos that\\nDigs users copied to “personal boards” and “Implicit Digs,”\\nbut which Zillow did not add to the searchable set.\\n\\n A “personal board” is a bespoke digital bulletin board of\\nimages that a user saves or uploads from the Listing\\nPlatform. Users can also upload their own images. These\\nboards are typically private, though users can share a link to\\ntheir boards. When a user saves a copy of an image with\\nevergreen rights to a personal board, that image\\nautomatically joins a queue for review by a Zillow\\nmoderator. The moderator then decides whether to\\ndesignate the photo for tagging so that it can be searchable\\non Digs, and thus select it for public display. Not all photos\\nthat are in the queue for moderation are reviewed.\\nAdditionally, in some instances, a user starts but does not\\nfinish the process of saving an image to this board.\\nBeginning in 2014, Zillow programmed these “clicked to\\nsave” images—called “Implicit Digs”—to enter the queue\\nfor moderator review in the same manner as if the photo had\\nbeen saved to a personal board.\\n\\n According to VHT, the jury could have found that Zillow\\n“caused the [2,093] images to be displayed . . . by subjecting\\nthe non-searchable VHT Photos to the potential for\\nmoderation.” This argument defies logic because the only\\ndisplay that occurred was triggered by the user. Any\\n\\f20 VHT, INC. V. ZILLOW GROUP\\n\\npotential for future display is purely speculative. As the\\ndistrict court explained, the possibility that images might be\\nmoderated and tagged—conduct that is volitional—is not\\nsufficient “to transform Zillow from a ‘passive host’ to a\\n‘direct[] cause’ of the display of VHT’s images.”\\n\\n Next, we consider VHT’s exclusive reproduction and\\nadaption rights in the 2,093 displayed, non-searchable\\nphotos. The Copyright Act grants copyright holders the\\nexclusive right “to reproduce the copyrighted work in copies\\nor phonorecords.” 17 U.S.C. § 106(1). Direct infringement\\nof the reproduction right “requires copying by the defendant,\\n. . . which [requires] that the defendant cause the copying.”\\nFox Broad. Co., Inc. v. Dish Network LLC, 747 F.3d 1060,\\n1067 (9th Cir. 2014) (quotation and citation omitted). The\\nadaptation right is the exclusive right “to prepare derivative\\nworks based upon the copyrighted work.” 17 U.S.C.\\n§ 106(2). Following Giganews, we conclude that Zillow’s\\nautomated processes storing or caching VHT’s photos are\\ninsufficiently volitional to establish that Zillow directly\\ninfringed VHT’s reproduction and adaption rights in these\\nnon-searchable photos. Unlike photos that Zillow curated,\\nselected, and tagged for searchable functionality—activities\\nthat amount to volitional conduct establishing direct\\nliability—these 2,093 photos were copied to “personal\\nboards” and “Implicit Digs” based on user actions, not the\\nconduct of Zillow or its moderators.\\n\\n Any volitional conduct with respect to these photos was\\ntaken by the users, not Zillow. Users, not Zillow,\\n“selecte[d]” images to add to their personal boards and\\n“instigate[d]” the automatic caching process by saving a\\nparticular image. See Giganews, 847 F.3d at 670. This\\narrangement is important because courts have found no\\ndirect liability where an online system “responds\\n\\f VHT, INC. V. ZILLOW GROUP 21\\n\\nautomatically to users’ input . . . without intervening\\nconduct” by the website owner. CoStar, 373 F.3d at 550;\\nsee also Giganews, 847 F.3d at 670. Under these conditions,\\ncourts have analogized online facilities, like Internet service\\nproviders, to a copy machine owner, who is not liable\\n“[w]hen a customer duplicates an infringing work.” CoStar,\\n373 F.3d at 550.\\n\\n Additionally, Zillow’s behind-the-scenes technical work\\non Digs photos is not evidence that Zillow “selected any\\nmaterial for upload, download, transmission, or storage.”\\nGiganews, 847 F.3d at 670. Zillow produced cached copies\\nof these Digs images, a process that automatically trims or\\npads images whose height and width did not match the target\\nresolution, for the purposes of accelerating website speed.\\nThis activity does not amount to volitional conduct. Nor can\\nZillow’s promotion of Digs, including encouraging users to\\nshare photos through its site, be seen as “instigat[ing]” user\\ncopying. Id.\\n\\n Zillow’s conduct with respect to these photos amounts\\nto, at most, passive participation in the alleged infringement\\nof reproduction and adaption rights and is not sufficient to\\ncross the volitional-conduct line. As in cases involving\\nInternet service providers, Zillow “affords its [users] an\\nInternet-based facility on which to post materials, but the\\nmaterials posted are of a type and kind selected by the [user]\\nand at a time initiated by the [user].” CoStar, 373 F.3d at\\n555. Zillow did not directly infringe VHT’s reproduction\\n\\f22 VHT, INC. V. ZILLOW GROUP\\n\\nand adaptation rights in the 2,093 displayed, non-searchable\\nphotos. 7\\n\\n * * *\\n\\n In sum, VHT did not present substantial evidence that\\nZillow, through the Digs platform, directly infringed its\\ndisplay, reproduction, or adaption rights in 22,109 not\\ndisplayed photos and 2,093 displayed but non-searchable\\nphotos. We affirm the district court’s grant of judgment\\nnotwithstanding the verdict as to these photos. We now turn\\nto Zillow’s fair use defense to direct infringement of 3,921\\ndisplayed, searchable photos.\\n\\n 2. Summary Judgment—Fair Use re\\n Searchable Photos\\n\\n Zillow does not appeal the jury’s finding of direct\\ninfringement with respect to the 3,921 displayed, searchable\\nphotos, but does assert a fair use defense for those photos.\\nZillow contends that Digs’ searchable functionality\\nconstitutes fair use, which the district court rejected as a\\nmatter of law at summary judgment.\\n\\n We recount the somewhat unusual history of the fair use\\nissue in the proceedings below. On summary judgment, the\\ndistrict court rejected Zillow’s argument that “the images\\nthat it has made searchable on Digs” are protected by fair use\\nand instead “conclude[d] as a matter of law that Digs’\\nsearchable functionality does not constitute a fair use.” At\\ntrial, the jury was generally instructed to consider the fair use\\n\\n 7\\n The same analysis applies to any potential violation of VHT’s\\nexclusive reproduction and adaption rights in the 22,109 photos that were\\nnot displayed.\\n\\f VHT, INC. V. ZILLOW GROUP 23\\n\\ndefense as to all VHT photos used on Digs that it found\\nZillow had directly or indirectly infringed. However, this set\\nof photos was carved out for separate treatment. The jury\\nconsidered only whether “reproduction, cropping, and\\nscaling” of these photos constituted fair use because the\\ncourt instructed the jury that the court “ha[d] determined,\\nand you are to take as proven, that the Digs searchable\\nfunctionality does not constitute fair use.” After finding that\\nZillow directly infringed all 28,125 VHT photos used on\\nDigs, the jury rejected Zillow’s fair use affirmative defense\\nfor all photos.\\n\\n The district court upheld the jury’s fair use verdict,\\nwhich Zillow does not appeal. Rather, Zillow’s appeal\\nreaches back to the district court’s summary judgment ruling\\nto argue that the Digs’ searchable functionality is fair use as\\na matter of law, and, as a result, Zillow bears no liability for\\nthe 3,921 searchable and displayed photos. We review de\\nnovo the district court’s grant of summary judgment to VHT\\non this mixed question of law and fact. Kelly, 336 F.3d at\\n817.\\n\\n a. Background on Fair Use\\n\\n Protection of copyrighted works is not absolute. “The\\nfair use defense permits the use of copyrighted works\\nwithout the copyright owner’s consent under certain\\nsituations.” Amazon, 508 F.3d at 1163. Fair use both fosters\\ninnovation and encourages iteration on others’ ideas, “thus\\nproviding a necessary counterbalance to the copyright law’s\\ngoal of protecting creators’ work product.” Id.; see\\nCampbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 575–76\\n(1994). Fair use also aligns with copyright’s larger purpose\\n“‘[t]o promote the Progress of Science and useful Arts,’ . . .\\nand to serve ‘the welfare of the public.’” Amazon, 508 F.3d\\n\\f24 VHT, INC. V. ZILLOW GROUP\\n\\nat 1163 (quoting U.S. Const. art. I, § 8, cl. 8, and Sony Corp.\\nof Am. v. Universal City Studios, Inc., 464 U.S. 417, 429 n.10\\n(1984)).\\n\\n Fittingly enough, a case involving a biography of George\\nWashington serves as a foundational source of fair use in\\nAmerican law. Folsom v. Marsh, 9 F. Cas. 342 (D. Mass.\\n1841) (No. 4901). Justice Story’s narrative description of\\ncopyright doctrine in that case “distilled the essence of law\\nand methodology from the earlier cases” and provided the\\nconceptual basis for the judge-made fair use doctrine.\\nCampbell, 510 U.S. at 576. Although the 1976 Copyright\\nAct codified those principles, it did little to elaborate on\\nJustice Story’s description or to clarify application of the\\nfactors. With minimal guidance or elucidation, Congress set\\nforth four factors for courts to consider when determining\\nwhether the use of a copyrighted work is a “fair use”:\\n\\n (1) the purpose and character of the use,\\n including whether such use is of a\\n commercial nature or is for nonprofit\\n educational purposes;\\n\\n (2) the nature of the copyrighted work;\\n\\n (3) the amount and substantiality of the\\n portion used in relation to the copyrighted\\n work as a whole; and\\n\\n (4) the effect of the use upon the potential\\n market for or value of the copyrighted\\n work.\\n\\n17 U.S.C. § 107.\\n\\f VHT, INC. V. ZILLOW GROUP 25\\n\\n Given license to apply these factors flexibly and to\\nconsider them in their totality, courts have been bedeviled by\\nthe fair use inquiry. See Campbell, 510 U.S. at 577–78. Fair\\nuse has been called “the most troublesome [doctrine] in the\\nwhole law of copyright” and commentators have criticized\\nthe factors as “billowing white goo.” Monge v. Maya\\nMagazines, Inc., 688 F.3d 1164, 1170–71 (9th Cir. 2012)\\n(citations and quotations omitted). In a (still ongoing) effort\\nto adapt the fair use analysis to a myriad of circumstances,\\ncourts have embellished and supplemented the factors. For\\nexample, the concept of “transformativeness” is found\\nnowhere in the statute, but appeared for the first time in the\\nSupreme Court in Campbell, where the Court endeavored to\\nrefine and crystalize the first statutory factor: the “purpose\\nand character of the use.” 8 510 U.S. at 579. The animating\\npurpose of the first factor is to determine,\\n\\n in Justice Story’s words, whether the new\\n work merely ‘supersede[s] the objects’ of the\\n original creation . . . or instead adds\\n something new, with a further purpose or\\n different character, altering the first with new\\n expression, meaning, or message; it asks, in\\n other words, whether and to what extent the\\n new work is “transformative.”\\n\\n\\n\\n 8\\n The concept of transformative use had appeared in earlier lower\\ncourt decisions, e.g., Twin Peaks Prods., Inc. v. Publ’ns Intern., Ltd., 996\\nF.2d 1366, 1375 (2d Cir. 1993) (“[I]t is not uncommon for works serving\\na fair use purpose to give at least a brief indication of the plot. . . . In\\nidentifying plot, the author of the second work may or may not be said\\nto have made . . . a ‘transformative’ use. . . . Such use would occur, for\\nexample, if a plot was briefly described for purposes of adding\\nsignificant criticism or comment about the author’s plotting technique.”).\\n\\f26 VHT, INC. V. ZILLOW GROUP\\n\\nId. (citations omitted). Because transformation advances\\ncopyright’s core goals, “the more transformative the new\\nwork, the less will be the significance of other factors.” Id.\\nLikewise, despite the absence of a textual hook, public\\npurpose also has been read into the statute. See Amazon, 508\\nF.3d at 1166; Kelly, 336 F.3d at 820. While we can discern\\ncertain animating principles bridging cases in this area, the\\ndoctrine has hardly followed a straight, or even slightly\\ncurved, line.\\n\\n The focus of the parties’ debate here is whether Zillow’s\\ntagging of 3,921 VHT photos for searchable functionality on\\nDigs was transformative and thus supported a finding of fair\\nuse. The purpose of Digs is to permit users to search for\\ncertain attributes or features, such as a marble countertop or\\nhardwood floor, and view photos of rooms with those\\nattributes or features. These photos are either uploaded by\\nusers to Digs, or selected manually or electronically by\\nZillow. Zillow then tags the photos to make them\\nsearchable. Of course, tagging makes it possible for a user’s\\nkeyword search to produce relevant results. Zillow refers to\\nthese tagged photos as “searchable images” or components\\nof the “searchable set.” VHT’s 3,921 photos are in the\\nsearchable set.\\n\\n Zillow contends that Digs is effectively a search engine,\\nwhich makes its use of VHT’s photos transformative, and\\ntherefore fair use. VHT responds that this is not fair use\\nbecause Digs is not a search engine and the tagging for\\nsearchable functionality is not transformative. Dueling\\n“search engine” characterizations do not resolve fair use\\nhere. Instead, we step back and assess the question\\nholistically, as we have been instructed to do by the statute\\nand the Supreme Court. We consider the reality of what is\\n\\f VHT, INC. V. ZILLOW GROUP 27\\n\\nhappening rather than resorting to labels. To do that, it is\\nhelpful to recount the history of the search engine cases.\\n\\n b. Evolution of Search Engine Cases\\n\\n Over the past two decades, search engines have emerged\\nas a significant technology that may qualify as a\\ntransformative fair use, making images and information that\\nwould otherwise be protected by copyright searchable on the\\nweb. See Amazon, 508 F.3d at 1166–67; Kelly, 336 F.3d at\\n818–22. In assessing fair use in the context of search\\nengines, courts have relied heavily on the first fair use factor,\\nand in particular “whether and to what extent the new work\\nis transformative.” Campbell, 510 U.S. at 579 (citation and\\nquotation omitted); see also Amazon, 508 F.3d at 1164\\n(explaining the Kelly court relied “primarily” on the first fair\\nuse factor when conducting its analysis); Authors Guild v.\\nGoogle, Inc., 804 F.3d 202, 220–221, 223 (2d Cir. 2015)\\n(offering a relatively abbreviated consideration of the\\nremaining three fair use factors, all of which were informed\\nby its analysis of the first factor).\\n\\n In an early opinion applying fair use principles in the\\ndigital age, we held that the now-defunct search engine\\nArriba’s creation and use of thumbnail versions of a\\nprofessional photographer’s copyrighted images was fair use\\nbecause the “smaller, lower-resolution images . . . served an\\nentirely different function than [the] original images.” Kelly,\\n336 F.3d at 815, 818. The original images served an artistic\\nor aesthetic purpose. Id. at 819. By contrast, the thumbnail\\nimages, which were provided in response to a user’s search\\nquery, were incorporated into the search engine’s overall\\nfunction “to help index and improve access to images on the\\ninternet and their related web sites.” Id. at 818. Investing\\nthe images with a new purpose made Arriba’s use\\n\\f28 VHT, INC. V. ZILLOW GROUP\\n\\ntransformative, not superseding. Indeed, the thumbnail\\nversions could not supersede the original use because the\\nthumbnails were grainy and low-resolution when enlarged.\\nId. Additionally, Arriba’s use of the thumbnail images\\n“promote[d] the goals of the Copyright Act and the fair use\\nexception” because they “benefit[ed] the public by\\nenhancing information-gathering techniques on the\\ninternet.” Id. at 820. Just as Campbell had drawn out the\\nprinciple of transformation from the first statutory factor, we\\ndrew out the principle of public benefit.\\n\\n Building on our reasoning in Kelly, in Amazon we held\\nthat Google’s use of thumbnail images in its search engine\\nis “highly transformative” and thus fair use. 508 F.3d at\\n1163–65. As in Kelly, we concluded that “a search engine\\nprovides social benefit by incorporating an original work\\ninto a new work, namely, an electronic reference tool.” Id.\\nat 1165. On a scale much greater than the search engine at\\nissue in Kelly, Google “improve[s] access to images on the\\ninternet and their related web sites” by “index[ing]” the\\ninternet and linking to the original source image generated\\nin the search results. Kelly, 336 F.3d at 815–16, 818. By\\nusing the thumbnail images in service of the search engine,\\nGoogle “transforms the image,” which might have been\\ncreated for an “entertainment, aesthetic, or informative\\nfunction,” “into a pointer directing a user to a source of\\ninformation.” Amazon, 508 F.3d at 1165. As a result of the\\nnew function that the image serves, Google’s use of the\\nentire image in its search engine results “does not diminish\\nthe transformative nature of [its] use.” Id. And, further\\ndeveloping the public benefit principle from Kelly, we\\nemphasized that Google’s search engine both “promotes the\\npurposes of copyright and serves the interests of the public,”\\nwhich significantly outweighed the superseding or\\n\\f VHT, INC. V. ZILLOW GROUP 29\\n\\ncommercial uses of the search engine, and strongly\\nsupported finding fair use. Id. at 1166.\\n\\n More recently, the Second Circuit considered whether\\nfair use protected the Google Books search engine, which\\nemploys digital, machine-readable copies of millions of\\ncopyright-protected books scanned by Google. Authors\\nGuild, 804 F.3d at 207–08. 9 The Google Books search\\nengine enables a full text search, which makes possible\\nsearching for a specific term, and then provides “snippets,”\\nor a part of a page, for users to read. Id. at 208–09, 216–17.\\nThe court held that both functions involve a “highly\\ntransformative purpose of identifying books of interest to the\\nsearcher.” Id. at 218. The search function “augments public\\nknowledge by making available information about\\nPlaintiffs’ books without providing the public with a\\nsubstantial substitute.” Id. at 207. And the search engine\\nmakes possible a new type of research known as “text\\nmining” or “data mining,” whereby users can search across\\nthe corpus of books to determine the frequency of specified\\nterms across time. Id. at 209, 217. Additionally, the\\n“snippet” view provides context for users to assess if a book\\nis relevant to them, without providing so much context as to\\nsupersede the original. Id. at 218. To boot, Google often\\nprovides a link to a page where the entire book can be found\\nat a library or purchased. Id. at 209. Concluding that\\nGoogle’s commercial motivation did not significantly\\noutweigh these transformative uses, the court held that the\\nfirst factor strongly supported a finding of fair use. Id. at\\n219.\\n\\n\\n 9\\n The Google Books search engine also featured millions of public\\ndomain texts. For obvious reasons, fair use was not at issue with those\\nworks.\\n\\f30 VHT, INC. V. ZILLOW GROUP\\n\\n What we divine from these cases is that the label “search\\nengine” is not a talismanic term that serves as an on-off\\nswitch as to fair use. Rather, these cases teach the\\nimportance of considering the details and function of a\\nwebsite’s operation in making a fair use determination. We\\nnow examine Digs with those lessons in mind.\\n\\n c. Application of Fair Use Principles\\n\\n As noted, the first factor assesses the character of the use,\\nincluding whether it is commercial in nature, and, critically,\\nwhether it is “transformative.” There is no dispute that\\nZillow’s use is for commercial purposes, a factor we cannot\\nignore. To determine if that use is transformative, we\\nconsider whether and to what extent it “adds something new,\\nwith a further purpose or different character, altering the first\\nwith new expression, meaning, or message[.]” Campbell,\\n510 U.S. at 579. Though we agree that Digs is a type of\\nsearch engine because it offers searchable functionality, it is\\nqualitatively different than Google and other open-universe\\nsearch engines, as well as different than the Google Books\\nsearch engine.\\n\\n Most simply, a search engine is a software program that\\nenables information retrieval by helping users find\\ninformation through the use of keyword queries. But not all\\nsearch engines are created equal. The search engines\\ncommonly used for day-to-day research are internet-wide\\nsearch engines, like Google, Yahoo, or Bing. These search\\nengines are programs powered by algorithms that search or\\n“crawl” the web. A search engine like Google then indexes\\nwebsites, stores them on a database, and runs users’ search\\nqueries against it. Search results are typically a mix of\\nimages and text, which include hyperlinks to sources of that\\ncontent elsewhere on the web. Amazon, 508 F.3d at 1155.\\n\\f VHT, INC. V. ZILLOW GROUP 31\\n\\n Unlike the internet-wide search engines considered in\\nAmazon and Kelly, Digs is a closed-universe search engine\\nthat does not “crawl” the web. Users can run searches on the\\n“searchable set” of images within Digs’ walled garden,\\nwhich includes VHT photos. The search results do not direct\\nusers to the original sources of the photos, such as VHT’s\\nwebsite. Rather, they link to other pages within Zillow’s\\nwebsite and, in some cases, to third-party merchants that sell\\nitems similar to those featured in the photo.\\n\\n That Digs makes these images searchable does not\\nfundamentally change their original purpose when produced\\nby VHT: to artfully depict rooms and properties.\\nAdditionally, Digs displays the entire VHT image, not\\nmerely a thumbnail. Unlike in Amazon, the new image does\\nnot serve a “different function” than the old one. Amazon,\\n508 F.3d at 1165. Zillow’s use preserves the photos’\\n“inherent character.” Monge, 688 F.3d at 1176. And Zillow\\n“simply supersed[es] [VHT’s] purpose” in creating the\\nimages in the first place. Kelly, 336 F.3d at 819–20; see\\nHarper & Row Publishers, Inc. v. Nation Enters., 471 U.S.\\n539, 550–51 (1985) (holding that if a new work\\n“supersede[s] the use of the original,” it is probably not a fair\\nuse).\\n\\n Comparing Digs to the Google Books search engine\\nfurther drives home this analysis. We agree with the Second\\nCircuit’s observation that the copyright dispute over the\\nGoogle Books search engine “tests the boundaries of fair\\nuse.” Authors Guild, 804 F.3d at 206. We conclude that\\nDigs goes one step further—and crosses the line.\\n\\n Like Digs, the Google Books search engine operates on\\na closed database comprised of complete digital copies of\\noriginal works. Id. at 217. But the similarities end there.\\n\\f32 VHT, INC. V. ZILLOW GROUP\\n\\nThe Google Books project makes it possible to search books\\nfor “identifying information instantaneously supplied [that]\\nwould otherwise not be obtainable in lifetimes of searching.”\\nId. at 209. With this broad purpose in mind, the court\\nrightfully observed that this system “augments public\\nknowledge.” Id. at 207. This rationale bears no comparison\\nto Digs.\\n\\n Nor is the limited transformation present on Digs\\nremotely comparable to the unprecedented text mining, word\\npattern, frequency of use, and other statistical analyses made\\npossible for the first time by Google Books. Google Books\\nsearch results provide “[a] brief description of each book,\\nentitled ‘About the Book,’” as well as, for some books, links\\nfor borrowing or purchasing the book. Id. at 209.\\nSignificantly, a Google Books search produces only a\\n“snippet” of the book, and sometimes it “disables snippet\\nview entirely.” Id. at 210. At the request of a rights holder,\\nGoogle “will exclude any book altogether from snippet\\nview.” Id.\\n\\n These features, in conjunction with other creative aspects\\nof Google Books, result in a categorically more\\ntransformative use than Zillow’s simple tagging and query\\nsystem that displays full-size copyrighted images serving the\\nsame purpose as the originals, with no option to opt out of\\nthe display, and with few, if any, transformative qualities.\\nAny transformation by Zillow pales in comparison to the\\nuses upheld in prior search engine cases. Such use also does\\nnothing to further the use of copyrighted works for the\\nsocially valuable purposes identified in the Copyright Act\\nitself, like “criticism, comment, news reporting,\\nteaching . . . , scholarship, or research.” 17 U.S.C. § 107;\\nsee also 4 NIMMER ON COPYRIGHT § 13.05[A]. The lack of\\ntransformation is especially significant because, as Kelly\\n\\f VHT, INC. V. ZILLOW GROUP 33\\n\\nteaches, “[t]he more transformative the new work, the less\\nimportant the other factors, including commercialism,\\nbecome.” 336 F.3d at 818 (citing Campbell, 510 U.S. at\\n579). If, as the Second Circuit suggested, Google’s use\\n“tests the boundaries of fair use,” Zillow’s efforts push Digs\\ninto the outer space of fair use. Authors Guild, 804 F.3d at\\n206. So while Google Books may inch across the boundary\\nof fair use, Zillow’s use does not approach the line.\\n\\n Our decisions in Amazon and Kelly provide a roadmap\\nfor analyzing the second factor, which focuses on the nature\\nof the copyrighted work. In those cases, we held that\\nphotographers’ images are creative, especially when they are\\ncreated for public viewing. Amazon, 508 F.3d at 1167;\\nKelly, 336 F.3d at 820. “Works that are creative in nature\\nare ‘closer to the core of intended copyright protection’ than\\nare more fact-based works.” Napster, 239 F.3d at 1016\\n(quoting Campbell, 510 U.S. at 586).\\n\\n So too here. VHT’s photos are aesthetically and\\ncreatively shot and edited by professional photographers.\\nThat Zillow’s curators select the most creative photos for the\\nDigs searchable set underscores the creative nature of the\\nworks. But, as the district court properly noted, this factor\\noperates “with less force” in favor of VHT because the\\nphotos had already been published on the Listing Platform.\\nSee Kelly, 336 F.3d at 820 (“The fact that a work is published\\nor unpublished also is a critical element of its nature.\\nPublished works are more likely to qualify as fair use\\nbecause the first appearance of the artist’s expression has\\nalready occurred.”). Ultimately, this factor only slightly\\nfavors VHT, further cutting against finding fair use.\\n\\n The third factor evaluates the amount and substantiality\\nof the copyrighted work that was used. “[C]opying an entire\\n\\f34 VHT, INC. V. ZILLOW GROUP\\n\\nwork militates against a finding of fair use.” Worldwide\\nChurch of God v. Phila. Church of God, Inc., 227 F.3d 1110,\\n1118 (9th Cir. 2000) (quotation and citation omitted).\\nHowever, this analysis is informed by the “purpose of the\\ncopying.” Campbell, 510 U.S. at 586. In that spirit, we have\\nfound that copying full works qualifies as fair use where “[i]t\\nwas necessary . . . to copy the entire image to allow users to\\nrecognize the image and decide whether to pursue more\\ninformation about the image or the originating web site.”\\nKelly, 336 F.2d at 821. In contrast to Amazon and Kelly,\\nnothing justifies Zillow’s full copy display of VHT’s photos\\non Digs.\\n\\n Finally, the fourth factor considers “the effect of the use\\nupon the potential market for or value of the copyrighted\\nwork.” 17 U.S.C. § 107(4). To defeat a fair use defense,\\n“one need only show that if the challenged use should\\nbecome widespread, it would adversely affect the potential\\nmarket for the copyrighted work.” Harper & Row\\nPublishers, 471 U.S. at 568 (citation and quotation omitted).\\n\\n Although VHT had licensed only a handful of photos for\\nsecondary uses (and none on a searchable database), that\\nmarket was more than “hypothetical.” See Amazon, 508\\nF.3d at 1168. Significantly, VHT was “actively exploring”\\nthe market for licensing its photos to home design websites\\nlike Digs—including with Zillow itself. This factor favors\\nVHT.\\n\\n Taken together, the nature of Zillow’s use, when\\nintegrated with the four factors, cuts against finding fair use\\nby Zillow. We affirm the district court’s grant of summary\\njudgment to VHT with respect to fair use.\\n\\f VHT, INC. V. ZILLOW GROUP 35\\n\\n II. Secondary Infringement—Digs\\n\\n Turning to VHT’s claim for secondary infringement, the\\ndistrict court correctly concluded that Zillow did not\\nsecondarily infringe VHT’s exclusive rights in the 28,125\\nphotos used on Digs, aside from 114 images created on Digs\\nafter VHT specifically identified them, which are not on\\nappeal. As noted before with regard to the district court’s\\nruling on direct infringement, “we apply the same standard\\nused by the district court in evaluating the jury’s verdict” and\\nuphold the verdict unless “the evidence permits only one\\nreasonable conclusion, and that conclusion is contrary to the\\njury’s verdict.” Wallace, 479 F.3d at 624; Fed. R. Civ. P.\\n50(a). The jury’s verdict here did not meet this standard.\\nWe affirm the district court’s grant of Zillow’s motion for\\njudgment notwithstanding the verdict with respect to\\nsecondary infringement, both contributory and vicarious\\ninfringement.\\n\\n A. Contributory Liability\\n\\n Contributory liability requires that a party “(1) has\\nknowledge of another’s infringement and (2) either\\n(a) materially contributes to or (b) induces that\\ninfringement.” Perfect 10, Inc. v. Visa Int’l Serv., Ass’n, 494\\nF.3d 788, 795 (9th Cir. 2007). Zillow’s actions do not satisfy\\nthe second prong—material contribution or inducement—so\\nwe do not address the first prong. See Giganews, 847 F.3d\\nat 671.\\n\\n In Giganews, we outlined the means of material\\ncontribution to infringement:\\n\\n In the online context, . . . a “computer system\\n operator” is liable under a material\\n contribution theory of infringement “if it has\\n\\f36 VHT, INC. V. ZILLOW GROUP\\n\\n actual knowledge that specific infringing\\n material is available using its system, and can\\n take simple measures to prevent further\\n damage to copyrighted works, yet continues\\n to provide access to infringing works.”\\n\\nId. at 671 (quoting Amazon, 508 F.3d at 1172); see also\\nEllison v. Robertson, 357 F.3d 1072, 1078 (9th Cir. 2004)\\n(applying this standard in the online context); Napster, 239\\nF.3d at 1022 (same); Religious Tech. Ctr. v. Netcom On-Line\\nComm’n Servs., Inc., 907 F. Supp. 1361, 1375 (N.D. Cal.\\n1995) (same). There is insufficient evidence to support the\\ncontributory infringement verdict under the “simple\\nmeasures” standard.\\n\\n VHT’s position that “the jury could have reasonably\\ndecided that Zillow in fact had the means to identify and\\nremove” the allegedly infringing images that VHT identified\\nby property address, as opposed to their website designation\\nor Uniform Resource Locator (“URL”), fails. Zillow\\ntestified throughout trial that, in order to systematically or\\nswiftly take down a large number of photos, it required the\\nZillow Image ID–a number that is in the URL for each\\nimage. This stands to reason, because “Zillow receives\\nmultiple copies of the same photograph, depicting the same\\nproperty, with the same listing agent, from different feeds.”\\nMerely identifying the physical property address in no way\\nidentified the proper feed or the correct photo. Thus, Zillow\\ndid not have appropriately “specific” information necessary\\nto take “simple measures” to remedy the violation.\\n\\n VHT’s argument that Zillow is liable for failing to ask\\nfor the URLs of the allegedly infringing photos also fails.\\nAsking for the URLs was not Zillow’s duty under the\\ncontributory liability standard: Zillow must have “actual\\n\\f VHT, INC. V. ZILLOW GROUP 37\\n\\nknowledge that specific infringing material is available using\\nits system.” Amazon, 508 F.3d at 1172 (citation and\\nquotation omitted). Zillow first reasonably asked to see the\\nlicenses between VHT and the feed providers; otherwise,\\nZillow could not assess ownership and rights in the\\nundefined images. That Zillow did not proactively request a\\nlist of URLs before VHT filed suit does not make Zillow\\ncontributorily liable.\\n\\n Additionally, Zillow’s failure to systematically use\\nwatermarking technology does not show there was a “simple\\nmeasure” available that it failed to use. Even assuming there\\nwere “reasonable and feasible means” for Zillow to employ\\nwatermark detection technology, in practice VHT rarely\\nwatermarked its photos. See Giganews, 847 F.3d at 672\\n(citation and quotation omitted).\\n\\n Nor did Zillow induce infringement. Inducement\\nliability requires evidence of “active steps . . . taken to\\nencourage direct infringement.” Metro-Goldwyn-Mayer\\nStudios Inc. v. Grokster, Ltd., 545 U.S. 913, 936 (2005)\\n(citation and quotation omitted)). Evidence of active steps\\nincludes “advertising an infringing use or instructing how to\\nengage in an infringing use,” because such evidence\\n“show[s] an affirmative intent that the product be used to\\ninfringe.” Id.; see also Columbia Pictures Industries, Inc. v.\\nFung, 710 F.3d 1020, 1032 (9th Cir. 2013) (inducement\\nliability requires “an object of promoting [the product’s] use\\nto infringe copyright”). The “improper object” of\\ninfringement “must be plain and must be affirmatively\\ncommunicated through words or actions[.]” Fung, 710 F.3d\\nat 1034.\\n\\n In view of the evidence, “no reasonable juror could\\nconclude [Zillow] distributed its product with the object of\\n\\f38 VHT, INC. V. ZILLOW GROUP\\n\\npromoting its use to infringe copyright.” Giganews, 847\\nF.3d at 672 (quotation and citation omitted). For example,\\nZillow’s generally applicable tools and messages for users\\nto save more photos from the Listing Platform to Digs does\\nnot “promote[] the use of [Digs] specifically to infringe\\ncopyrights.” Amazon, 508 F.3d at 1170 n.11. Nor does\\nevidence that Zillow sometimes makes mistakes about the\\ndisplay rights in a feed plainly communicate an improper\\nobject of infringement. Zillow corrects these inadvertent\\nerrors when it learns of them. Because a “failure to take\\naffirmative steps to prevent infringement” alone cannot\\ntrigger inducement liability, the inducement claim is a\\nparticularly poor fit for Zillow’s real estate and home design\\nwebsites, which have “substantial noninfringing uses.”\\nGrokster, 545 U.S. at 939 n.12.\\n\\n B. Vicarious Liability\\n\\n Neither does the vicarious liability theory fit the Zillow\\nplatform. To prevail on a vicarious liability claim, “[VHT]\\nmust prove ‘[Zillow] has (1) the right and ability to supervise\\nthe infringing conduct and (2) a direct financial interest in\\nthe infringing activity.’” Giganews, 847 F.3d at 673\\n(quoting Visa, 494 F.3d at 802). The first element requires\\n“both a legal right to stop or limit the directly infringing\\nconduct, as well as the practical ability to do so.” Amazon,\\n508 F.3d at 1173. VHT’s vicarious infringement argument\\nfails because, as the district court found, “Zillow ‘lack[ed]\\nthe practical ability to police’ its users’ infringing conduct”\\non Digs.\\n\\n As discussed with respect to contributory infringement,\\nthere was insufficient evidence that Zillow had the technical\\nability to screen out or identify infringing VHT photos\\namong the many photos that users saved or uploaded daily\\n\\f VHT, INC. V. ZILLOW GROUP 39\\n\\nto Digs. Regardless, such allegations do not fall under the\\nvicarious liability rubric: Zillow’s “failure to change its\\noperations to avoid assisting [users] to distribute . . .\\ninfringing content . . . is not the same as declining to exercise\\na right and ability to make [third parties] stop their direct\\ninfringement.” Amazon, 508 F.3d at 1175.\\n\\n Our conclusion is consistent with earlier dicta that “the\\nvicarious liability standard applied in Napster can be met by\\nmerely having the general ability to locate infringing\\nmaterial and terminate users’ access.” UMG Recordings,\\nInc. v. Shelter Capital Partners LLC, 718 F.3d 1006, 1030\\n(9th Cir. 2013) (emphasis added). Once VHT photos were\\nuploaded to the Listing Platform with appropriate\\ncertification of rights, ferreting out claimed infringement\\nthrough use on Digs was beyond hunting for a needle in a\\nhaystack. As the district court concluded, “the trial record\\nlacks substantial evidence of a practical ability to limit direct\\ninfringement for the same reasons it lacks substantial\\nevidence of simple measures to remove infringing material.”\\nAnd linking a claimed infringement to a feed provider was\\neven more of an impossibility.\\n\\n We affirm the district court’s judgment notwithstanding\\nthe verdict concluding that substantial evidence did not\\nsupport the claim that Zillow secondarily infringed VHT’s\\nexclusive rights in its photos.\\n\\n III. Damages\\n\\n A. Compilation\\n\\n The size of the damages award hinges on whether VHT’s\\nphotos used on Digs are part of a “compilation” or if they are\\nindividual photos. This distinction makes a difference. If\\nthe VHT photo database is a “compilation,” and therefore\\n\\f40 VHT, INC. V. ZILLOW GROUP\\n\\none “work” for the purposes of the Copyright Act, then VHT\\nwould be limited to a single award of statutory damages for\\nZillow’s use of thousands of photos on Digs. 17 U.S.C.\\n§ 504(c)(1). But if the database is not a compilation, then\\nVHT could seek damages for each photo that Zillow used.\\n\\n In lieu of actual damages, a copyright holder may elect\\nto receive statutory damages under the Copyright Act. 17\\nU.S.C. § 504(c)(1). VHT did so. The Act provides for “an\\naward of statutory damages for all infringements involved in\\nthe action, with respect to any one work . . . in a sum of not\\nless than $750 or more than $30,000.” Id. This provision\\nties statutory damages to the term “work,” which is\\nundefined, except in a circular manner: copyright law\\nprotects “original works of authorship.” 17 U.S.C. § 102(a).\\nFortunately, there is a definition of compilation: “a work\\nformed by the collection and assembling of preexisting\\nmaterials or of data that are selected, coordinated, or\\narranged in such a way that the resulting work as a whole\\nconstitutes an original work of authorship.” 17 U.S.C.\\n§ 101. For purposes of statutory damages, “all the parts of a\\ncompilation . . . constitute one work.” 17 U.S.C.\\n§ 504(c)(1). “The question of whether a work constitutes a\\n‘compilation’ for the purposes of statutory damages pursuant\\nto Section 504(c)(1) of the Copyright Act is a mixed question\\nof law and fact.” Bryant v. Media Right Prods., Inc., 603\\nF.3d 135, 140 (2d Cir. 2010).\\n\\n Whether various VHT photo collections comprise one or\\nmore compilations is a threshold damages question. Before\\ntrial, Zillow asked for a legal determination on the\\ncompilation issue. That motion was denied. However, the\\ndistrict court did not make an explicit determination about\\ncompilation and the specifics of compilation were not put\\nbefore the jury. In fairness to the district court, we might\\n\\f VHT, INC. V. ZILLOW GROUP 41\\n\\ninfer from the transcript that there was an implicit\\ndetermination as to whether VHT’s photos are part of a\\ncompilation, but we are left in doubt. Instead, the jury was\\ninstructed that “[e]ach VHT Photo that has independent\\neconomic value constitutes a separate work.” On the verdict\\nform, the jury was asked which “photographs have\\nindependent economic value.”\\n\\n The notion of “independent economic value” derives not\\nfrom the statute, but from case law. In the Ninth Circuit, the\\nquestion of whether something—like a photo, television\\nepisode, or so forth—has “independent economic value”\\ninforms our analysis of whether the photo or episode is a\\nwork, though it is not a dispositive factor. See Columbia\\nPictures Television, Inc. v. Krypton Broad. of Birmingham,\\nInc., 259 F.3d 1186, 1193 (9th Cir. 2001). But consideration\\nof the independent economic value factor does not answer\\nthe question whether something is a compilation. That\\nquestion remains unanswered here.\\n\\n On appeal, the parties have polar opposite views on\\nwhether there was a compilation (Zillow’s position) or\\nwhether each photo is entitled to a separate damages award\\n(VHT’s position). VHT registered thousands of photos as\\ncompilations. But the Copyright Office warns that such a\\nregistration “may” limit the copyright holder “to claim only\\none award of statutory damages in an infringement action,\\neven if the defendant infringed all of the component works\\ncovered by the registration.” U.S. COPYRIGHT OFFICE,\\nCOMPENDIUM OF U.S. COPYRIGHT OFFICE PRACTICES\\n§ 1104.5 (Sept. 2017), https://www.copyright.gov/comp3/\\ndocs/compendium.pdf. Though the registration label is not\\ncontrolling, it may be considered by the court when\\nassessing whether a work is a compilation. See Yellow\\nPages Photos, Inc. v. Ziplocal, LP, 795 F.3d 1255, 1277\\n\\f42 VHT, INC. V. ZILLOW GROUP\\n\\n(11th Cir. 2015) (“Although the manner of copyright\\nregistration is not dispositive of the works issue, this Court\\nhas previously considered it to be at least a relevant factor.”).\\nUltimately, what counts is the statutory definition.\\n\\n Because there were at least ten different copyright\\nregistrations, thousands of photos, and no explicit\\ndetermination on compilation, we decline to sort out the\\ncompilation issue on appeal. We remand to the district court\\nfor further proceedings as to whether the VHT photos\\nremaining at issue were a compilation.\\n\\n B. Willfulness\\n\\n The jury found that Zillow willfully infringed exclusive\\nrights to 3,373 searchable VHT photos that were eligible for\\nstatutory damages. The district court largely upheld the\\nwillfulness finding in its post-trial motions order. However,\\nthe court granted Zillow judgment notwithstanding the\\nverdict on 673 images that were not displayed, so the court’s\\nfinal willfulness judgment applied to 2,700 searchable\\nphotos on Digs.\\n\\n To uphold a jury’s willfulness finding, there must be\\nsubstantial evidence “(1) that the [the infringing party] was\\nactually aware of the infringing activity, or (2) that the\\n[infringing party’s] actions were the result of reckless\\ndisregard for, or willful blindness to, the copyright holder’s\\nrights.” Unicolors, 853 F.3d at 991 (citation and quotation\\nomitted). Under the second prong, willful blindness requires\\nthat the infringing party “(1) subjectively believed that\\ninfringement was likely occurring on their networks and that\\nthey (2) took deliberate actions to avoid learning about the\\ninfringement.” Luvdarts, LLC v. AT&T Mobility, LLC, 710\\nF.3d 1068, 1073 (9th Cir. 2013). Reckless disregard can be\\ndemonstrated, for example, when a party “refus[es], as a\\n\\f VHT, INC. V. ZILLOW GROUP 43\\n\\nmatter of policy, to even investigate or attempt to determine\\nwhether particular [photos] are subject to copyright\\nprotections.” Unicolors, 853 F.3d at 992. A finding of\\nwillfulness has significant financial consequences—the jury\\nmay increase damages up to $150,000 per violation. See 17\\nU.S.C. § 504(c)(2).\\n\\n As noted with respect to infringement, we do not take\\nlightly the decision to reverse a jury verdict, nor do we\\ncavalierly set aside the district court’s thoughtful analysis.\\nBut here, we are compelled to disagree with both because\\nsubstantial evidence does not support willfulness as to the\\n2,700 photos.\\n\\n The test for willfulness is in the alternative: either actual\\nnotice or recklessness shown by reckless disregard or\\nturning a blind eye to infringement. We turn first to actual\\nnotice. That VHT provided Zillow with minimal notice of\\ninfringement does not itself establish that any subsequent\\ninfringement was willful. Rather, “[c]ontinued use of a work\\neven after one has been notified of his or her alleged\\ninfringement does not constitute willfulness so long as one\\nbelieves reasonably, and in good faith, that he or she is not\\ninfringing.” Evergreen Safety Council v. RSA Network Inc.,\\n697 F.3d 1221, 1228 (9th Cir. 2012); see also Danjaq LLC\\nv. Sony Corp., 263 F.3d 942, 959 (9th Cir. 2001) (“It would\\nseem to follow that one who has been notified that his\\nconduct constitutes copyright infringement, but who\\nreasonably and in good faith believes the contrary, is not\\n‘willful’ for these purposes.”) (quoting 4 NIMMER ON\\nCOPYRIGHT § 14.04[B][3]).\\n\\n Such is the case here. Zillow’s agreements with its feed\\nproviders grant it an express license to use, copy, distribute,\\npublicly display, and create derivative works for each photo,\\n\\f44 VHT, INC. V. ZILLOW GROUP\\n\\nand the agreements include unambiguous representations by\\nthe feed providers that they have the authority to assign such\\nrights. Zillow developed procedures to identify ex ante the\\nscope of its license for each uploaded photo and employed\\nautomated protocols to manage the use of each photo\\nconsistent with its evergreen or deciduous designation. At\\nno point during their year of communications prior to\\nissuance of the notice letter did VHT raise the specter of\\ninfringement. Notably, VHT’s eventual notice was minimal:\\none letter with a list of allegedly infringing photos,\\ndesignated by residential street address, not web address.\\n\\n The notion that Zillow failed to take appropriate\\nresponsive measures after receiving this notice is belied by\\nthe record. Zillow immediately requested information to\\nconfirm VHT’s copyright ownership and cross-reference the\\nphotos with licensing information. VHT was not\\nforthcoming with that information. Rather, in response,\\nVHT offered merely an unsigned form contract. Instead of\\nproviding helpful information, VHT then filed suit. Given\\nthe limited information provided by VHT, Zillow could not\\nreasonably be expected to have promptly and unilaterally\\nremoved each flagged photo. As the district court noted,\\nVHT failed to demonstrate there were simple measures\\navailable for the removal of infringing photos or that Zillow\\nhad any “practical ability to independently identify\\ninfringing images.”\\n\\n Thus, we are compelled to conclude that substantial\\nevidence does not show Zillow was “actually aware” of its\\ninfringing activity. See Evergreen Safety Council, 697 F.3d\\nat 1228. Zillow’s belief that feed providers had properly\\nlicensed its uses and that its system effectively respected\\nthose rights was reasonable. And, as the district court\\nobserved, “[t]he record suggests no reason to conclude that\\n\\f VHT, INC. V. ZILLOW GROUP 45\\n\\nZillow maintained that position in bad faith, and Zillow’s\\nnon-infringement contention proved accurate as to most of\\nthe images at issue in this lawsuit.”\\n\\n We reach the same conclusion as to whether Zillow\\nrecklessly disregarded or willfully blinded itself to its\\ninfringement. In reaching an opposite conclusion, the\\ndistrict court observed that Zillow did not “perform[] further\\ninvestigation into the rights each [feed provider] possesses,”\\nnor did it “t[ake] responsive measures to obtain further\\ninformation” after VHT provided the minimal notice of\\npotential infringement. That conclusion is at odds with the\\nevidence, for the reasons outlined above.\\n\\n VHT’s argument that Zillow, a sophisticated business\\nwith a robust legal team, should have known that its feed\\nprovider license agreements were invalid is unavailing.\\nVHT argues that when Zillow saw the non-exclusive grant\\nof rights in VHT’s unsigned form contract, showing that the\\nfeed providers did not have a right to sublicense, Zillow\\nshould have known the licenses were invalid. Despite\\nrequests for such information, Zillow did not have access to\\nVHT’s executed licenses with the feed providers who\\nfurnished VHT’s photos to Zillow. Access to a blank form\\ncontract (that the district court earlier found ambiguous as a\\nmatter of law) is not enough. We conclude that substantial\\nevidence does not show Zillow was “reckless or willfully\\nblind” as to its infringement. We reverse the district court\\nand vacate the jury’s finding of willful infringement.\\n\\n IV. Conclusion\\n\\n We affirm the district court’s summary judgment in\\nfavor of Zillow on direct infringement of the Listing\\nPlatform photos. With respect to direct liability on the Digs\\nphotos, we affirm the district court’s grant of judgment\\n\\f46 VHT, INC. V. ZILLOW GROUP\\n\\nnotwithstanding the verdict on the 22,109 non-displayed\\nphotos and the 2,093 displayed but not searchable photos.\\nWe uphold summary judgment in favor of VHT on the 3,921\\ndisplayed, searchable photos.\\n\\n We affirm the district court’s judgment notwithstanding\\nthe verdict on secondary liability, both contributory and\\nvicarious, on the Digs photos.\\n\\n We reverse the district court’s denial of judgment\\nnotwithstanding the verdict on the issue of willfulness and\\nvacate the jury’s finding on willfulness.\\n\\n We remand consideration of the compilation issue to the\\ndistrict court.\\n\\n Affirmed in part, reversed in part and remanded.\\nEach party shall pay its own costs on appeal.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4380667/\", \"author_raw\": \"McKEOWN, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"M MARGARET MCKEOWN"},"judge2":{"kind":"string","value":"WILLIAM A FLETCHER"},"judge3":{"kind":"string","value":"RONALD M GOULD"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4603414/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Vht, Inc. v. Zillow Group, Inc.\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-15\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nVHT, INC., a Delaware corporation, Nos. 17-35587\n Plaintiff-Appellee/ 17-35588\n Cross-Appellant,\n\n v. D.C. No.\n 2:15-cv-01096-\nZILLOW GROUP, INC., a Washington JLR\ncorporation; ZILLOW, INC., a\nWashington corporation,\n Defendants-Appellants/ OPINION\n Cross-Appellees.\n\n Appeal from the United States District Court\n for the Western District of Washington\n James L. Robart, Senior District Judge, Presiding\n\n Argued and Submitted August 28, 2018\n Seattle, Washington\n\n Filed March 15, 2019\n\nBefore: M. Margaret McKeown, William A. Fletcher, and\n Ronald M. Gould, Circuit Judges.\n\n Opinion by Judge McKeown\n\f2 VHT, INC. V. ZILLOW GROUP\n\n SUMMARY *\n\n\n Copyright Law\n\n The panel affirmed in part and reversed in part the\ndistrict court’s judgment after a jury trial and remanded in a\ncopyright infringement action brought by VHT, Inc., a real\nestate photography studio, against Zillow Group, Inc., and\nZillow, Inc., an online real estate marketplace.\n\n VHT alleged that Zillow’s use of photos on the “Listing\nPlatform” and “Digs” parts of its website exceeded the scope\nof VHT’s licenses to brokers, agents, and listing services that\nprovided those photos to Zillow. The district court granted\npartial summary judgment on a limited set of claims. The\njury found in favor of VHT on most remaining claims,\nawarding over $8.27 million in damages. The district court\npartially granted Zillow’s post-trial motion for judgment\nnotwithstanding the verdict, reversing in part the jury verdict\nand reducing total damages to approximately $4 million.\n\n The panel affirmed the district court’s summary\njudgment in favor of Zillow on direct infringement of the\nListing Platform photos. The panel held that VHT failed to\nestablish that Zillow engaged in volitional conduct by\nexercising control over the content of the Listing Platform.\n\n With respect to direct liability on the Digs photos, the\npanel affirmed the district court’s grant in favor of Zillow of\njudgment notwithstanding the verdict on 22,109 non-\ndisplayed photos and 2,093 displayed but not searchable\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f VHT, INC. V. ZILLOW GROUP 3\n\nphotos. The panel held that VHT did not present substantial\nevidence that Zillow, through the Digs platform, directly\ninfringed its display, reproduction, or adaption rights.\n\n The panel upheld summary judgment in favor of VHT\non 3,921 displayed, searchable Digs photos. The panel held\nthat fair use did not absolve Zillow of liability because\nZillow’s tagging of the photos for searchable functionality\nwas not a transformative fair use.\n\n The panel affirmed the district court’s grant in favor of\nZillow of judgment notwithstanding the verdict on\nsecondary liability, both contributory and vicarious, on the\nDigs photos.\n\n As to damages, the panel remanded consideration of the\nissue whether VHT’s photos used on Digs were part of a\ncompilation or were individual photos.\n\n The panel reversed the district court’s denial of judgment\nnotwithstanding the verdict on the issue of willfulness and\nvacated the jury’s finding on willfulness. The panel\nconcluded that substantial evidence did not show that Zillow\nwas “actually aware” of its infringing activity, nor that\nZillow recklessly disregarded or willfully blinded itself to its\ninfringement.\n\n\n COUNSEL\n\nIan B. Crosby (argued), Edgar G. Sargent, Genevieve Vose\nWallace, and Jenna G. Farleigh, Susman Godfrey LLP,\nSeattle, Washington, for Defendants-Appellants/Cross-\nAppellees.\n\f4 VHT, INC. V. ZILLOW GROUP\n\nStephen M. Rummage (argued), Marcia B. Paul, Jonathan\nM. Lloyd, James E. Howard, and Max B. Hensley, Davis\nWright Tremaine LLP, Seattle, Washington, for Plaintiff-\nAppellee/Cross-Appellant.\n\nKeith Kupferschmid and Terry Hart, Copyright Alliance,\nWashington, D.C.; Eleanor M. Lackman and Lindsay W.\nBowen, Cowan, DeBaets, Abrahams & Sheppard LLP, New\nYork, New York; for Amicus Curiae Copyright Alliance.\n\nAlicia Calzada, Alicia Wagner Calzada, PLLC, San\nAntonio, Texas, for Amici Curiae The American Society of\nMedia Photographers, Inc., Digital Media Licensing\nAssociation, Inc., Graphic Artist Guild, Inc., and National\nPress Photographers Association, Inc.\n\nThomas G. Hentoff and Chanakya A. Sethi, Williams &\nConnolly LLP, Washington D. C., for Amici Curiae Disney\nEnterprises, Inc., Twentieth Century Fox Film Corporation,\nand Recording Industry Association of America, Inc.\n\nMitchell L. Stoltz, Electronic Frontier Foundation, San\nFrancisco, California, for Amicus Curiae Electronic Frontier\nFoundation.\n\nBrian M. Willen, Wilson Sonsini Goodrich & Rosati, P.C.,\nNew York, New York; Ryan T. O’Hollaren, Wilson Sonsini\nGoodrich & Rosati, P.C., Palo Alto, California; for Amici\nCuriae Internet Association and Computer &\nCommunications Industry Association.\n\f VHT, INC. V. ZILLOW GROUP 5\n\n OPINION\n\nMcKEOWN, Circuit Judge:\n\n Zillow, an online real estate marketplace, has become a\npopular website for homeowners and others to check\nestimated valuations of their property, look for houses and\ncondominiums for sale and rent, and see photographs of a\nwide range of properties. Thousands of those copyrighted\nphotos come from VHT, the largest professional real estate\nphotography studio in the country.\n\n The copyright claims on appeal concern Zillow’s use of\nVHT’s photos on two parts of Zillow’s website: the “Listing\nPlatform” and “Digs.” The Listing Platform is the core of\nthe website, featuring photos and information about real\nestate properties, both on and off the market. Zillow claims\nthat the site includes “most homes in America.” Digs\nfeatures photos of artfully-designed rooms in some of those\nproperties and is geared toward home improvement and\nremodeling. Zillow tags photos on the Listing Platform so\nthat Digs users can search the database by various criteria,\nlike room type, style, cost, and color.\n\n Real estate brokers, listing services, and agents hire VHT\nto take professional photos of new listings for marketing\npurposes. A VHT photographer takes the photos and sends\nthem to the company’s studio for touch-up, where they are\nsaved to VHT’s electronic photo database, and then\ndelivered to the client for use under license. Each license\nagreement between VHT and its clients differs slightly, but\neach contract generally grants the requesting client the right\nto use the photos in the sale or marketing of the featured\nproperty. Zillow receives these photos and other data in\nfeeds from various real estate-related sources.\n\f6 VHT, INC. V. ZILLOW GROUP\n\n In 2015, VHT sued Zillow Group, Inc., and Zillow, Inc.,\n(collectively “Zillow”) for copyright infringement, alleging\nthat Zillow’s use of photos on the Listing Platform and Digs\nexceeded the scope of VHT’s licenses to brokers, agents, and\nlisting services who provided those photos to Zillow. The\ndistrict court granted partial summary judgment on a limited\nset of claims, while other claims advanced to trial. The jury\nfound in favor of VHT on most remaining claims, awarding\nover $8.27 million in damages. The district court partially\ngranted Zillow’s post-trial motion for judgment\nnotwithstanding the verdict, reversing in part the jury verdict\nand reducing total damages to approximately $4 million.\n\n The parties cross-appealed issues stemming from partial\nsummary judgment, the jury verdict, and judgment\nnotwithstanding the verdict. We affirm in part and reverse\nin part. 1\n\n To simplify and make sense of the various claims, this\nopinion does not split out the appeal and cross-appeal as was\ndone in the briefing to the court. Instead, the opinion\nseparately addresses liability for each of the categories of\nphotos at issue, followed by a discussion of damages. In\nview of the multiple theories of liability and categories of\nphotos, following is an overview of the opinion.\n\n\n\n\n 1\n In connection with these proceedings, we received amicus curiae\nbriefs from a broad array of interested parties, including nonprofit groups\nand associations representing a diverse set of industry, technology, and\nartistic interests. The briefs were helpful to our understanding of the\nimplications of this case from various points of view. We thank amici\nfor their participation.\n\f VHT, INC. V. ZILLOW GROUP 7\n\n I. Direct Infringement\n A. Direct Infringement—Listing Platform Photos\n B. Direct Infringement—Digs Photos\n 1. Jury Verdict—Direct Infringement\n 2. Summary Judgment—Fair Use re Searchable\n Photos\n a. Background on Fair Use\n b. Evolution of Search Engine Cases\n c. Application of Fair Use Principles\n II. Secondary Infringement—Digs\n A. Contributory Liability\n B. Vicarious Liability\n III. Damages\n A. Compilation\n B. Willfulness\n IV. Conclusion\n\n ANALYSIS\n\n The heart of this dispute is Zillow’s copyright liability\nfor use of VHT photos. VHT argues that Zillow directly\ninfringed its copyrighted photos, both those on the Listing\nPlatform and Digs. VHT also argues that Zillow indirectly\ninfringed through use of the photos on Digs. These claims\npertain to different images, focus on different features of\nZillow’s website, and have different procedural postures, so\nwe consider the various categories of photos separately.\n\n I. Direct Infringement\n\n VHT’s key claim is that Zillow is directly liable for\ninfringing VHT’s copyright on photos that were posted on\nthe Listing Platform and Digs. To prevail on a claim of\ndirect copyright infringement, VHT must establish\n“ownership of the allegedly infringed material” and that\nZillow “violate[d] at least one exclusive right granted to”\n\f8 VHT, INC. V. ZILLOW GROUP\n\nVHT under 17 U.S.C. § 106. A&M Records, Inc. v. Napster,\nInc., 239 F.3d 1004, 1013 (9th Cir. 2001). It is undisputed\nthat VHT is the copyright holder of the allegedly infringed\nphotos and therefore has the exclusive right to reproduce,\nadapt, and display them. 2 17 U.S.C. § 106.\n\n VHT must also establish causation, which is commonly\nreferred to as the “volitional-conduct requirement.” See\nPerfect 10, Inc. v. Giganews, Inc., 847 F.3d 657, 666 (9th\nCir. 2017). As we set out in Giganews—decided on the first\nday of the VHT/Zillow trial and the closest circuit precedent\non point—“volition in this context does not really mean an\nact of willing or choosing or an act of deciding”; rather, “it\nsimply stands for the unremarkable proposition that\nproximate causation historically underlines copyright\ninfringement liability no less than other torts.” Id. (internal\ncitations omitted). Stated differently, “direct liability must\nbe premised on conduct that can reasonably be described as\nthe direct cause of the infringement.” Id. (citation omitted).\nThis prerequisite takes on greater importance in cases\ninvolving automated systems, like the Zillow website.\n\n In addressing this concept, Justice Scalia noted that\n“[e]very Court of Appeals to have considered an automated-\nservice provider’s direct liability for copyright infringement\nhas adopted [the volitional-conduct] rule.” Am. Broad. Cos.,\nInc. v. Aereo, Inc., 573 U.S. 431, 453 (2014) (Scalia, J.,\ndissenting). 3 He went on to explain that while “most direct-\n\n 2\n VHT does not appeal the district court’s finding that there was\ninsufficient evidence that Zillow violated VHT’s distribution rights.\n 3\n Although the majority opinion in Aereo does not reference the\nvolitional-conduct requirement, Justice Scalia’s dissent offers instructive\nbackground on the doctrine. In Giganews, we embraced the principle\n\f VHT, INC. V. ZILLOW GROUP 9\n\ninfringement cases” do not present this issue, “it comes right\nto the fore when a direct-infringement claim is lodged\nagainst a defendant who does nothing more than operate an\nautomated, user-controlled system. . . . Most of the time that\nissue will come down to who selects the copyrighted\ncontent: the defendant or its customers.” Id. at 454–55\n(internal citations omitted).\n\n Giganews, Aereo, and out-of-circuit precedent counsel\nthat direct copyright liability for website owners arises when\nthey are actively involved in the infringement. “‘[T]he\ndistinction between active and passive participation’” in the\nalleged infringement is “‘central’” to the legal analysis.\nGiganews, 847 F.3d at 667 (quoting Fox Broad. Co. v. Dish\nNetwork LLC, 160 F. Supp. 3d 1139, 1160 (C.D. Cal. 2015)).\n\n That “direct” infringement requires “active”\ninvolvement is hardly surprising, given the correlation\nbetween the words “active” and “direct.” As the Fourth\nCircuit held, “[t]here must be actual infringing conduct with\na nexus sufficiently close and causal to the illegal copying\nthat one could conclude that the machine owner himself\ntrespassed on the exclusive domain of the copyright owner.”\nCoStar Grp., Inc. v. LoopNet, Inc., 373 F.3d 544, 550 (4th\nCir. 2004). By contrast, activities that fall on the other side\nof the line, such as “‘automatic copying, storage, and\ntransmission of copyrighted materials, when instigated by\nothers, do[] not render an [Internet service provider] strictly\nliable for copyright infringement[.]’” Giganews, 847 F.3d at\n670 (quoting CoStar, 373 F.3d at 555).\n\n\n\nand held that it is “consistent with the Aereo majority opinion,” which\nleft the requirement “intact.” 847 F.3d at 666–67.\n\f10 VHT, INC. V. ZILLOW GROUP\n\n In other words, to demonstrate volitional conduct, a party\nlike VHT must provide some “evidence showing [the alleged\ninfringer] exercised control (other than by general operation\nof [its website]); selected any material for upload, download,\ntransmission, or storage; or instigated any copying, storage,\nor distribution” of its photos. Id. at 666, 670. VHT failed to\nsatisfy that burden with respect to either the photos on the\nListing Platform or on Digs.\n\n A. Direct Infringement—Listing Platform\n Photos\n\n VHT asserted that Zillow directly infringed the photos\ndisplayed on the Listing Platform after a real estate property\nwas sold because VHT’s license agreements only authorized\nuse of those photos in relation to the sale of the property.\nThis claim, involving 54,257 non-searchable photos, was\nresolved on summary judgment. The Listing Platform is the\ncore of Zillow’s online real estate marketplace. It features\nphotos and information about properties, which Zillow\nreceives through digital feeds from real estate agents,\nbrokerages, and multiple listing services, among others\n(collectively “feed providers”).\n\n Zillow has agreements with its feed providers granting it\nan express license to use, copy, distribute, publicly display,\nand create derivative works from the feed data on its\nwebsites. Feed providers represent that they “ha[ve] all\nnecessary rights and authority to enter into” the agreements,\nand that “Zillow’s exercise of the rights granted [t]hereunder\nwill not violate the intellectual property rights, or any other\nrights of any third party.”\n\n These agreements provide Zillow with either\n“evergreen” or “deciduous” rights in the photos provided\nthrough the feeds. An evergreen right permits use of a photo\n\f VHT, INC. V. ZILLOW GROUP 11\n\nwithout any time restriction, “on and in connection with the\noperation, marketing and promotion of the web sites and\nother properties, owned, operated or powered by Zillow or\nits authorized licensees.” By contrast, a deciduous right is\ntemporally limited: Zillow may use the photo when the real-\nestate listing for its corresponding property is active, but\nonce the listing is removed (for example, when the property\nsells), the photo must be taken down from Zillow’s websites.\nTo treat each photo consistently with its deciduous or\nevergreen designation, Zillow developed automated\n“trumping” rules to determine which photos to display on the\nListing Platform.\n\n VHT argues that Zillow “designed its system to . . .\ncause[] the reproduction, display, and adaptation of VHT\nphotographs post-sale on the Listing Platform,” and “chose\nto simply ignore VHT’s notices that post-sale use was\nbeyond the scope of VHT’s licenses.” The district court\ngranted summary judgment to Zillow, concluding that it did\nnot engage in volitional conduct and therefore did not\ndirectly infringe VHT’s copyrights in 54,257 photos by\ndisplaying them on the Listing Platform after a real estate\nproperty was sold.\n\n On de novo review, we agree with the district court’s\nanalysis and affirm. Kelly v. Arriba Soft Corp., 336 F.3d\n811, 817 (9th Cir. 2003). Although the district court did not\nhave the benefit of Giganews at the time of summary\njudgment, its careful reasoning was prescient in invoking the\nsame principles.\n\n Zillow did not engage in volitional conduct necessary to\nsupport a finding of direct liability. The content of the\nListing Platform is populated with data submitted by third-\nparty sources that attested to the permissible use of that data,\n\f12 VHT, INC. V. ZILLOW GROUP\n\nand Zillow’s system for managing photos on the Listing\nPlatform was constructed in a copyright-protective way.\nThe feed providers themselves select and upload every\nphoto, along with the evergreen or deciduous designations,\nthat wind up on the Listing Platform. As a result, the photos\non the Listing Platform were not “selected” by Zillow. See\nGiganews, 847 F.3d at 670. Nor did Zillow “exercise[]\ncontrol” over these photos beyond the “general operation of\n[its website].” Id. Zillow required feed providers to certify\nthe extent of their rights to use each photo. Consistent with\nthese designations, Zillow’s system classified each photo as\ndeciduous or evergreen and programmed its automated\nsystems to treat each photo consistently with that scope of\nuse certified to by the third party.\n\n Further, when multiple versions of the same photo were\nsubmitted through the various feeds, Zillow invoked its\ncopyright-protective “trumping” rules. For example, one\nrule might prefer a photo provided by an agent over one\nprovided by a multiple listing service, and another might\nprefer a local broker to an international one. Zillow used a\nrule that gave preference to photos with evergreen rights\nover photos with deciduous rights in the same image. As the\ndistrict court recognized, “trumping” is a reasonable way to\ndesign a system to manage multiple versions of the same\nphoto when the authorized use varies across versions. These\nrules, along with other features of the system, facilitate\nkeeping the photos with evergreen rights on the website and\nremoving the photos with deciduous rights once a property\nhas sold. Thus, Zillow actively designed its system to avoid\nand eliminate copyright infringement.\n\n Notably, VHT’s argument is primarily cast in terms of\nZillow facilitating or enabling infringement by VHT’s\nclients that are Zillow’s feed providers. But this type of\n\f VHT, INC. V. ZILLOW GROUP 13\n\nclaim more properly falls in the category of secondary\ninfringement, a claim not advanced by VHT with respect to\nthe Listing Platform photos.\n\n VHT also asserts that Zillow failed to remove photos\nonce it received notice that infringing content was on the\nListing Platform, a conscious choice that amounts to\nvolitional conduct on Zillow’s part. This claim is unavailing\nbecause, once VHT put Zillow on notice of claimed\ninfringement, Zillow took affirmative action to address the\nclaims. Additionally, VHT’s assertion that it “repeatedly\nnotified Zillow that it was infringing” is unsupported in the\nrecord.\n\n In July 2014, VHT sent Zillow a takedown notice letter\nwith a list of thousands of allegedly infringing photos by\nresidential street address (but not by web address). Zillow\npromptly requested all executed license agreements between\nVHT and the feed providers who had provided photos to\nZillow, as well as license agreements between VHT and its\nphotographers, so that Zillow could evaluate whether VHT\npossessed exclusive rights to the photos on the Listing\nPlatform. VHT responded with an unsigned form contract,\nwhich it stated was used with many feed providers, but\nwhich was not tied to any specific photos on Zillow’s\nwebsite. Zillow again reiterated its need to see the specific\ncontracts governing the contested photos. Instead of\nresponding with the contracts, VHT filed suit. Zillow’s\nreasonable response to VHT’s single formal inquiry\n(supplemented in a follow-on email) can hardly be\ncharacterized as rising to the level of volitional conduct or\nturning a blind eye.\n\n In sum, VHT failed to “provide[] . . . evidence showing\n[Zillow] exercised control (other than by general operation\n\f14 VHT, INC. V. ZILLOW GROUP\n\nof [its website]); selected any material for upload, download,\ntransmission, or storage; or instigated any copying, storage,\nor distribution” of these photos. See Giganews, 847 F.3d at\n670; see also CoStar, 373 F.3d at 555. Thus, we affirm the\ndistrict court and conclude Zillow did not directly infringe\nVHT’s copyrights in photos displayed on the Listing\nPlatform post-sale.\n\n B. Direct Infringement—Digs Photos\n\n VHT also claimed that Zillow directly infringed\nthousands of photos used on Digs. The jury concluded that\nZillow directly infringed 28,125 photos and rejected its fair\nuse defense. Following trial, the district court granted in\nsubstantial part Zillow’s motion for judgment\nnotwithstanding the verdict on the ground that insufficient\nevidence supported Zillow’s direct infringement of 22,109\nphotos that were not displayed on Digs and 2,093 4 photos\nthat were displayed but not searchable on Digs.\n\n By contrast, the court upheld the jury’s determination\nthat Zillow directly infringed a set of 3,921 images that were\nselected and tagged by Zillow moderators for searchable\nfunctionality and displayed on Digs. Zillow does not appeal\nthis ruling. However, Zillow argues that fair use insulates it\nfrom liability as to this subset of photos. The jury was\ninstructed not to consider this legal theory as to these photos\nbecause the district court had determined pretrial that, as a\nmatter of law, the searchability function did not constitute\nfair use. It is that summary judgment ruling that Zillow\nchallenges on appeal. Because we agree with the district\n\n 4\n The displayed but non-searchable set includes 2,094 photos. The\ndistrict court affirmed the jury verdict with respect to one of those\nphotos, which Zillow also distributed via email.\n\f VHT, INC. V. ZILLOW GROUP 15\n\ncourt that the fair use defense does not absolve Zillow of\ndirect liability for these searchable photos, this portion of the\njury verdict remains intact.\n\n The following chart clarifies the status of the Digs photos\nrelevant to the direct infringement claims.\n\n Photos Jury Verdict Post-Trial Party\n Determination Bringing\n Appeal\n22,109 not Direct Overturned jury VHT\ndisplayed 5 infringement verdict\n2,094 Direct Overturned jury VHT\ndisplayed, not infringement verdict\nsearchable (except 1 email\n photo)\n3,921 Direct Upheld jury Zillow:\ndisplayed, infringement verdict direct\nsearchable infringement\n not appealed;\n appeals only\n summary\n judgment on\n no fair use\n1 blog post Direct Upheld jury Not appealed\nphoto (not on infringement verdict\nDigs)\n\n\n\n\n 5\n Searchable photos numbered 1,694; 20,415 photos were not\nsearchable.\n\f16 VHT, INC. V. ZILLOW GROUP\n\n 1. Jury Verdict—Direct Infringement\n\n VHT’s claim that Zillow directly infringed photos on\nDigs went to the jury. The jury verdict form was framed in\ngeneral terms, asking only whether “VHT has proven its\ndirect copyright infringement claim as to one or more of the\nVHT Photos[.]” The jury answered “yes,” and was asked to\nspecify how many VHT photos were directly infringed. The\njury answered “28,125”—in other words, all of them.\nHowever, the jury was not asked to specify which copyright\nrights—display, reproduction, or adaption—were\n 6\ninfringed. Following trial, Zillow moved for judgment\nnotwithstanding the verdict or for a new trial. The task fell\nto the district court to examine the evidence as to each right.\n\n On de novo review, “we apply the same standard used\nby the district court in evaluating the jury’s verdict” and\nuphold the verdict unless “the evidence permits only one\nreasonable conclusion, and that conclusion is contrary to the\njury’s verdict.” Wallace v. City of San Diego, 479 F.3d 616,\n624 (9th Cir. 2007); Fed. R. Civ. P. 50(a). Specifically, we\n“ask[] whether the verdict is supported by substantial\nevidence,” “which is evidence adequate to support the jury’s\nconclusion, even if it is also possible to draw a contrary\nconclusion.” Unicolors, Inc. v. Urban Outfitters, Inc., 853\nF.3d 980, 984, 991 (9th Cir. 2017) (quotation and citation\nomitted). Given the sanctity of the jury process, we\nundertake this review with special care and reluctance to\noverturn a verdict. However, because the verdict here did\nnot meet this standard, we affirm the district court’s grant of\nZillow’s motion for judgment notwithstanding the verdict\n\n\n 6\n As noted earlier, the distribution right was not contested and is not\nan issue on appeal.\n\f VHT, INC. V. ZILLOW GROUP 17\n\nwith respect to direct infringement of 22,109 non-displayed\nphotos and 2,093 displayed but not searchable photos.\n\n We first consider display rights. As background for our\nanalysis, it is useful to consider the direct infringement by\nZillow that the district court upheld and that Zillow did not\nappeal. The district court found substantial evidence that\nZillow directly infringed VHT’s display rights in 3,921\nphotos displayed on Digs that Zillow moderators selected\nand tagged for searchable functionality. Based on testimony,\ncharts, and statistics, the court found that “the jury could\nhave reasonably concluded that users accessed those images\nthrough Digs’s search function.” The court went on to\nreason that “the jury could have reasonably concluded that\nZillow’s moderation efforts, which rendered those images\nsearchable, proximately caused the copying.” Put\ndifferently, active conduct by Zillow met the volitional-\nconduct requirement for direct infringement. Zillow does\nnot appeal this ruling upholding the jury’s verdict as to the\n3,921 displayed, searchable photos.\n\n On appeal, VHT attempts to shoehorn an additional\n1,694 photos into this category. This effort falls flat both as\na factual and legal matter because substantial evidence does\nnot support direct infringement of VHT’s display rights in\nthe 1,694 searchable images that were not displayed.\n\n VHT posits that the jury could have found that these\nphotos were displayed because of circumstantial evidence\nand because Zillow failed to record whether they were\ndisplayed. Not so. This argument is foreclosed by the\nparties’ stipulated spreadsheet that categorized each photo.\nThe column labeled “DISPLAYED” included an entry for\n“Y” (yes) or “N” (no). The jury was instructed to “treat\nevery fact on this spreadsheet as proven,” so VHT cannot\n\f18 VHT, INC. V. ZILLOW GROUP\n\nrecast the facts retroactively and now claim that 1,694\nsearchable images stipulated as “N[OT] DISPLAYED”\nwere in fact displayed or made available for display. Up is\nnot down.\n\n VHT’s contention that the jury could have reasonably\ninferred that Zillow made “available for public display” all\n22,109 “N[OT] DISPLAYED” images similarly fails. This\ntheory presumes that the Copyright Act’s display right\nencompasses an exclusive right to “make available for\ndisplay,” a position neither supported by the statute nor\nembraced by this court. See Perfect 10, Inc. v. Amazon.com,\nInc., 508 F.3d 1146, 1160 (9th Cir. 2007) (“[B]ased on the\nplain language of the statute, a person displays a\nphotographic image by using a computer to fill a computer\nscreen with a copy of the photographic image fixed in the\ncomputer’s memory.”); 17 U.S.C. § 101 (“To ‘display’ a\nwork means to show a copy of it, either directly or by means\nof a film, slide, television image, or any other device or\nprocess or, in the case of a motion picture or other\naudiovisual work, to show individual images\nnonsequentially.”). To be sure, the Copyright Office notes\nthat the outer limits of the public display right have yet to be\ndefined. U.S. COPYRIGHT OFFICE, THE MAKING AVAILABLE\nRIGHT IN THE UNITED STATES 47–51 (Feb. 2016),\nhttps://www.copyright.gov/docs/making_available/making-\navailable-right.pdf.\n\n What is most important here, though, is that VHT’s\nargument comes too late. The jury was never instructed on\nthe “made available” theory, nor did VHT raise this issue in\nits proposed jury instructions or in objections to the final\ninstructions. See Sinclair v. Long Island R.R., 985 F.2d 74,\n78 (2d Cir. 1993) (“the verdict . . . cannot be sustained on a\ntheory that was never presented to the jury”); Ramona Equip.\n\f VHT, INC. V. ZILLOW GROUP 19\n\nRental, Inc. ex rel. U.S. v. Carolina Cas. Ins. Co., 755 F.3d\n1063, 1070 (9th Cir. 2014) (an argument raised for first time\nin a post-trial motion is waived). For these reasons,\nsubstantial evidence did not support a finding of direct\ninfringement of VHT’s public display rights in these 22,109\nphotos.\n\n Zillow also did not violate VHT’s display rights in 2,093\ndisplayed, non-searchable photos. These are photos that\nDigs users copied to “personal boards” and “Implicit Digs,”\nbut which Zillow did not add to the searchable set.\n\n A “personal board” is a bespoke digital bulletin board of\nimages that a user saves or uploads from the Listing\nPlatform. Users can also upload their own images. These\nboards are typically private, though users can share a link to\ntheir boards. When a user saves a copy of an image with\nevergreen rights to a personal board, that image\nautomatically joins a queue for review by a Zillow\nmoderator. The moderator then decides whether to\ndesignate the photo for tagging so that it can be searchable\non Digs, and thus select it for public display. Not all photos\nthat are in the queue for moderation are reviewed.\nAdditionally, in some instances, a user starts but does not\nfinish the process of saving an image to this board.\nBeginning in 2014, Zillow programmed these “clicked to\nsave” images—called “Implicit Digs”—to enter the queue\nfor moderator review in the same manner as if the photo had\nbeen saved to a personal board.\n\n According to VHT, the jury could have found that Zillow\n“caused the [2,093] images to be displayed . . . by subjecting\nthe non-searchable VHT Photos to the potential for\nmoderation.” This argument defies logic because the only\ndisplay that occurred was triggered by the user. Any\n\f20 VHT, INC. V. ZILLOW GROUP\n\npotential for future display is purely speculative. As the\ndistrict court explained, the possibility that images might be\nmoderated and tagged—conduct that is volitional—is not\nsufficient “to transform Zillow from a ‘passive host’ to a\n‘direct[] cause’ of the display of VHT’s images.”\n\n Next, we consider VHT’s exclusive reproduction and\nadaption rights in the 2,093 displayed, non-searchable\nphotos. The Copyright Act grants copyright holders the\nexclusive right “to reproduce the copyrighted work in copies\nor phonorecords.” 17 U.S.C. § 106(1). Direct infringement\nof the reproduction right “requires copying by the defendant,\n. . . which [requires] that the defendant cause the copying.”\nFox Broad. Co., Inc. v. Dish Network LLC, 747 F.3d 1060,\n1067 (9th Cir. 2014) (quotation and citation omitted). The\nadaptation right is the exclusive right “to prepare derivative\nworks based upon the copyrighted work.” 17 U.S.C.\n§ 106(2). Following Giganews, we conclude that Zillow’s\nautomated processes storing or caching VHT’s photos are\ninsufficiently volitional to establish that Zillow directly\ninfringed VHT’s reproduction and adaption rights in these\nnon-searchable photos. Unlike photos that Zillow curated,\nselected, and tagged for searchable functionality—activities\nthat amount to volitional conduct establishing direct\nliability—these 2,093 photos were copied to “personal\nboards” and “Implicit Digs” based on user actions, not the\nconduct of Zillow or its moderators.\n\n Any volitional conduct with respect to these photos was\ntaken by the users, not Zillow. Users, not Zillow,\n“selecte[d]” images to add to their personal boards and\n“instigate[d]” the automatic caching process by saving a\nparticular image. See Giganews, 847 F.3d at 670. This\narrangement is important because courts have found no\ndirect liability where an online system “responds\n\f VHT, INC. V. ZILLOW GROUP 21\n\nautomatically to users’ input . . . without intervening\nconduct” by the website owner. CoStar, 373 F.3d at 550;\nsee also Giganews, 847 F.3d at 670. Under these conditions,\ncourts have analogized online facilities, like Internet service\nproviders, to a copy machine owner, who is not liable\n“[w]hen a customer duplicates an infringing work.” CoStar,\n373 F.3d at 550.\n\n Additionally, Zillow’s behind-the-scenes technical work\non Digs photos is not evidence that Zillow “selected any\nmaterial for upload, download, transmission, or storage.”\nGiganews, 847 F.3d at 670. Zillow produced cached copies\nof these Digs images, a process that automatically trims or\npads images whose height and width did not match the target\nresolution, for the purposes of accelerating website speed.\nThis activity does not amount to volitional conduct. Nor can\nZillow’s promotion of Digs, including encouraging users to\nshare photos through its site, be seen as “instigat[ing]” user\ncopying. Id.\n\n Zillow’s conduct with respect to these photos amounts\nto, at most, passive participation in the alleged infringement\nof reproduction and adaption rights and is not sufficient to\ncross the volitional-conduct line. As in cases involving\nInternet service providers, Zillow “affords its [users] an\nInternet-based facility on which to post materials, but the\nmaterials posted are of a type and kind selected by the [user]\nand at a time initiated by the [user].” CoStar, 373 F.3d at\n555. Zillow did not directly infringe VHT’s reproduction\n\f22 VHT, INC. V. ZILLOW GROUP\n\nand adaptation rights in the 2,093 displayed, non-searchable\nphotos. 7\n\n * * *\n\n In sum, VHT did not present substantial evidence that\nZillow, through the Digs platform, directly infringed its\ndisplay, reproduction, or adaption rights in 22,109 not\ndisplayed photos and 2,093 displayed but non-searchable\nphotos. We affirm the district court’s grant of judgment\nnotwithstanding the verdict as to these photos. We now turn\nto Zillow’s fair use defense to direct infringement of 3,921\ndisplayed, searchable photos.\n\n 2. Summary Judgment—Fair Use re\n Searchable Photos\n\n Zillow does not appeal the jury’s finding of direct\ninfringement with respect to the 3,921 displayed, searchable\nphotos, but does assert a fair use defense for those photos.\nZillow contends that Digs’ searchable functionality\nconstitutes fair use, which the district court rejected as a\nmatter of law at summary judgment.\n\n We recount the somewhat unusual history of the fair use\nissue in the proceedings below. On summary judgment, the\ndistrict court rejected Zillow’s argument that “the images\nthat it has made searchable on Digs” are protected by fair use\nand instead “conclude[d] as a matter of law that Digs’\nsearchable functionality does not constitute a fair use.” At\ntrial, the jury was generally instructed to consider the fair use\n\n 7\n The same analysis applies to any potential violation of VHT’s\nexclusive reproduction and adaption rights in the 22,109 photos that were\nnot displayed.\n\f VHT, INC. V. ZILLOW GROUP 23\n\ndefense as to all VHT photos used on Digs that it found\nZillow had directly or indirectly infringed. However, this set\nof photos was carved out for separate treatment. The jury\nconsidered only whether “reproduction, cropping, and\nscaling” of these photos constituted fair use because the\ncourt instructed the jury that the court “ha[d] determined,\nand you are to take as proven, that the Digs searchable\nfunctionality does not constitute fair use.” After finding that\nZillow directly infringed all 28,125 VHT photos used on\nDigs, the jury rejected Zillow’s fair use affirmative defense\nfor all photos.\n\n The district court upheld the jury’s fair use verdict,\nwhich Zillow does not appeal. Rather, Zillow’s appeal\nreaches back to the district court’s summary judgment ruling\nto argue that the Digs’ searchable functionality is fair use as\na matter of law, and, as a result, Zillow bears no liability for\nthe 3,921 searchable and displayed photos. We review de\nnovo the district court’s grant of summary judgment to VHT\non this mixed question of law and fact. Kelly, 336 F.3d at\n817.\n\n a. Background on Fair Use\n\n Protection of copyrighted works is not absolute. “The\nfair use defense permits the use of copyrighted works\nwithout the copyright owner’s consent under certain\nsituations.” Amazon, 508 F.3d at 1163. Fair use both fosters\ninnovation and encourages iteration on others’ ideas, “thus\nproviding a necessary counterbalance to the copyright law’s\ngoal of protecting creators’ work product.” Id.; see\nCampbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 575–76\n(1994). Fair use also aligns with copyright’s larger purpose\n“‘[t]o promote the Progress of Science and useful Arts,’ . . .\nand to serve ‘the welfare of the public.’” Amazon, 508 F.3d\n\f24 VHT, INC. V. ZILLOW GROUP\n\nat 1163 (quoting U.S. Const. art. I, § 8, cl. 8, and Sony Corp.\nof Am. v. Universal City Studios, Inc., 464 U.S. 417, 429 n.10\n(1984)).\n\n Fittingly enough, a case involving a biography of George\nWashington serves as a foundational source of fair use in\nAmerican law. Folsom v. Marsh, 9 F. Cas. 342 (D. Mass.\n1841) (No. 4901). Justice Story’s narrative description of\ncopyright doctrine in that case “distilled the essence of law\nand methodology from the earlier cases” and provided the\nconceptual basis for the judge-made fair use doctrine.\nCampbell, 510 U.S. at 576. Although the 1976 Copyright\nAct codified those principles, it did little to elaborate on\nJustice Story’s description or to clarify application of the\nfactors. With minimal guidance or elucidation, Congress set\nforth four factors for courts to consider when determining\nwhether the use of a copyrighted work is a “fair use”:\n\n (1) the purpose and character of the use,\n including whether such use is of a\n commercial nature or is for nonprofit\n educational purposes;\n\n (2) the nature of the copyrighted work;\n\n (3) the amount and substantiality of the\n portion used in relation to the copyrighted\n work as a whole; and\n\n (4) the effect of the use upon the potential\n market for or value of the copyrighted\n work.\n\n17 U.S.C. § 107.\n\f VHT, INC. V. ZILLOW GROUP 25\n\n Given license to apply these factors flexibly and to\nconsider them in their totality, courts have been bedeviled by\nthe fair use inquiry. See Campbell, 510 U.S. at 577–78. Fair\nuse has been called “the most troublesome [doctrine] in the\nwhole law of copyright” and commentators have criticized\nthe factors as “billowing white goo.” Monge v. Maya\nMagazines, Inc., 688 F.3d 1164, 1170–71 (9th Cir. 2012)\n(citations and quotations omitted). In a (still ongoing) effort\nto adapt the fair use analysis to a myriad of circumstances,\ncourts have embellished and supplemented the factors. For\nexample, the concept of “transformativeness” is found\nnowhere in the statute, but appeared for the first time in the\nSupreme Court in Campbell, where the Court endeavored to\nrefine and crystalize the first statutory factor: the “purpose\nand character of the use.” 8 510 U.S. at 579. The animating\npurpose of the first factor is to determine,\n\n in Justice Story’s words, whether the new\n work merely ‘supersede[s] the objects’ of the\n original creation . . . or instead adds\n something new, with a further purpose or\n different character, altering the first with new\n expression, meaning, or message; it asks, in\n other words, whether and to what extent the\n new work is “transformative.”\n\n\n\n 8\n The concept of transformative use had appeared in earlier lower\ncourt decisions, e.g., Twin Peaks Prods., Inc. v. Publ’ns Intern., Ltd., 996\nF.2d 1366, 1375 (2d Cir. 1993) (“[I]t is not uncommon for works serving\na fair use purpose to give at least a brief indication of the plot. . . . In\nidentifying plot, the author of the second work may or may not be said\nto have made . . . a ‘transformative’ use. . . . Such use would occur, for\nexample, if a plot was briefly described for purposes of adding\nsignificant criticism or comment about the author’s plotting technique.”).\n\f26 VHT, INC. V. ZILLOW GROUP\n\nId. (citations omitted). Because transformation advances\ncopyright’s core goals, “the more transformative the new\nwork, the less will be the significance of other factors.” Id.\nLikewise, despite the absence of a textual hook, public\npurpose also has been read into the statute. See Amazon, 508\nF.3d at 1166; Kelly, 336 F.3d at 820. While we can discern\ncertain animating principles bridging cases in this area, the\ndoctrine has hardly followed a straight, or even slightly\ncurved, line.\n\n The focus of the parties’ debate here is whether Zillow’s\ntagging of 3,921 VHT photos for searchable functionality on\nDigs was transformative and thus supported a finding of fair\nuse. The purpose of Digs is to permit users to search for\ncertain attributes or features, such as a marble countertop or\nhardwood floor, and view photos of rooms with those\nattributes or features. These photos are either uploaded by\nusers to Digs, or selected manually or electronically by\nZillow. Zillow then tags the photos to make them\nsearchable. Of course, tagging makes it possible for a user’s\nkeyword search to produce relevant results. Zillow refers to\nthese tagged photos as “searchable images” or components\nof the “searchable set.” VHT’s 3,921 photos are in the\nsearchable set.\n\n Zillow contends that Digs is effectively a search engine,\nwhich makes its use of VHT’s photos transformative, and\ntherefore fair use. VHT responds that this is not fair use\nbecause Digs is not a search engine and the tagging for\nsearchable functionality is not transformative. Dueling\n“search engine” characterizations do not resolve fair use\nhere. Instead, we step back and assess the question\nholistically, as we have been instructed to do by the statute\nand the Supreme Court. We consider the reality of what is\n\f VHT, INC. V. ZILLOW GROUP 27\n\nhappening rather than resorting to labels. To do that, it is\nhelpful to recount the history of the search engine cases.\n\n b. Evolution of Search Engine Cases\n\n Over the past two decades, search engines have emerged\nas a significant technology that may qualify as a\ntransformative fair use, making images and information that\nwould otherwise be protected by copyright searchable on the\nweb. See Amazon, 508 F.3d at 1166–67; Kelly, 336 F.3d at\n818–22. In assessing fair use in the context of search\nengines, courts have relied heavily on the first fair use factor,\nand in particular “whether and to what extent the new work\nis transformative.” Campbell, 510 U.S. at 579 (citation and\nquotation omitted); see also Amazon, 508 F.3d at 1164\n(explaining the Kelly court relied “primarily” on the first fair\nuse factor when conducting its analysis); Authors Guild v.\nGoogle, Inc., 804 F.3d 202, 220–221, 223 (2d Cir. 2015)\n(offering a relatively abbreviated consideration of the\nremaining three fair use factors, all of which were informed\nby its analysis of the first factor).\n\n In an early opinion applying fair use principles in the\ndigital age, we held that the now-defunct search engine\nArriba’s creation and use of thumbnail versions of a\nprofessional photographer’s copyrighted images was fair use\nbecause the “smaller, lower-resolution images . . . served an\nentirely different function than [the] original images.” Kelly,\n336 F.3d at 815, 818. The original images served an artistic\nor aesthetic purpose. Id. at 819. By contrast, the thumbnail\nimages, which were provided in response to a user’s search\nquery, were incorporated into the search engine’s overall\nfunction “to help index and improve access to images on the\ninternet and their related web sites.” Id. at 818. Investing\nthe images with a new purpose made Arriba’s use\n\f28 VHT, INC. V. ZILLOW GROUP\n\ntransformative, not superseding. Indeed, the thumbnail\nversions could not supersede the original use because the\nthumbnails were grainy and low-resolution when enlarged.\nId. Additionally, Arriba’s use of the thumbnail images\n“promote[d] the goals of the Copyright Act and the fair use\nexception” because they “benefit[ed] the public by\nenhancing information-gathering techniques on the\ninternet.” Id. at 820. Just as Campbell had drawn out the\nprinciple of transformation from the first statutory factor, we\ndrew out the principle of public benefit.\n\n Building on our reasoning in Kelly, in Amazon we held\nthat Google’s use of thumbnail images in its search engine\nis “highly transformative” and thus fair use. 508 F.3d at\n1163–65. As in Kelly, we concluded that “a search engine\nprovides social benefit by incorporating an original work\ninto a new work, namely, an electronic reference tool.” Id.\nat 1165. On a scale much greater than the search engine at\nissue in Kelly, Google “improve[s] access to images on the\ninternet and their related web sites” by “index[ing]” the\ninternet and linking to the original source image generated\nin the search results. Kelly, 336 F.3d at 815–16, 818. By\nusing the thumbnail images in service of the search engine,\nGoogle “transforms the image,” which might have been\ncreated for an “entertainment, aesthetic, or informative\nfunction,” “into a pointer directing a user to a source of\ninformation.” Amazon, 508 F.3d at 1165. As a result of the\nnew function that the image serves, Google’s use of the\nentire image in its search engine results “does not diminish\nthe transformative nature of [its] use.” Id. And, further\ndeveloping the public benefit principle from Kelly, we\nemphasized that Google’s search engine both “promotes the\npurposes of copyright and serves the interests of the public,”\nwhich significantly outweighed the superseding or\n\f VHT, INC. V. ZILLOW GROUP 29\n\ncommercial uses of the search engine, and strongly\nsupported finding fair use. Id. at 1166.\n\n More recently, the Second Circuit considered whether\nfair use protected the Google Books search engine, which\nemploys digital, machine-readable copies of millions of\ncopyright-protected books scanned by Google. Authors\nGuild, 804 F.3d at 207–08. 9 The Google Books search\nengine enables a full text search, which makes possible\nsearching for a specific term, and then provides “snippets,”\nor a part of a page, for users to read. Id. at 208–09, 216–17.\nThe court held that both functions involve a “highly\ntransformative purpose of identifying books of interest to the\nsearcher.” Id. at 218. The search function “augments public\nknowledge by making available information about\nPlaintiffs’ books without providing the public with a\nsubstantial substitute.” Id. at 207. And the search engine\nmakes possible a new type of research known as “text\nmining” or “data mining,” whereby users can search across\nthe corpus of books to determine the frequency of specified\nterms across time. Id. at 209, 217. Additionally, the\n“snippet” view provides context for users to assess if a book\nis relevant to them, without providing so much context as to\nsupersede the original. Id. at 218. To boot, Google often\nprovides a link to a page where the entire book can be found\nat a library or purchased. Id. at 209. Concluding that\nGoogle’s commercial motivation did not significantly\noutweigh these transformative uses, the court held that the\nfirst factor strongly supported a finding of fair use. Id. at\n219.\n\n\n 9\n The Google Books search engine also featured millions of public\ndomain texts. For obvious reasons, fair use was not at issue with those\nworks.\n\f30 VHT, INC. V. ZILLOW GROUP\n\n What we divine from these cases is that the label “search\nengine” is not a talismanic term that serves as an on-off\nswitch as to fair use. Rather, these cases teach the\nimportance of considering the details and function of a\nwebsite’s operation in making a fair use determination. We\nnow examine Digs with those lessons in mind.\n\n c. Application of Fair Use Principles\n\n As noted, the first factor assesses the character of the use,\nincluding whether it is commercial in nature, and, critically,\nwhether it is “transformative.” There is no dispute that\nZillow’s use is for commercial purposes, a factor we cannot\nignore. To determine if that use is transformative, we\nconsider whether and to what extent it “adds something new,\nwith a further purpose or different character, altering the first\nwith new expression, meaning, or message[.]” Campbell,\n510 U.S. at 579. Though we agree that Digs is a type of\nsearch engine because it offers searchable functionality, it is\nqualitatively different than Google and other open-universe\nsearch engines, as well as different than the Google Books\nsearch engine.\n\n Most simply, a search engine is a software program that\nenables information retrieval by helping users find\ninformation through the use of keyword queries. But not all\nsearch engines are created equal. The search engines\ncommonly used for day-to-day research are internet-wide\nsearch engines, like Google, Yahoo, or Bing. These search\nengines are programs powered by algorithms that search or\n“crawl” the web. A search engine like Google then indexes\nwebsites, stores them on a database, and runs users’ search\nqueries against it. Search results are typically a mix of\nimages and text, which include hyperlinks to sources of that\ncontent elsewhere on the web. Amazon, 508 F.3d at 1155.\n\f VHT, INC. V. ZILLOW GROUP 31\n\n Unlike the internet-wide search engines considered in\nAmazon and Kelly, Digs is a closed-universe search engine\nthat does not “crawl” the web. Users can run searches on the\n“searchable set” of images within Digs’ walled garden,\nwhich includes VHT photos. The search results do not direct\nusers to the original sources of the photos, such as VHT’s\nwebsite. Rather, they link to other pages within Zillow’s\nwebsite and, in some cases, to third-party merchants that sell\nitems similar to those featured in the photo.\n\n That Digs makes these images searchable does not\nfundamentally change their original purpose when produced\nby VHT: to artfully depict rooms and properties.\nAdditionally, Digs displays the entire VHT image, not\nmerely a thumbnail. Unlike in Amazon, the new image does\nnot serve a “different function” than the old one. Amazon,\n508 F.3d at 1165. Zillow’s use preserves the photos’\n“inherent character.” Monge, 688 F.3d at 1176. And Zillow\n“simply supersed[es] [VHT’s] purpose” in creating the\nimages in the first place. Kelly, 336 F.3d at 819–20; see\nHarper & Row Publishers, Inc. v. Nation Enters., 471 U.S.\n539, 550–51 (1985) (holding that if a new work\n“supersede[s] the use of the original,” it is probably not a fair\nuse).\n\n Comparing Digs to the Google Books search engine\nfurther drives home this analysis. We agree with the Second\nCircuit’s observation that the copyright dispute over the\nGoogle Books search engine “tests the boundaries of fair\nuse.” Authors Guild, 804 F.3d at 206. We conclude that\nDigs goes one step further—and crosses the line.\n\n Like Digs, the Google Books search engine operates on\na closed database comprised of complete digital copies of\noriginal works. Id. at 217. But the similarities end there.\n\f32 VHT, INC. V. ZILLOW GROUP\n\nThe Google Books project makes it possible to search books\nfor “identifying information instantaneously supplied [that]\nwould otherwise not be obtainable in lifetimes of searching.”\nId. at 209. With this broad purpose in mind, the court\nrightfully observed that this system “augments public\nknowledge.” Id. at 207. This rationale bears no comparison\nto Digs.\n\n Nor is the limited transformation present on Digs\nremotely comparable to the unprecedented text mining, word\npattern, frequency of use, and other statistical analyses made\npossible for the first time by Google Books. Google Books\nsearch results provide “[a] brief description of each book,\nentitled ‘About the Book,’” as well as, for some books, links\nfor borrowing or purchasing the book. Id. at 209.\nSignificantly, a Google Books search produces only a\n“snippet” of the book, and sometimes it “disables snippet\nview entirely.” Id. at 210. At the request of a rights holder,\nGoogle “will exclude any book altogether from snippet\nview.” Id.\n\n These features, in conjunction with other creative aspects\nof Google Books, result in a categorically more\ntransformative use than Zillow’s simple tagging and query\nsystem that displays full-size copyrighted images serving the\nsame purpose as the originals, with no option to opt out of\nthe display, and with few, if any, transformative qualities.\nAny transformation by Zillow pales in comparison to the\nuses upheld in prior search engine cases. Such use also does\nnothing to further the use of copyrighted works for the\nsocially valuable purposes identified in the Copyright Act\nitself, like “criticism, comment, news reporting,\nteaching . . . , scholarship, or research.” 17 U.S.C. § 107;\nsee also 4 NIMMER ON COPYRIGHT § 13.05[A]. The lack of\ntransformation is especially significant because, as Kelly\n\f VHT, INC. V. ZILLOW GROUP 33\n\nteaches, “[t]he more transformative the new work, the less\nimportant the other factors, including commercialism,\nbecome.” 336 F.3d at 818 (citing Campbell, 510 U.S. at\n579). If, as the Second Circuit suggested, Google’s use\n“tests the boundaries of fair use,” Zillow’s efforts push Digs\ninto the outer space of fair use. Authors Guild, 804 F.3d at\n206. So while Google Books may inch across the boundary\nof fair use, Zillow’s use does not approach the line.\n\n Our decisions in Amazon and Kelly provide a roadmap\nfor analyzing the second factor, which focuses on the nature\nof the copyrighted work. In those cases, we held that\nphotographers’ images are creative, especially when they are\ncreated for public viewing. Amazon, 508 F.3d at 1167;\nKelly, 336 F.3d at 820. “Works that are creative in nature\nare ‘closer to the core of intended copyright protection’ than\nare more fact-based works.” Napster, 239 F.3d at 1016\n(quoting Campbell, 510 U.S. at 586).\n\n So too here. VHT’s photos are aesthetically and\ncreatively shot and edited by professional photographers.\nThat Zillow’s curators select the most creative photos for the\nDigs searchable set underscores the creative nature of the\nworks. But, as the district court properly noted, this factor\noperates “with less force” in favor of VHT because the\nphotos had already been published on the Listing Platform.\nSee Kelly, 336 F.3d at 820 (“The fact that a work is published\nor unpublished also is a critical element of its nature.\nPublished works are more likely to qualify as fair use\nbecause the first appearance of the artist’s expression has\nalready occurred.”). Ultimately, this factor only slightly\nfavors VHT, further cutting against finding fair use.\n\n The third factor evaluates the amount and substantiality\nof the copyrighted work that was used. “[C]opying an entire\n\f34 VHT, INC. V. ZILLOW GROUP\n\nwork militates against a finding of fair use.” Worldwide\nChurch of God v. Phila. Church of God, Inc., 227 F.3d 1110,\n1118 (9th Cir. 2000) (quotation and citation omitted).\nHowever, this analysis is informed by the “purpose of the\ncopying.” Campbell, 510 U.S. at 586. In that spirit, we have\nfound that copying full works qualifies as fair use where “[i]t\nwas necessary . . . to copy the entire image to allow users to\nrecognize the image and decide whether to pursue more\ninformation about the image or the originating web site.”\nKelly, 336 F.2d at 821. In contrast to Amazon and Kelly,\nnothing justifies Zillow’s full copy display of VHT’s photos\non Digs.\n\n Finally, the fourth factor considers “the effect of the use\nupon the potential market for or value of the copyrighted\nwork.” 17 U.S.C. § 107(4). To defeat a fair use defense,\n“one need only show that if the challenged use should\nbecome widespread, it would adversely affect the potential\nmarket for the copyrighted work.” Harper & Row\nPublishers, 471 U.S. at 568 (citation and quotation omitted).\n\n Although VHT had licensed only a handful of photos for\nsecondary uses (and none on a searchable database), that\nmarket was more than “hypothetical.” See Amazon, 508\nF.3d at 1168. Significantly, VHT was “actively exploring”\nthe market for licensing its photos to home design websites\nlike Digs—including with Zillow itself. This factor favors\nVHT.\n\n Taken together, the nature of Zillow’s use, when\nintegrated with the four factors, cuts against finding fair use\nby Zillow. We affirm the district court’s grant of summary\njudgment to VHT with respect to fair use.\n\f VHT, INC. V. ZILLOW GROUP 35\n\n II. Secondary Infringement—Digs\n\n Turning to VHT’s claim for secondary infringement, the\ndistrict court correctly concluded that Zillow did not\nsecondarily infringe VHT’s exclusive rights in the 28,125\nphotos used on Digs, aside from 114 images created on Digs\nafter VHT specifically identified them, which are not on\nappeal. As noted before with regard to the district court’s\nruling on direct infringement, “we apply the same standard\nused by the district court in evaluating the jury’s verdict” and\nuphold the verdict unless “the evidence permits only one\nreasonable conclusion, and that conclusion is contrary to the\njury’s verdict.” Wallace, 479 F.3d at 624; Fed. R. Civ. P.\n50(a). The jury’s verdict here did not meet this standard.\nWe affirm the district court’s grant of Zillow’s motion for\njudgment notwithstanding the verdict with respect to\nsecondary infringement, both contributory and vicarious\ninfringement.\n\n A. Contributory Liability\n\n Contributory liability requires that a party “(1) has\nknowledge of another’s infringement and (2) either\n(a) materially contributes to or (b) induces that\ninfringement.” Perfect 10, Inc. v. Visa Int’l Serv., Ass’n, 494\nF.3d 788, 795 (9th Cir. 2007). Zillow’s actions do not satisfy\nthe second prong—material contribution or inducement—so\nwe do not address the first prong. See Giganews, 847 F.3d\nat 671.\n\n In Giganews, we outlined the means of material\ncontribution to infringement:\n\n In the online context, . . . a “computer system\n operator” is liable under a material\n contribution theory of infringement “if it has\n\f36 VHT, INC. V. ZILLOW GROUP\n\n actual knowledge that specific infringing\n material is available using its system, and can\n take simple measures to prevent further\n damage to copyrighted works, yet continues\n to provide access to infringing works.”\n\nId. at 671 (quoting Amazon, 508 F.3d at 1172); see also\nEllison v. Robertson, 357 F.3d 1072, 1078 (9th Cir. 2004)\n(applying this standard in the online context); Napster, 239\nF.3d at 1022 (same); Religious Tech. Ctr. v. Netcom On-Line\nComm’n Servs., Inc., 907 F. Supp. 1361, 1375 (N.D. Cal.\n1995) (same). There is insufficient evidence to support the\ncontributory infringement verdict under the “simple\nmeasures” standard.\n\n VHT’s position that “the jury could have reasonably\ndecided that Zillow in fact had the means to identify and\nremove” the allegedly infringing images that VHT identified\nby property address, as opposed to their website designation\nor Uniform Resource Locator (“URL”), fails. Zillow\ntestified throughout trial that, in order to systematically or\nswiftly take down a large number of photos, it required the\nZillow Image ID–a number that is in the URL for each\nimage. This stands to reason, because “Zillow receives\nmultiple copies of the same photograph, depicting the same\nproperty, with the same listing agent, from different feeds.”\nMerely identifying the physical property address in no way\nidentified the proper feed or the correct photo. Thus, Zillow\ndid not have appropriately “specific” information necessary\nto take “simple measures” to remedy the violation.\n\n VHT’s argument that Zillow is liable for failing to ask\nfor the URLs of the allegedly infringing photos also fails.\nAsking for the URLs was not Zillow’s duty under the\ncontributory liability standard: Zillow must have “actual\n\f VHT, INC. V. ZILLOW GROUP 37\n\nknowledge that specific infringing material is available using\nits system.” Amazon, 508 F.3d at 1172 (citation and\nquotation omitted). Zillow first reasonably asked to see the\nlicenses between VHT and the feed providers; otherwise,\nZillow could not assess ownership and rights in the\nundefined images. That Zillow did not proactively request a\nlist of URLs before VHT filed suit does not make Zillow\ncontributorily liable.\n\n Additionally, Zillow’s failure to systematically use\nwatermarking technology does not show there was a “simple\nmeasure” available that it failed to use. Even assuming there\nwere “reasonable and feasible means” for Zillow to employ\nwatermark detection technology, in practice VHT rarely\nwatermarked its photos. See Giganews, 847 F.3d at 672\n(citation and quotation omitted).\n\n Nor did Zillow induce infringement. Inducement\nliability requires evidence of “active steps . . . taken to\nencourage direct infringement.” Metro-Goldwyn-Mayer\nStudios Inc. v. Grokster, Ltd., 545 U.S. 913, 936 (2005)\n(citation and quotation omitted)). Evidence of active steps\nincludes “advertising an infringing use or instructing how to\nengage in an infringing use,” because such evidence\n“show[s] an affirmative intent that the product be used to\ninfringe.” Id.; see also Columbia Pictures Industries, Inc. v.\nFung, 710 F.3d 1020, 1032 (9th Cir. 2013) (inducement\nliability requires “an object of promoting [the product’s] use\nto infringe copyright”). The “improper object” of\ninfringement “must be plain and must be affirmatively\ncommunicated through words or actions[.]” Fung, 710 F.3d\nat 1034.\n\n In view of the evidence, “no reasonable juror could\nconclude [Zillow] distributed its product with the object of\n\f38 VHT, INC. V. ZILLOW GROUP\n\npromoting its use to infringe copyright.” Giganews, 847\nF.3d at 672 (quotation and citation omitted). For example,\nZillow’s generally applicable tools and messages for users\nto save more photos from the Listing Platform to Digs does\nnot “promote[] the use of [Digs] specifically to infringe\ncopyrights.” Amazon, 508 F.3d at 1170 n.11. Nor does\nevidence that Zillow sometimes makes mistakes about the\ndisplay rights in a feed plainly communicate an improper\nobject of infringement. Zillow corrects these inadvertent\nerrors when it learns of them. Because a “failure to take\naffirmative steps to prevent infringement” alone cannot\ntrigger inducement liability, the inducement claim is a\nparticularly poor fit for Zillow’s real estate and home design\nwebsites, which have “substantial noninfringing uses.”\nGrokster, 545 U.S. at 939 n.12.\n\n B. Vicarious Liability\n\n Neither does the vicarious liability theory fit the Zillow\nplatform. To prevail on a vicarious liability claim, “[VHT]\nmust prove ‘[Zillow] has (1) the right and ability to supervise\nthe infringing conduct and (2) a direct financial interest in\nthe infringing activity.’” Giganews, 847 F.3d at 673\n(quoting Visa, 494 F.3d at 802). The first element requires\n“both a legal right to stop or limit the directly infringing\nconduct, as well as the practical ability to do so.” Amazon,\n508 F.3d at 1173. VHT’s vicarious infringement argument\nfails because, as the district court found, “Zillow ‘lack[ed]\nthe practical ability to police’ its users’ infringing conduct”\non Digs.\n\n As discussed with respect to contributory infringement,\nthere was insufficient evidence that Zillow had the technical\nability to screen out or identify infringing VHT photos\namong the many photos that users saved or uploaded daily\n\f VHT, INC. V. ZILLOW GROUP 39\n\nto Digs. Regardless, such allegations do not fall under the\nvicarious liability rubric: Zillow’s “failure to change its\noperations to avoid assisting [users] to distribute . . .\ninfringing content . . . is not the same as declining to exercise\na right and ability to make [third parties] stop their direct\ninfringement.” Amazon, 508 F.3d at 1175.\n\n Our conclusion is consistent with earlier dicta that “the\nvicarious liability standard applied in Napster can be met by\nmerely having the general ability to locate infringing\nmaterial and terminate users’ access.” UMG Recordings,\nInc. v. Shelter Capital Partners LLC, 718 F.3d 1006, 1030\n(9th Cir. 2013) (emphasis added). Once VHT photos were\nuploaded to the Listing Platform with appropriate\ncertification of rights, ferreting out claimed infringement\nthrough use on Digs was beyond hunting for a needle in a\nhaystack. As the district court concluded, “the trial record\nlacks substantial evidence of a practical ability to limit direct\ninfringement for the same reasons it lacks substantial\nevidence of simple measures to remove infringing material.”\nAnd linking a claimed infringement to a feed provider was\neven more of an impossibility.\n\n We affirm the district court’s judgment notwithstanding\nthe verdict concluding that substantial evidence did not\nsupport the claim that Zillow secondarily infringed VHT’s\nexclusive rights in its photos.\n\n III. Damages\n\n A. Compilation\n\n The size of the damages award hinges on whether VHT’s\nphotos used on Digs are part of a “compilation” or if they are\nindividual photos. This distinction makes a difference. If\nthe VHT photo database is a “compilation,” and therefore\n\f40 VHT, INC. V. ZILLOW GROUP\n\none “work” for the purposes of the Copyright Act, then VHT\nwould be limited to a single award of statutory damages for\nZillow’s use of thousands of photos on Digs. 17 U.S.C.\n§ 504(c)(1). But if the database is not a compilation, then\nVHT could seek damages for each photo that Zillow used.\n\n In lieu of actual damages, a copyright holder may elect\nto receive statutory damages under the Copyright Act. 17\nU.S.C. § 504(c)(1). VHT did so. The Act provides for “an\naward of statutory damages for all infringements involved in\nthe action, with respect to any one work . . . in a sum of not\nless than $750 or more than $30,000.” Id. This provision\nties statutory damages to the term “work,” which is\nundefined, except in a circular manner: copyright law\nprotects “original works of authorship.” 17 U.S.C. § 102(a).\nFortunately, there is a definition of compilation: “a work\nformed by the collection and assembling of preexisting\nmaterials or of data that are selected, coordinated, or\narranged in such a way that the resulting work as a whole\nconstitutes an original work of authorship.” 17 U.S.C.\n§ 101. For purposes of statutory damages, “all the parts of a\ncompilation . . . constitute one work.” 17 U.S.C.\n§ 504(c)(1). “The question of whether a work constitutes a\n‘compilation’ for the purposes of statutory damages pursuant\nto Section 504(c)(1) of the Copyright Act is a mixed question\nof law and fact.” Bryant v. Media Right Prods., Inc., 603\nF.3d 135, 140 (2d Cir. 2010).\n\n Whether various VHT photo collections comprise one or\nmore compilations is a threshold damages question. Before\ntrial, Zillow asked for a legal determination on the\ncompilation issue. That motion was denied. However, the\ndistrict court did not make an explicit determination about\ncompilation and the specifics of compilation were not put\nbefore the jury. In fairness to the district court, we might\n\f VHT, INC. V. ZILLOW GROUP 41\n\ninfer from the transcript that there was an implicit\ndetermination as to whether VHT’s photos are part of a\ncompilation, but we are left in doubt. Instead, the jury was\ninstructed that “[e]ach VHT Photo that has independent\neconomic value constitutes a separate work.” On the verdict\nform, the jury was asked which “photographs have\nindependent economic value.”\n\n The notion of “independent economic value” derives not\nfrom the statute, but from case law. In the Ninth Circuit, the\nquestion of whether something—like a photo, television\nepisode, or so forth—has “independent economic value”\ninforms our analysis of whether the photo or episode is a\nwork, though it is not a dispositive factor. See Columbia\nPictures Television, Inc. v. Krypton Broad. of Birmingham,\nInc., 259 F.3d 1186, 1193 (9th Cir. 2001). But consideration\nof the independent economic value factor does not answer\nthe question whether something is a compilation. That\nquestion remains unanswered here.\n\n On appeal, the parties have polar opposite views on\nwhether there was a compilation (Zillow’s position) or\nwhether each photo is entitled to a separate damages award\n(VHT’s position). VHT registered thousands of photos as\ncompilations. But the Copyright Office warns that such a\nregistration “may” limit the copyright holder “to claim only\none award of statutory damages in an infringement action,\neven if the defendant infringed all of the component works\ncovered by the registration.” U.S. COPYRIGHT OFFICE,\nCOMPENDIUM OF U.S. COPYRIGHT OFFICE PRACTICES\n§ 1104.5 (Sept. 2017), https://www.copyright.gov/comp3/\ndocs/compendium.pdf. Though the registration label is not\ncontrolling, it may be considered by the court when\nassessing whether a work is a compilation. See Yellow\nPages Photos, Inc. v. Ziplocal, LP, 795 F.3d 1255, 1277\n\f42 VHT, INC. V. ZILLOW GROUP\n\n(11th Cir. 2015) (“Although the manner of copyright\nregistration is not dispositive of the works issue, this Court\nhas previously considered it to be at least a relevant factor.”).\nUltimately, what counts is the statutory definition.\n\n Because there were at least ten different copyright\nregistrations, thousands of photos, and no explicit\ndetermination on compilation, we decline to sort out the\ncompilation issue on appeal. We remand to the district court\nfor further proceedings as to whether the VHT photos\nremaining at issue were a compilation.\n\n B. Willfulness\n\n The jury found that Zillow willfully infringed exclusive\nrights to 3,373 searchable VHT photos that were eligible for\nstatutory damages. The district court largely upheld the\nwillfulness finding in its post-trial motions order. However,\nthe court granted Zillow judgment notwithstanding the\nverdict on 673 images that were not displayed, so the court’s\nfinal willfulness judgment applied to 2,700 searchable\nphotos on Digs.\n\n To uphold a jury’s willfulness finding, there must be\nsubstantial evidence “(1) that the [the infringing party] was\nactually aware of the infringing activity, or (2) that the\n[infringing party’s] actions were the result of reckless\ndisregard for, or willful blindness to, the copyright holder’s\nrights.” Unicolors, 853 F.3d at 991 (citation and quotation\nomitted). Under the second prong, willful blindness requires\nthat the infringing party “(1) subjectively believed that\ninfringement was likely occurring on their networks and that\nthey (2) took deliberate actions to avoid learning about the\ninfringement.” Luvdarts, LLC v. AT&T Mobility, LLC, 710\nF.3d 1068, 1073 (9th Cir. 2013). Reckless disregard can be\ndemonstrated, for example, when a party “refus[es], as a\n\f VHT, INC. V. ZILLOW GROUP 43\n\nmatter of policy, to even investigate or attempt to determine\nwhether particular [photos] are subject to copyright\nprotections.” Unicolors, 853 F.3d at 992. A finding of\nwillfulness has significant financial consequences—the jury\nmay increase damages up to $150,000 per violation. See 17\nU.S.C. § 504(c)(2).\n\n As noted with respect to infringement, we do not take\nlightly the decision to reverse a jury verdict, nor do we\ncavalierly set aside the district court’s thoughtful analysis.\nBut here, we are compelled to disagree with both because\nsubstantial evidence does not support willfulness as to the\n2,700 photos.\n\n The test for willfulness is in the alternative: either actual\nnotice or recklessness shown by reckless disregard or\nturning a blind eye to infringement. We turn first to actual\nnotice. That VHT provided Zillow with minimal notice of\ninfringement does not itself establish that any subsequent\ninfringement was willful. Rather, “[c]ontinued use of a work\neven after one has been notified of his or her alleged\ninfringement does not constitute willfulness so long as one\nbelieves reasonably, and in good faith, that he or she is not\ninfringing.” Evergreen Safety Council v. RSA Network Inc.,\n697 F.3d 1221, 1228 (9th Cir. 2012); see also Danjaq LLC\nv. Sony Corp., 263 F.3d 942, 959 (9th Cir. 2001) (“It would\nseem to follow that one who has been notified that his\nconduct constitutes copyright infringement, but who\nreasonably and in good faith believes the contrary, is not\n‘willful’ for these purposes.”) (quoting 4 NIMMER ON\nCOPYRIGHT § 14.04[B][3]).\n\n Such is the case here. Zillow’s agreements with its feed\nproviders grant it an express license to use, copy, distribute,\npublicly display, and create derivative works for each photo,\n\f44 VHT, INC. V. ZILLOW GROUP\n\nand the agreements include unambiguous representations by\nthe feed providers that they have the authority to assign such\nrights. Zillow developed procedures to identify ex ante the\nscope of its license for each uploaded photo and employed\nautomated protocols to manage the use of each photo\nconsistent with its evergreen or deciduous designation. At\nno point during their year of communications prior to\nissuance of the notice letter did VHT raise the specter of\ninfringement. Notably, VHT’s eventual notice was minimal:\none letter with a list of allegedly infringing photos,\ndesignated by residential street address, not web address.\n\n The notion that Zillow failed to take appropriate\nresponsive measures after receiving this notice is belied by\nthe record. Zillow immediately requested information to\nconfirm VHT’s copyright ownership and cross-reference the\nphotos with licensing information. VHT was not\nforthcoming with that information. Rather, in response,\nVHT offered merely an unsigned form contract. Instead of\nproviding helpful information, VHT then filed suit. Given\nthe limited information provided by VHT, Zillow could not\nreasonably be expected to have promptly and unilaterally\nremoved each flagged photo. As the district court noted,\nVHT failed to demonstrate there were simple measures\navailable for the removal of infringing photos or that Zillow\nhad any “practical ability to independently identify\ninfringing images.”\n\n Thus, we are compelled to conclude that substantial\nevidence does not show Zillow was “actually aware” of its\ninfringing activity. See Evergreen Safety Council, 697 F.3d\nat 1228. Zillow’s belief that feed providers had properly\nlicensed its uses and that its system effectively respected\nthose rights was reasonable. And, as the district court\nobserved, “[t]he record suggests no reason to conclude that\n\f VHT, INC. V. ZILLOW GROUP 45\n\nZillow maintained that position in bad faith, and Zillow’s\nnon-infringement contention proved accurate as to most of\nthe images at issue in this lawsuit.”\n\n We reach the same conclusion as to whether Zillow\nrecklessly disregarded or willfully blinded itself to its\ninfringement. In reaching an opposite conclusion, the\ndistrict court observed that Zillow did not “perform[] further\ninvestigation into the rights each [feed provider] possesses,”\nnor did it “t[ake] responsive measures to obtain further\ninformation” after VHT provided the minimal notice of\npotential infringement. That conclusion is at odds with the\nevidence, for the reasons outlined above.\n\n VHT’s argument that Zillow, a sophisticated business\nwith a robust legal team, should have known that its feed\nprovider license agreements were invalid is unavailing.\nVHT argues that when Zillow saw the non-exclusive grant\nof rights in VHT’s unsigned form contract, showing that the\nfeed providers did not have a right to sublicense, Zillow\nshould have known the licenses were invalid. Despite\nrequests for such information, Zillow did not have access to\nVHT’s executed licenses with the feed providers who\nfurnished VHT’s photos to Zillow. Access to a blank form\ncontract (that the district court earlier found ambiguous as a\nmatter of law) is not enough. We conclude that substantial\nevidence does not show Zillow was “reckless or willfully\nblind” as to its infringement. We reverse the district court\nand vacate the jury’s finding of willful infringement.\n\n IV. Conclusion\n\n We affirm the district court’s summary judgment in\nfavor of Zillow on direct infringement of the Listing\nPlatform photos. With respect to direct liability on the Digs\nphotos, we affirm the district court’s grant of judgment\n\f46 VHT, INC. V. ZILLOW GROUP\n\nnotwithstanding the verdict on the 22,109 non-displayed\nphotos and the 2,093 displayed but not searchable photos.\nWe uphold summary judgment in favor of VHT on the 3,921\ndisplayed, searchable photos.\n\n We affirm the district court’s judgment notwithstanding\nthe verdict on secondary liability, both contributory and\nvicarious, on the Digs photos.\n\n We reverse the district court’s denial of judgment\nnotwithstanding the verdict on the issue of willfulness and\nvacate the jury’s finding on willfulness.\n\n We remand consideration of the compilation issue to the\ndistrict court.\n\n Affirmed in part, reversed in part and remanded.\nEach party shall pay its own costs on appeal.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Vht, Inc. v. Zillow Group, Inc.\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-15\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nVHT, INC., a Delaware corporation, Nos. 17-35587\\n Plaintiff-Appellee/ 17-35588\\n Cross-Appellant,\\n\\n v. D.C. No.\\n 2:15-cv-01096-\\nZILLOW GROUP, INC., a Washington JLR\\ncorporation; ZILLOW, INC., a\\nWashington corporation,\\n Defendants-Appellants/ OPINION\\n Cross-Appellees.\\n\\n Appeal from the United States District Court\\n for the Western District of Washington\\n James L. Robart, Senior District Judge, Presiding\\n\\n Argued and Submitted August 28, 2018\\n Seattle, Washington\\n\\n Filed March 15, 2019\\n\\nBefore: M. Margaret McKeown, William A. Fletcher, and\\n Ronald M. Gould, Circuit Judges.\\n\\n Opinion by Judge McKeown\\n\\f2 VHT, INC. V. ZILLOW GROUP\\n\\n SUMMARY *\\n\\n\\n Copyright Law\\n\\n The panel affirmed in part and reversed in part the\\ndistrict court’s judgment after a jury trial and remanded in a\\ncopyright infringement action brought by VHT, Inc., a real\\nestate photography studio, against Zillow Group, Inc., and\\nZillow, Inc., an online real estate marketplace.\\n\\n VHT alleged that Zillow’s use of photos on the “Listing\\nPlatform” and “Digs” parts of its website exceeded the scope\\nof VHT’s licenses to brokers, agents, and listing services that\\nprovided those photos to Zillow. The district court granted\\npartial summary judgment on a limited set of claims. The\\njury found in favor of VHT on most remaining claims,\\nawarding over $8.27 million in damages. The district court\\npartially granted Zillow’s post-trial motion for judgment\\nnotwithstanding the verdict, reversing in part the jury verdict\\nand reducing total damages to approximately $4 million.\\n\\n The panel affirmed the district court’s summary\\njudgment in favor of Zillow on direct infringement of the\\nListing Platform photos. The panel held that VHT failed to\\nestablish that Zillow engaged in volitional conduct by\\nexercising control over the content of the Listing Platform.\\n\\n With respect to direct liability on the Digs photos, the\\npanel affirmed the district court’s grant in favor of Zillow of\\njudgment notwithstanding the verdict on 22,109 non-\\ndisplayed photos and 2,093 displayed but not searchable\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f VHT, INC. V. ZILLOW GROUP 3\\n\\nphotos. The panel held that VHT did not present substantial\\nevidence that Zillow, through the Digs platform, directly\\ninfringed its display, reproduction, or adaption rights.\\n\\n The panel upheld summary judgment in favor of VHT\\non 3,921 displayed, searchable Digs photos. The panel held\\nthat fair use did not absolve Zillow of liability because\\nZillow’s tagging of the photos for searchable functionality\\nwas not a transformative fair use.\\n\\n The panel affirmed the district court’s grant in favor of\\nZillow of judgment notwithstanding the verdict on\\nsecondary liability, both contributory and vicarious, on the\\nDigs photos.\\n\\n As to damages, the panel remanded consideration of the\\nissue whether VHT’s photos used on Digs were part of a\\ncompilation or were individual photos.\\n\\n The panel reversed the district court’s denial of judgment\\nnotwithstanding the verdict on the issue of willfulness and\\nvacated the jury’s finding on willfulness. The panel\\nconcluded that substantial evidence did not show that Zillow\\nwas “actually aware” of its infringing activity, nor that\\nZillow recklessly disregarded or willfully blinded itself to its\\ninfringement.\\n\\n\\n COUNSEL\\n\\nIan B. Crosby (argued), Edgar G. Sargent, Genevieve Vose\\nWallace, and Jenna G. Farleigh, Susman Godfrey LLP,\\nSeattle, Washington, for Defendants-Appellants/Cross-\\nAppellees.\\n\\f4 VHT, INC. V. ZILLOW GROUP\\n\\nStephen M. Rummage (argued), Marcia B. Paul, Jonathan\\nM. Lloyd, James E. Howard, and Max B. Hensley, Davis\\nWright Tremaine LLP, Seattle, Washington, for Plaintiff-\\nAppellee/Cross-Appellant.\\n\\nKeith Kupferschmid and Terry Hart, Copyright Alliance,\\nWashington, D.C.; Eleanor M. Lackman and Lindsay W.\\nBowen, Cowan, DeBaets, Abrahams & Sheppard LLP, New\\nYork, New York; for Amicus Curiae Copyright Alliance.\\n\\nAlicia Calzada, Alicia Wagner Calzada, PLLC, San\\nAntonio, Texas, for Amici Curiae The American Society of\\nMedia Photographers, Inc., Digital Media Licensing\\nAssociation, Inc., Graphic Artist Guild, Inc., and National\\nPress Photographers Association, Inc.\\n\\nThomas G. Hentoff and Chanakya A. Sethi, Williams &\\nConnolly LLP, Washington D. C., for Amici Curiae Disney\\nEnterprises, Inc., Twentieth Century Fox Film Corporation,\\nand Recording Industry Association of America, Inc.\\n\\nMitchell L. Stoltz, Electronic Frontier Foundation, San\\nFrancisco, California, for Amicus Curiae Electronic Frontier\\nFoundation.\\n\\nBrian M. Willen, Wilson Sonsini Goodrich & Rosati, P.C.,\\nNew York, New York; Ryan T. O’Hollaren, Wilson Sonsini\\nGoodrich & Rosati, P.C., Palo Alto, California; for Amici\\nCuriae Internet Association and Computer &\\nCommunications Industry Association.\\n\\f VHT, INC. V. ZILLOW GROUP 5\\n\\n OPINION\\n\\nMcKEOWN, Circuit Judge:\\n\\n Zillow, an online real estate marketplace, has become a\\npopular website for homeowners and others to check\\nestimated valuations of their property, look for houses and\\ncondominiums for sale and rent, and see photographs of a\\nwide range of properties. Thousands of those copyrighted\\nphotos come from VHT, the largest professional real estate\\nphotography studio in the country.\\n\\n The copyright claims on appeal concern Zillow’s use of\\nVHT’s photos on two parts of Zillow’s website: the “Listing\\nPlatform” and “Digs.” The Listing Platform is the core of\\nthe website, featuring photos and information about real\\nestate properties, both on and off the market. Zillow claims\\nthat the site includes “most homes in America.” Digs\\nfeatures photos of artfully-designed rooms in some of those\\nproperties and is geared toward home improvement and\\nremodeling. Zillow tags photos on the Listing Platform so\\nthat Digs users can search the database by various criteria,\\nlike room type, style, cost, and color.\\n\\n Real estate brokers, listing services, and agents hire VHT\\nto take professional photos of new listings for marketing\\npurposes. A VHT photographer takes the photos and sends\\nthem to the company’s studio for touch-up, where they are\\nsaved to VHT’s electronic photo database, and then\\ndelivered to the client for use under license. Each license\\nagreement between VHT and its clients differs slightly, but\\neach contract generally grants the requesting client the right\\nto use the photos in the sale or marketing of the featured\\nproperty. Zillow receives these photos and other data in\\nfeeds from various real estate-related sources.\\n\\f6 VHT, INC. V. ZILLOW GROUP\\n\\n In 2015, VHT sued Zillow Group, Inc., and Zillow, Inc.,\\n(collectively “Zillow”) for copyright infringement, alleging\\nthat Zillow’s use of photos on the Listing Platform and Digs\\nexceeded the scope of VHT’s licenses to brokers, agents, and\\nlisting services who provided those photos to Zillow. The\\ndistrict court granted partial summary judgment on a limited\\nset of claims, while other claims advanced to trial. The jury\\nfound in favor of VHT on most remaining claims, awarding\\nover $8.27 million in damages. The district court partially\\ngranted Zillow’s post-trial motion for judgment\\nnotwithstanding the verdict, reversing in part the jury verdict\\nand reducing total damages to approximately $4 million.\\n\\n The parties cross-appealed issues stemming from partial\\nsummary judgment, the jury verdict, and judgment\\nnotwithstanding the verdict. We affirm in part and reverse\\nin part. 1\\n\\n To simplify and make sense of the various claims, this\\nopinion does not split out the appeal and cross-appeal as was\\ndone in the briefing to the court. Instead, the opinion\\nseparately addresses liability for each of the categories of\\nphotos at issue, followed by a discussion of damages. In\\nview of the multiple theories of liability and categories of\\nphotos, following is an overview of the opinion.\\n\\n\\n\\n\\n 1\\n In connection with these proceedings, we received amicus curiae\\nbriefs from a broad array of interested parties, including nonprofit groups\\nand associations representing a diverse set of industry, technology, and\\nartistic interests. The briefs were helpful to our understanding of the\\nimplications of this case from various points of view. We thank amici\\nfor their participation.\\n\\f VHT, INC. V. ZILLOW GROUP 7\\n\\n I. Direct Infringement\\n A. Direct Infringement—Listing Platform Photos\\n B. Direct Infringement—Digs Photos\\n 1. Jury Verdict—Direct Infringement\\n 2. Summary Judgment—Fair Use re Searchable\\n Photos\\n a. Background on Fair Use\\n b. Evolution of Search Engine Cases\\n c. Application of Fair Use Principles\\n II. Secondary Infringement—Digs\\n A. Contributory Liability\\n B. Vicarious Liability\\n III. Damages\\n A. Compilation\\n B. Willfulness\\n IV. Conclusion\\n\\n ANALYSIS\\n\\n The heart of this dispute is Zillow’s copyright liability\\nfor use of VHT photos. VHT argues that Zillow directly\\ninfringed its copyrighted photos, both those on the Listing\\nPlatform and Digs. VHT also argues that Zillow indirectly\\ninfringed through use of the photos on Digs. These claims\\npertain to different images, focus on different features of\\nZillow’s website, and have different procedural postures, so\\nwe consider the various categories of photos separately.\\n\\n I. Direct Infringement\\n\\n VHT’s key claim is that Zillow is directly liable for\\ninfringing VHT’s copyright on photos that were posted on\\nthe Listing Platform and Digs. To prevail on a claim of\\ndirect copyright infringement, VHT must establish\\n“ownership of the allegedly infringed material” and that\\nZillow “violate[d] at least one exclusive right granted to”\\n\\f8 VHT, INC. V. ZILLOW GROUP\\n\\nVHT under 17 U.S.C. § 106. A&M Records, Inc. v. Napster,\\nInc., 239 F.3d 1004, 1013 (9th Cir. 2001). It is undisputed\\nthat VHT is the copyright holder of the allegedly infringed\\nphotos and therefore has the exclusive right to reproduce,\\nadapt, and display them. 2 17 U.S.C. § 106.\\n\\n VHT must also establish causation, which is commonly\\nreferred to as the “volitional-conduct requirement.” See\\nPerfect 10, Inc. v. Giganews, Inc., 847 F.3d 657, 666 (9th\\nCir. 2017). As we set out in Giganews—decided on the first\\nday of the VHT/Zillow trial and the closest circuit precedent\\non point—“volition in this context does not really mean an\\nact of willing or choosing or an act of deciding”; rather, “it\\nsimply stands for the unremarkable proposition that\\nproximate causation historically underlines copyright\\ninfringement liability no less than other torts.” Id. (internal\\ncitations omitted). Stated differently, “direct liability must\\nbe premised on conduct that can reasonably be described as\\nthe direct cause of the infringement.” Id. (citation omitted).\\nThis prerequisite takes on greater importance in cases\\ninvolving automated systems, like the Zillow website.\\n\\n In addressing this concept, Justice Scalia noted that\\n“[e]very Court of Appeals to have considered an automated-\\nservice provider’s direct liability for copyright infringement\\nhas adopted [the volitional-conduct] rule.” Am. Broad. Cos.,\\nInc. v. Aereo, Inc., 573 U.S. 431, 453 (2014) (Scalia, J.,\\ndissenting). 3 He went on to explain that while “most direct-\\n\\n 2\\n VHT does not appeal the district court’s finding that there was\\ninsufficient evidence that Zillow violated VHT’s distribution rights.\\n 3\\n Although the majority opinion in Aereo does not reference the\\nvolitional-conduct requirement, Justice Scalia’s dissent offers instructive\\nbackground on the doctrine. In Giganews, we embraced the principle\\n\\f VHT, INC. V. ZILLOW GROUP 9\\n\\ninfringement cases” do not present this issue, “it comes right\\nto the fore when a direct-infringement claim is lodged\\nagainst a defendant who does nothing more than operate an\\nautomated, user-controlled system. . . . Most of the time that\\nissue will come down to who selects the copyrighted\\ncontent: the defendant or its customers.” Id. at 454–55\\n(internal citations omitted).\\n\\n Giganews, Aereo, and out-of-circuit precedent counsel\\nthat direct copyright liability for website owners arises when\\nthey are actively involved in the infringement. “‘[T]he\\ndistinction between active and passive participation’” in the\\nalleged infringement is “‘central’” to the legal analysis.\\nGiganews, 847 F.3d at 667 (quoting Fox Broad. Co. v. Dish\\nNetwork LLC, 160 F. Supp. 3d 1139, 1160 (C.D. Cal. 2015)).\\n\\n That “direct” infringement requires “active”\\ninvolvement is hardly surprising, given the correlation\\nbetween the words “active” and “direct.” As the Fourth\\nCircuit held, “[t]here must be actual infringing conduct with\\na nexus sufficiently close and causal to the illegal copying\\nthat one could conclude that the machine owner himself\\ntrespassed on the exclusive domain of the copyright owner.”\\nCoStar Grp., Inc. v. LoopNet, Inc., 373 F.3d 544, 550 (4th\\nCir. 2004). By contrast, activities that fall on the other side\\nof the line, such as “‘automatic copying, storage, and\\ntransmission of copyrighted materials, when instigated by\\nothers, do[] not render an [Internet service provider] strictly\\nliable for copyright infringement[.]’” Giganews, 847 F.3d at\\n670 (quoting CoStar, 373 F.3d at 555).\\n\\n\\n\\nand held that it is “consistent with the Aereo majority opinion,” which\\nleft the requirement “intact.” 847 F.3d at 666–67.\\n\\f10 VHT, INC. V. ZILLOW GROUP\\n\\n In other words, to demonstrate volitional conduct, a party\\nlike VHT must provide some “evidence showing [the alleged\\ninfringer] exercised control (other than by general operation\\nof [its website]); selected any material for upload, download,\\ntransmission, or storage; or instigated any copying, storage,\\nor distribution” of its photos. Id. at 666, 670. VHT failed to\\nsatisfy that burden with respect to either the photos on the\\nListing Platform or on Digs.\\n\\n A. Direct Infringement—Listing Platform\\n Photos\\n\\n VHT asserted that Zillow directly infringed the photos\\ndisplayed on the Listing Platform after a real estate property\\nwas sold because VHT’s license agreements only authorized\\nuse of those photos in relation to the sale of the property.\\nThis claim, involving 54,257 non-searchable photos, was\\nresolved on summary judgment. The Listing Platform is the\\ncore of Zillow’s online real estate marketplace. It features\\nphotos and information about properties, which Zillow\\nreceives through digital feeds from real estate agents,\\nbrokerages, and multiple listing services, among others\\n(collectively “feed providers”).\\n\\n Zillow has agreements with its feed providers granting it\\nan express license to use, copy, distribute, publicly display,\\nand create derivative works from the feed data on its\\nwebsites. Feed providers represent that they “ha[ve] all\\nnecessary rights and authority to enter into” the agreements,\\nand that “Zillow’s exercise of the rights granted [t]hereunder\\nwill not violate the intellectual property rights, or any other\\nrights of any third party.”\\n\\n These agreements provide Zillow with either\\n“evergreen” or “deciduous” rights in the photos provided\\nthrough the feeds. An evergreen right permits use of a photo\\n\\f VHT, INC. V. ZILLOW GROUP 11\\n\\nwithout any time restriction, “on and in connection with the\\noperation, marketing and promotion of the web sites and\\nother properties, owned, operated or powered by Zillow or\\nits authorized licensees.” By contrast, a deciduous right is\\ntemporally limited: Zillow may use the photo when the real-\\nestate listing for its corresponding property is active, but\\nonce the listing is removed (for example, when the property\\nsells), the photo must be taken down from Zillow’s websites.\\nTo treat each photo consistently with its deciduous or\\nevergreen designation, Zillow developed automated\\n“trumping” rules to determine which photos to display on the\\nListing Platform.\\n\\n VHT argues that Zillow “designed its system to . . .\\ncause[] the reproduction, display, and adaptation of VHT\\nphotographs post-sale on the Listing Platform,” and “chose\\nto simply ignore VHT’s notices that post-sale use was\\nbeyond the scope of VHT’s licenses.” The district court\\ngranted summary judgment to Zillow, concluding that it did\\nnot engage in volitional conduct and therefore did not\\ndirectly infringe VHT’s copyrights in 54,257 photos by\\ndisplaying them on the Listing Platform after a real estate\\nproperty was sold.\\n\\n On de novo review, we agree with the district court’s\\nanalysis and affirm. Kelly v. Arriba Soft Corp., 336 F.3d\\n811, 817 (9th Cir. 2003). Although the district court did not\\nhave the benefit of Giganews at the time of summary\\njudgment, its careful reasoning was prescient in invoking the\\nsame principles.\\n\\n Zillow did not engage in volitional conduct necessary to\\nsupport a finding of direct liability. The content of the\\nListing Platform is populated with data submitted by third-\\nparty sources that attested to the permissible use of that data,\\n\\f12 VHT, INC. V. ZILLOW GROUP\\n\\nand Zillow’s system for managing photos on the Listing\\nPlatform was constructed in a copyright-protective way.\\nThe feed providers themselves select and upload every\\nphoto, along with the evergreen or deciduous designations,\\nthat wind up on the Listing Platform. As a result, the photos\\non the Listing Platform were not “selected” by Zillow. See\\nGiganews, 847 F.3d at 670. Nor did Zillow “exercise[]\\ncontrol” over these photos beyond the “general operation of\\n[its website].” Id. Zillow required feed providers to certify\\nthe extent of their rights to use each photo. Consistent with\\nthese designations, Zillow’s system classified each photo as\\ndeciduous or evergreen and programmed its automated\\nsystems to treat each photo consistently with that scope of\\nuse certified to by the third party.\\n\\n Further, when multiple versions of the same photo were\\nsubmitted through the various feeds, Zillow invoked its\\ncopyright-protective “trumping” rules. For example, one\\nrule might prefer a photo provided by an agent over one\\nprovided by a multiple listing service, and another might\\nprefer a local broker to an international one. Zillow used a\\nrule that gave preference to photos with evergreen rights\\nover photos with deciduous rights in the same image. As the\\ndistrict court recognized, “trumping” is a reasonable way to\\ndesign a system to manage multiple versions of the same\\nphoto when the authorized use varies across versions. These\\nrules, along with other features of the system, facilitate\\nkeeping the photos with evergreen rights on the website and\\nremoving the photos with deciduous rights once a property\\nhas sold. Thus, Zillow actively designed its system to avoid\\nand eliminate copyright infringement.\\n\\n Notably, VHT’s argument is primarily cast in terms of\\nZillow facilitating or enabling infringement by VHT’s\\nclients that are Zillow’s feed providers. But this type of\\n\\f VHT, INC. V. ZILLOW GROUP 13\\n\\nclaim more properly falls in the category of secondary\\ninfringement, a claim not advanced by VHT with respect to\\nthe Listing Platform photos.\\n\\n VHT also asserts that Zillow failed to remove photos\\nonce it received notice that infringing content was on the\\nListing Platform, a conscious choice that amounts to\\nvolitional conduct on Zillow’s part. This claim is unavailing\\nbecause, once VHT put Zillow on notice of claimed\\ninfringement, Zillow took affirmative action to address the\\nclaims. Additionally, VHT’s assertion that it “repeatedly\\nnotified Zillow that it was infringing” is unsupported in the\\nrecord.\\n\\n In July 2014, VHT sent Zillow a takedown notice letter\\nwith a list of thousands of allegedly infringing photos by\\nresidential street address (but not by web address). Zillow\\npromptly requested all executed license agreements between\\nVHT and the feed providers who had provided photos to\\nZillow, as well as license agreements between VHT and its\\nphotographers, so that Zillow could evaluate whether VHT\\npossessed exclusive rights to the photos on the Listing\\nPlatform. VHT responded with an unsigned form contract,\\nwhich it stated was used with many feed providers, but\\nwhich was not tied to any specific photos on Zillow’s\\nwebsite. Zillow again reiterated its need to see the specific\\ncontracts governing the contested photos. Instead of\\nresponding with the contracts, VHT filed suit. Zillow’s\\nreasonable response to VHT’s single formal inquiry\\n(supplemented in a follow-on email) can hardly be\\ncharacterized as rising to the level of volitional conduct or\\nturning a blind eye.\\n\\n In sum, VHT failed to “provide[] . . . evidence showing\\n[Zillow] exercised control (other than by general operation\\n\\f14 VHT, INC. V. ZILLOW GROUP\\n\\nof [its website]); selected any material for upload, download,\\ntransmission, or storage; or instigated any copying, storage,\\nor distribution” of these photos. See Giganews, 847 F.3d at\\n670; see also CoStar, 373 F.3d at 555. Thus, we affirm the\\ndistrict court and conclude Zillow did not directly infringe\\nVHT’s copyrights in photos displayed on the Listing\\nPlatform post-sale.\\n\\n B. Direct Infringement—Digs Photos\\n\\n VHT also claimed that Zillow directly infringed\\nthousands of photos used on Digs. The jury concluded that\\nZillow directly infringed 28,125 photos and rejected its fair\\nuse defense. Following trial, the district court granted in\\nsubstantial part Zillow’s motion for judgment\\nnotwithstanding the verdict on the ground that insufficient\\nevidence supported Zillow’s direct infringement of 22,109\\nphotos that were not displayed on Digs and 2,093 4 photos\\nthat were displayed but not searchable on Digs.\\n\\n By contrast, the court upheld the jury’s determination\\nthat Zillow directly infringed a set of 3,921 images that were\\nselected and tagged by Zillow moderators for searchable\\nfunctionality and displayed on Digs. Zillow does not appeal\\nthis ruling. However, Zillow argues that fair use insulates it\\nfrom liability as to this subset of photos. The jury was\\ninstructed not to consider this legal theory as to these photos\\nbecause the district court had determined pretrial that, as a\\nmatter of law, the searchability function did not constitute\\nfair use. It is that summary judgment ruling that Zillow\\nchallenges on appeal. Because we agree with the district\\n\\n 4\\n The displayed but non-searchable set includes 2,094 photos. The\\ndistrict court affirmed the jury verdict with respect to one of those\\nphotos, which Zillow also distributed via email.\\n\\f VHT, INC. V. ZILLOW GROUP 15\\n\\ncourt that the fair use defense does not absolve Zillow of\\ndirect liability for these searchable photos, this portion of the\\njury verdict remains intact.\\n\\n The following chart clarifies the status of the Digs photos\\nrelevant to the direct infringement claims.\\n\\n Photos Jury Verdict Post-Trial Party\\n Determination Bringing\\n Appeal\\n22,109 not Direct Overturned jury VHT\\ndisplayed 5 infringement verdict\\n2,094 Direct Overturned jury VHT\\ndisplayed, not infringement verdict\\nsearchable (except 1 email\\n photo)\\n3,921 Direct Upheld jury Zillow:\\ndisplayed, infringement verdict direct\\nsearchable infringement\\n not appealed;\\n appeals only\\n summary\\n judgment on\\n no fair use\\n1 blog post Direct Upheld jury Not appealed\\nphoto (not on infringement verdict\\nDigs)\\n\\n\\n\\n\\n 5\\n Searchable photos numbered 1,694; 20,415 photos were not\\nsearchable.\\n\\f16 VHT, INC. V. ZILLOW GROUP\\n\\n 1. Jury Verdict—Direct Infringement\\n\\n VHT’s claim that Zillow directly infringed photos on\\nDigs went to the jury. The jury verdict form was framed in\\ngeneral terms, asking only whether “VHT has proven its\\ndirect copyright infringement claim as to one or more of the\\nVHT Photos[.]” The jury answered “yes,” and was asked to\\nspecify how many VHT photos were directly infringed. The\\njury answered “28,125”—in other words, all of them.\\nHowever, the jury was not asked to specify which copyright\\nrights—display, reproduction, or adaption—were\\n 6\\ninfringed. Following trial, Zillow moved for judgment\\nnotwithstanding the verdict or for a new trial. The task fell\\nto the district court to examine the evidence as to each right.\\n\\n On de novo review, “we apply the same standard used\\nby the district court in evaluating the jury’s verdict” and\\nuphold the verdict unless “the evidence permits only one\\nreasonable conclusion, and that conclusion is contrary to the\\njury’s verdict.” Wallace v. City of San Diego, 479 F.3d 616,\\n624 (9th Cir. 2007); Fed. R. Civ. P. 50(a). Specifically, we\\n“ask[] whether the verdict is supported by substantial\\nevidence,” “which is evidence adequate to support the jury’s\\nconclusion, even if it is also possible to draw a contrary\\nconclusion.” Unicolors, Inc. v. Urban Outfitters, Inc., 853\\nF.3d 980, 984, 991 (9th Cir. 2017) (quotation and citation\\nomitted). Given the sanctity of the jury process, we\\nundertake this review with special care and reluctance to\\noverturn a verdict. However, because the verdict here did\\nnot meet this standard, we affirm the district court’s grant of\\nZillow’s motion for judgment notwithstanding the verdict\\n\\n\\n 6\\n As noted earlier, the distribution right was not contested and is not\\nan issue on appeal.\\n\\f VHT, INC. V. ZILLOW GROUP 17\\n\\nwith respect to direct infringement of 22,109 non-displayed\\nphotos and 2,093 displayed but not searchable photos.\\n\\n We first consider display rights. As background for our\\nanalysis, it is useful to consider the direct infringement by\\nZillow that the district court upheld and that Zillow did not\\nappeal. The district court found substantial evidence that\\nZillow directly infringed VHT’s display rights in 3,921\\nphotos displayed on Digs that Zillow moderators selected\\nand tagged for searchable functionality. Based on testimony,\\ncharts, and statistics, the court found that “the jury could\\nhave reasonably concluded that users accessed those images\\nthrough Digs’s search function.” The court went on to\\nreason that “the jury could have reasonably concluded that\\nZillow’s moderation efforts, which rendered those images\\nsearchable, proximately caused the copying.” Put\\ndifferently, active conduct by Zillow met the volitional-\\nconduct requirement for direct infringement. Zillow does\\nnot appeal this ruling upholding the jury’s verdict as to the\\n3,921 displayed, searchable photos.\\n\\n On appeal, VHT attempts to shoehorn an additional\\n1,694 photos into this category. This effort falls flat both as\\na factual and legal matter because substantial evidence does\\nnot support direct infringement of VHT’s display rights in\\nthe 1,694 searchable images that were not displayed.\\n\\n VHT posits that the jury could have found that these\\nphotos were displayed because of circumstantial evidence\\nand because Zillow failed to record whether they were\\ndisplayed. Not so. This argument is foreclosed by the\\nparties’ stipulated spreadsheet that categorized each photo.\\nThe column labeled “DISPLAYED” included an entry for\\n“Y” (yes) or “N” (no). The jury was instructed to “treat\\nevery fact on this spreadsheet as proven,” so VHT cannot\\n\\f18 VHT, INC. V. ZILLOW GROUP\\n\\nrecast the facts retroactively and now claim that 1,694\\nsearchable images stipulated as “N[OT] DISPLAYED”\\nwere in fact displayed or made available for display. Up is\\nnot down.\\n\\n VHT’s contention that the jury could have reasonably\\ninferred that Zillow made “available for public display” all\\n22,109 “N[OT] DISPLAYED” images similarly fails. This\\ntheory presumes that the Copyright Act’s display right\\nencompasses an exclusive right to “make available for\\ndisplay,” a position neither supported by the statute nor\\nembraced by this court. See Perfect 10, Inc. v. Amazon.com,\\nInc., 508 F.3d 1146, 1160 (9th Cir. 2007) (“[B]ased on the\\nplain language of the statute, a person displays a\\nphotographic image by using a computer to fill a computer\\nscreen with a copy of the photographic image fixed in the\\ncomputer’s memory.”); 17 U.S.C. § 101 (“To ‘display’ a\\nwork means to show a copy of it, either directly or by means\\nof a film, slide, television image, or any other device or\\nprocess or, in the case of a motion picture or other\\naudiovisual work, to show individual images\\nnonsequentially.”). To be sure, the Copyright Office notes\\nthat the outer limits of the public display right have yet to be\\ndefined. U.S. COPYRIGHT OFFICE, THE MAKING AVAILABLE\\nRIGHT IN THE UNITED STATES 47–51 (Feb. 2016),\\nhttps://www.copyright.gov/docs/making_available/making-\\navailable-right.pdf.\\n\\n What is most important here, though, is that VHT’s\\nargument comes too late. The jury was never instructed on\\nthe “made available” theory, nor did VHT raise this issue in\\nits proposed jury instructions or in objections to the final\\ninstructions. See Sinclair v. Long Island R.R., 985 F.2d 74,\\n78 (2d Cir. 1993) (“the verdict . . . cannot be sustained on a\\ntheory that was never presented to the jury”); Ramona Equip.\\n\\f VHT, INC. V. ZILLOW GROUP 19\\n\\nRental, Inc. ex rel. U.S. v. Carolina Cas. Ins. Co., 755 F.3d\\n1063, 1070 (9th Cir. 2014) (an argument raised for first time\\nin a post-trial motion is waived). For these reasons,\\nsubstantial evidence did not support a finding of direct\\ninfringement of VHT’s public display rights in these 22,109\\nphotos.\\n\\n Zillow also did not violate VHT’s display rights in 2,093\\ndisplayed, non-searchable photos. These are photos that\\nDigs users copied to “personal boards” and “Implicit Digs,”\\nbut which Zillow did not add to the searchable set.\\n\\n A “personal board” is a bespoke digital bulletin board of\\nimages that a user saves or uploads from the Listing\\nPlatform. Users can also upload their own images. These\\nboards are typically private, though users can share a link to\\ntheir boards. When a user saves a copy of an image with\\nevergreen rights to a personal board, that image\\nautomatically joins a queue for review by a Zillow\\nmoderator. The moderator then decides whether to\\ndesignate the photo for tagging so that it can be searchable\\non Digs, and thus select it for public display. Not all photos\\nthat are in the queue for moderation are reviewed.\\nAdditionally, in some instances, a user starts but does not\\nfinish the process of saving an image to this board.\\nBeginning in 2014, Zillow programmed these “clicked to\\nsave” images—called “Implicit Digs”—to enter the queue\\nfor moderator review in the same manner as if the photo had\\nbeen saved to a personal board.\\n\\n According to VHT, the jury could have found that Zillow\\n“caused the [2,093] images to be displayed . . . by subjecting\\nthe non-searchable VHT Photos to the potential for\\nmoderation.” This argument defies logic because the only\\ndisplay that occurred was triggered by the user. Any\\n\\f20 VHT, INC. V. ZILLOW GROUP\\n\\npotential for future display is purely speculative. As the\\ndistrict court explained, the possibility that images might be\\nmoderated and tagged—conduct that is volitional—is not\\nsufficient “to transform Zillow from a ‘passive host’ to a\\n‘direct[] cause’ of the display of VHT’s images.”\\n\\n Next, we consider VHT’s exclusive reproduction and\\nadaption rights in the 2,093 displayed, non-searchable\\nphotos. The Copyright Act grants copyright holders the\\nexclusive right “to reproduce the copyrighted work in copies\\nor phonorecords.” 17 U.S.C. § 106(1). Direct infringement\\nof the reproduction right “requires copying by the defendant,\\n. . . which [requires] that the defendant cause the copying.”\\nFox Broad. Co., Inc. v. Dish Network LLC, 747 F.3d 1060,\\n1067 (9th Cir. 2014) (quotation and citation omitted). The\\nadaptation right is the exclusive right “to prepare derivative\\nworks based upon the copyrighted work.” 17 U.S.C.\\n§ 106(2). Following Giganews, we conclude that Zillow’s\\nautomated processes storing or caching VHT’s photos are\\ninsufficiently volitional to establish that Zillow directly\\ninfringed VHT’s reproduction and adaption rights in these\\nnon-searchable photos. Unlike photos that Zillow curated,\\nselected, and tagged for searchable functionality—activities\\nthat amount to volitional conduct establishing direct\\nliability—these 2,093 photos were copied to “personal\\nboards” and “Implicit Digs” based on user actions, not the\\nconduct of Zillow or its moderators.\\n\\n Any volitional conduct with respect to these photos was\\ntaken by the users, not Zillow. Users, not Zillow,\\n“selecte[d]” images to add to their personal boards and\\n“instigate[d]” the automatic caching process by saving a\\nparticular image. See Giganews, 847 F.3d at 670. This\\narrangement is important because courts have found no\\ndirect liability where an online system “responds\\n\\f VHT, INC. V. ZILLOW GROUP 21\\n\\nautomatically to users’ input . . . without intervening\\nconduct” by the website owner. CoStar, 373 F.3d at 550;\\nsee also Giganews, 847 F.3d at 670. Under these conditions,\\ncourts have analogized online facilities, like Internet service\\nproviders, to a copy machine owner, who is not liable\\n“[w]hen a customer duplicates an infringing work.” CoStar,\\n373 F.3d at 550.\\n\\n Additionally, Zillow’s behind-the-scenes technical work\\non Digs photos is not evidence that Zillow “selected any\\nmaterial for upload, download, transmission, or storage.”\\nGiganews, 847 F.3d at 670. Zillow produced cached copies\\nof these Digs images, a process that automatically trims or\\npads images whose height and width did not match the target\\nresolution, for the purposes of accelerating website speed.\\nThis activity does not amount to volitional conduct. Nor can\\nZillow’s promotion of Digs, including encouraging users to\\nshare photos through its site, be seen as “instigat[ing]” user\\ncopying. Id.\\n\\n Zillow’s conduct with respect to these photos amounts\\nto, at most, passive participation in the alleged infringement\\nof reproduction and adaption rights and is not sufficient to\\ncross the volitional-conduct line. As in cases involving\\nInternet service providers, Zillow “affords its [users] an\\nInternet-based facility on which to post materials, but the\\nmaterials posted are of a type and kind selected by the [user]\\nand at a time initiated by the [user].” CoStar, 373 F.3d at\\n555. Zillow did not directly infringe VHT’s reproduction\\n\\f22 VHT, INC. V. ZILLOW GROUP\\n\\nand adaptation rights in the 2,093 displayed, non-searchable\\nphotos. 7\\n\\n * * *\\n\\n In sum, VHT did not present substantial evidence that\\nZillow, through the Digs platform, directly infringed its\\ndisplay, reproduction, or adaption rights in 22,109 not\\ndisplayed photos and 2,093 displayed but non-searchable\\nphotos. We affirm the district court’s grant of judgment\\nnotwithstanding the verdict as to these photos. We now turn\\nto Zillow’s fair use defense to direct infringement of 3,921\\ndisplayed, searchable photos.\\n\\n 2. Summary Judgment—Fair Use re\\n Searchable Photos\\n\\n Zillow does not appeal the jury’s finding of direct\\ninfringement with respect to the 3,921 displayed, searchable\\nphotos, but does assert a fair use defense for those photos.\\nZillow contends that Digs’ searchable functionality\\nconstitutes fair use, which the district court rejected as a\\nmatter of law at summary judgment.\\n\\n We recount the somewhat unusual history of the fair use\\nissue in the proceedings below. On summary judgment, the\\ndistrict court rejected Zillow’s argument that “the images\\nthat it has made searchable on Digs” are protected by fair use\\nand instead “conclude[d] as a matter of law that Digs’\\nsearchable functionality does not constitute a fair use.” At\\ntrial, the jury was generally instructed to consider the fair use\\n\\n 7\\n The same analysis applies to any potential violation of VHT’s\\nexclusive reproduction and adaption rights in the 22,109 photos that were\\nnot displayed.\\n\\f VHT, INC. V. ZILLOW GROUP 23\\n\\ndefense as to all VHT photos used on Digs that it found\\nZillow had directly or indirectly infringed. However, this set\\nof photos was carved out for separate treatment. The jury\\nconsidered only whether “reproduction, cropping, and\\nscaling” of these photos constituted fair use because the\\ncourt instructed the jury that the court “ha[d] determined,\\nand you are to take as proven, that the Digs searchable\\nfunctionality does not constitute fair use.” After finding that\\nZillow directly infringed all 28,125 VHT photos used on\\nDigs, the jury rejected Zillow’s fair use affirmative defense\\nfor all photos.\\n\\n The district court upheld the jury’s fair use verdict,\\nwhich Zillow does not appeal. Rather, Zillow’s appeal\\nreaches back to the district court’s summary judgment ruling\\nto argue that the Digs’ searchable functionality is fair use as\\na matter of law, and, as a result, Zillow bears no liability for\\nthe 3,921 searchable and displayed photos. We review de\\nnovo the district court’s grant of summary judgment to VHT\\non this mixed question of law and fact. Kelly, 336 F.3d at\\n817.\\n\\n a. Background on Fair Use\\n\\n Protection of copyrighted works is not absolute. “The\\nfair use defense permits the use of copyrighted works\\nwithout the copyright owner’s consent under certain\\nsituations.” Amazon, 508 F.3d at 1163. Fair use both fosters\\ninnovation and encourages iteration on others’ ideas, “thus\\nproviding a necessary counterbalance to the copyright law’s\\ngoal of protecting creators’ work product.” Id.; see\\nCampbell v. Acuff-Rose Music, Inc., 510 U.S. 569, 575–76\\n(1994). Fair use also aligns with copyright’s larger purpose\\n“‘[t]o promote the Progress of Science and useful Arts,’ . . .\\nand to serve ‘the welfare of the public.’” Amazon, 508 F.3d\\n\\f24 VHT, INC. V. ZILLOW GROUP\\n\\nat 1163 (quoting U.S. Const. art. I, § 8, cl. 8, and Sony Corp.\\nof Am. v. Universal City Studios, Inc., 464 U.S. 417, 429 n.10\\n(1984)).\\n\\n Fittingly enough, a case involving a biography of George\\nWashington serves as a foundational source of fair use in\\nAmerican law. Folsom v. Marsh, 9 F. Cas. 342 (D. Mass.\\n1841) (No. 4901). Justice Story’s narrative description of\\ncopyright doctrine in that case “distilled the essence of law\\nand methodology from the earlier cases” and provided the\\nconceptual basis for the judge-made fair use doctrine.\\nCampbell, 510 U.S. at 576. Although the 1976 Copyright\\nAct codified those principles, it did little to elaborate on\\nJustice Story’s description or to clarify application of the\\nfactors. With minimal guidance or elucidation, Congress set\\nforth four factors for courts to consider when determining\\nwhether the use of a copyrighted work is a “fair use”:\\n\\n (1) the purpose and character of the use,\\n including whether such use is of a\\n commercial nature or is for nonprofit\\n educational purposes;\\n\\n (2) the nature of the copyrighted work;\\n\\n (3) the amount and substantiality of the\\n portion used in relation to the copyrighted\\n work as a whole; and\\n\\n (4) the effect of the use upon the potential\\n market for or value of the copyrighted\\n work.\\n\\n17 U.S.C. § 107.\\n\\f VHT, INC. V. ZILLOW GROUP 25\\n\\n Given license to apply these factors flexibly and to\\nconsider them in their totality, courts have been bedeviled by\\nthe fair use inquiry. See Campbell, 510 U.S. at 577–78. Fair\\nuse has been called “the most troublesome [doctrine] in the\\nwhole law of copyright” and commentators have criticized\\nthe factors as “billowing white goo.” Monge v. Maya\\nMagazines, Inc., 688 F.3d 1164, 1170–71 (9th Cir. 2012)\\n(citations and quotations omitted). In a (still ongoing) effort\\nto adapt the fair use analysis to a myriad of circumstances,\\ncourts have embellished and supplemented the factors. For\\nexample, the concept of “transformativeness” is found\\nnowhere in the statute, but appeared for the first time in the\\nSupreme Court in Campbell, where the Court endeavored to\\nrefine and crystalize the first statutory factor: the “purpose\\nand character of the use.” 8 510 U.S. at 579. The animating\\npurpose of the first factor is to determine,\\n\\n in Justice Story’s words, whether the new\\n work merely ‘supersede[s] the objects’ of the\\n original creation . . . or instead adds\\n something new, with a further purpose or\\n different character, altering the first with new\\n expression, meaning, or message; it asks, in\\n other words, whether and to what extent the\\n new work is “transformative.”\\n\\n\\n\\n 8\\n The concept of transformative use had appeared in earlier lower\\ncourt decisions, e.g., Twin Peaks Prods., Inc. v. Publ’ns Intern., Ltd., 996\\nF.2d 1366, 1375 (2d Cir. 1993) (“[I]t is not uncommon for works serving\\na fair use purpose to give at least a brief indication of the plot. . . . In\\nidentifying plot, the author of the second work may or may not be said\\nto have made . . . a ‘transformative’ use. . . . Such use would occur, for\\nexample, if a plot was briefly described for purposes of adding\\nsignificant criticism or comment about the author’s plotting technique.”).\\n\\f26 VHT, INC. V. ZILLOW GROUP\\n\\nId. (citations omitted). Because transformation advances\\ncopyright’s core goals, “the more transformative the new\\nwork, the less will be the significance of other factors.” Id.\\nLikewise, despite the absence of a textual hook, public\\npurpose also has been read into the statute. See Amazon, 508\\nF.3d at 1166; Kelly, 336 F.3d at 820. While we can discern\\ncertain animating principles bridging cases in this area, the\\ndoctrine has hardly followed a straight, or even slightly\\ncurved, line.\\n\\n The focus of the parties’ debate here is whether Zillow’s\\ntagging of 3,921 VHT photos for searchable functionality on\\nDigs was transformative and thus supported a finding of fair\\nuse. The purpose of Digs is to permit users to search for\\ncertain attributes or features, such as a marble countertop or\\nhardwood floor, and view photos of rooms with those\\nattributes or features. These photos are either uploaded by\\nusers to Digs, or selected manually or electronically by\\nZillow. Zillow then tags the photos to make them\\nsearchable. Of course, tagging makes it possible for a user’s\\nkeyword search to produce relevant results. Zillow refers to\\nthese tagged photos as “searchable images” or components\\nof the “searchable set.” VHT’s 3,921 photos are in the\\nsearchable set.\\n\\n Zillow contends that Digs is effectively a search engine,\\nwhich makes its use of VHT’s photos transformative, and\\ntherefore fair use. VHT responds that this is not fair use\\nbecause Digs is not a search engine and the tagging for\\nsearchable functionality is not transformative. Dueling\\n“search engine” characterizations do not resolve fair use\\nhere. Instead, we step back and assess the question\\nholistically, as we have been instructed to do by the statute\\nand the Supreme Court. We consider the reality of what is\\n\\f VHT, INC. V. ZILLOW GROUP 27\\n\\nhappening rather than resorting to labels. To do that, it is\\nhelpful to recount the history of the search engine cases.\\n\\n b. Evolution of Search Engine Cases\\n\\n Over the past two decades, search engines have emerged\\nas a significant technology that may qualify as a\\ntransformative fair use, making images and information that\\nwould otherwise be protected by copyright searchable on the\\nweb. See Amazon, 508 F.3d at 1166–67; Kelly, 336 F.3d at\\n818–22. In assessing fair use in the context of search\\nengines, courts have relied heavily on the first fair use factor,\\nand in particular “whether and to what extent the new work\\nis transformative.” Campbell, 510 U.S. at 579 (citation and\\nquotation omitted); see also Amazon, 508 F.3d at 1164\\n(explaining the Kelly court relied “primarily” on the first fair\\nuse factor when conducting its analysis); Authors Guild v.\\nGoogle, Inc., 804 F.3d 202, 220–221, 223 (2d Cir. 2015)\\n(offering a relatively abbreviated consideration of the\\nremaining three fair use factors, all of which were informed\\nby its analysis of the first factor).\\n\\n In an early opinion applying fair use principles in the\\ndigital age, we held that the now-defunct search engine\\nArriba’s creation and use of thumbnail versions of a\\nprofessional photographer’s copyrighted images was fair use\\nbecause the “smaller, lower-resolution images . . . served an\\nentirely different function than [the] original images.” Kelly,\\n336 F.3d at 815, 818. The original images served an artistic\\nor aesthetic purpose. Id. at 819. By contrast, the thumbnail\\nimages, which were provided in response to a user’s search\\nquery, were incorporated into the search engine’s overall\\nfunction “to help index and improve access to images on the\\ninternet and their related web sites.” Id. at 818. Investing\\nthe images with a new purpose made Arriba’s use\\n\\f28 VHT, INC. V. ZILLOW GROUP\\n\\ntransformative, not superseding. Indeed, the thumbnail\\nversions could not supersede the original use because the\\nthumbnails were grainy and low-resolution when enlarged.\\nId. Additionally, Arriba’s use of the thumbnail images\\n“promote[d] the goals of the Copyright Act and the fair use\\nexception” because they “benefit[ed] the public by\\nenhancing information-gathering techniques on the\\ninternet.” Id. at 820. Just as Campbell had drawn out the\\nprinciple of transformation from the first statutory factor, we\\ndrew out the principle of public benefit.\\n\\n Building on our reasoning in Kelly, in Amazon we held\\nthat Google’s use of thumbnail images in its search engine\\nis “highly transformative” and thus fair use. 508 F.3d at\\n1163–65. As in Kelly, we concluded that “a search engine\\nprovides social benefit by incorporating an original work\\ninto a new work, namely, an electronic reference tool.” Id.\\nat 1165. On a scale much greater than the search engine at\\nissue in Kelly, Google “improve[s] access to images on the\\ninternet and their related web sites” by “index[ing]” the\\ninternet and linking to the original source image generated\\nin the search results. Kelly, 336 F.3d at 815–16, 818. By\\nusing the thumbnail images in service of the search engine,\\nGoogle “transforms the image,” which might have been\\ncreated for an “entertainment, aesthetic, or informative\\nfunction,” “into a pointer directing a user to a source of\\ninformation.” Amazon, 508 F.3d at 1165. As a result of the\\nnew function that the image serves, Google’s use of the\\nentire image in its search engine results “does not diminish\\nthe transformative nature of [its] use.” Id. And, further\\ndeveloping the public benefit principle from Kelly, we\\nemphasized that Google’s search engine both “promotes the\\npurposes of copyright and serves the interests of the public,”\\nwhich significantly outweighed the superseding or\\n\\f VHT, INC. V. ZILLOW GROUP 29\\n\\ncommercial uses of the search engine, and strongly\\nsupported finding fair use. Id. at 1166.\\n\\n More recently, the Second Circuit considered whether\\nfair use protected the Google Books search engine, which\\nemploys digital, machine-readable copies of millions of\\ncopyright-protected books scanned by Google. Authors\\nGuild, 804 F.3d at 207–08. 9 The Google Books search\\nengine enables a full text search, which makes possible\\nsearching for a specific term, and then provides “snippets,”\\nor a part of a page, for users to read. Id. at 208–09, 216–17.\\nThe court held that both functions involve a “highly\\ntransformative purpose of identifying books of interest to the\\nsearcher.” Id. at 218. The search function “augments public\\nknowledge by making available information about\\nPlaintiffs’ books without providing the public with a\\nsubstantial substitute.” Id. at 207. And the search engine\\nmakes possible a new type of research known as “text\\nmining” or “data mining,” whereby users can search across\\nthe corpus of books to determine the frequency of specified\\nterms across time. Id. at 209, 217. Additionally, the\\n“snippet” view provides context for users to assess if a book\\nis relevant to them, without providing so much context as to\\nsupersede the original. Id. at 218. To boot, Google often\\nprovides a link to a page where the entire book can be found\\nat a library or purchased. Id. at 209. Concluding that\\nGoogle’s commercial motivation did not significantly\\noutweigh these transformative uses, the court held that the\\nfirst factor strongly supported a finding of fair use. Id. at\\n219.\\n\\n\\n 9\\n The Google Books search engine also featured millions of public\\ndomain texts. For obvious reasons, fair use was not at issue with those\\nworks.\\n\\f30 VHT, INC. V. ZILLOW GROUP\\n\\n What we divine from these cases is that the label “search\\nengine” is not a talismanic term that serves as an on-off\\nswitch as to fair use. Rather, these cases teach the\\nimportance of considering the details and function of a\\nwebsite’s operation in making a fair use determination. We\\nnow examine Digs with those lessons in mind.\\n\\n c. Application of Fair Use Principles\\n\\n As noted, the first factor assesses the character of the use,\\nincluding whether it is commercial in nature, and, critically,\\nwhether it is “transformative.” There is no dispute that\\nZillow’s use is for commercial purposes, a factor we cannot\\nignore. To determine if that use is transformative, we\\nconsider whether and to what extent it “adds something new,\\nwith a further purpose or different character, altering the first\\nwith new expression, meaning, or message[.]” Campbell,\\n510 U.S. at 579. Though we agree that Digs is a type of\\nsearch engine because it offers searchable functionality, it is\\nqualitatively different than Google and other open-universe\\nsearch engines, as well as different than the Google Books\\nsearch engine.\\n\\n Most simply, a search engine is a software program that\\nenables information retrieval by helping users find\\ninformation through the use of keyword queries. But not all\\nsearch engines are created equal. The search engines\\ncommonly used for day-to-day research are internet-wide\\nsearch engines, like Google, Yahoo, or Bing. These search\\nengines are programs powered by algorithms that search or\\n“crawl” the web. A search engine like Google then indexes\\nwebsites, stores them on a database, and runs users’ search\\nqueries against it. Search results are typically a mix of\\nimages and text, which include hyperlinks to sources of that\\ncontent elsewhere on the web. Amazon, 508 F.3d at 1155.\\n\\f VHT, INC. V. ZILLOW GROUP 31\\n\\n Unlike the internet-wide search engines considered in\\nAmazon and Kelly, Digs is a closed-universe search engine\\nthat does not “crawl” the web. Users can run searches on the\\n“searchable set” of images within Digs’ walled garden,\\nwhich includes VHT photos. The search results do not direct\\nusers to the original sources of the photos, such as VHT’s\\nwebsite. Rather, they link to other pages within Zillow’s\\nwebsite and, in some cases, to third-party merchants that sell\\nitems similar to those featured in the photo.\\n\\n That Digs makes these images searchable does not\\nfundamentally change their original purpose when produced\\nby VHT: to artfully depict rooms and properties.\\nAdditionally, Digs displays the entire VHT image, not\\nmerely a thumbnail. Unlike in Amazon, the new image does\\nnot serve a “different function” than the old one. Amazon,\\n508 F.3d at 1165. Zillow’s use preserves the photos’\\n“inherent character.” Monge, 688 F.3d at 1176. And Zillow\\n“simply supersed[es] [VHT’s] purpose” in creating the\\nimages in the first place. Kelly, 336 F.3d at 819–20; see\\nHarper & Row Publishers, Inc. v. Nation Enters., 471 U.S.\\n539, 550–51 (1985) (holding that if a new work\\n“supersede[s] the use of the original,” it is probably not a fair\\nuse).\\n\\n Comparing Digs to the Google Books search engine\\nfurther drives home this analysis. We agree with the Second\\nCircuit’s observation that the copyright dispute over the\\nGoogle Books search engine “tests the boundaries of fair\\nuse.” Authors Guild, 804 F.3d at 206. We conclude that\\nDigs goes one step further—and crosses the line.\\n\\n Like Digs, the Google Books search engine operates on\\na closed database comprised of complete digital copies of\\noriginal works. Id. at 217. But the similarities end there.\\n\\f32 VHT, INC. V. ZILLOW GROUP\\n\\nThe Google Books project makes it possible to search books\\nfor “identifying information instantaneously supplied [that]\\nwould otherwise not be obtainable in lifetimes of searching.”\\nId. at 209. With this broad purpose in mind, the court\\nrightfully observed that this system “augments public\\nknowledge.” Id. at 207. This rationale bears no comparison\\nto Digs.\\n\\n Nor is the limited transformation present on Digs\\nremotely comparable to the unprecedented text mining, word\\npattern, frequency of use, and other statistical analyses made\\npossible for the first time by Google Books. Google Books\\nsearch results provide “[a] brief description of each book,\\nentitled ‘About the Book,’” as well as, for some books, links\\nfor borrowing or purchasing the book. Id. at 209.\\nSignificantly, a Google Books search produces only a\\n“snippet” of the book, and sometimes it “disables snippet\\nview entirely.” Id. at 210. At the request of a rights holder,\\nGoogle “will exclude any book altogether from snippet\\nview.” Id.\\n\\n These features, in conjunction with other creative aspects\\nof Google Books, result in a categorically more\\ntransformative use than Zillow’s simple tagging and query\\nsystem that displays full-size copyrighted images serving the\\nsame purpose as the originals, with no option to opt out of\\nthe display, and with few, if any, transformative qualities.\\nAny transformation by Zillow pales in comparison to the\\nuses upheld in prior search engine cases. Such use also does\\nnothing to further the use of copyrighted works for the\\nsocially valuable purposes identified in the Copyright Act\\nitself, like “criticism, comment, news reporting,\\nteaching . . . , scholarship, or research.” 17 U.S.C. § 107;\\nsee also 4 NIMMER ON COPYRIGHT § 13.05[A]. The lack of\\ntransformation is especially significant because, as Kelly\\n\\f VHT, INC. V. ZILLOW GROUP 33\\n\\nteaches, “[t]he more transformative the new work, the less\\nimportant the other factors, including commercialism,\\nbecome.” 336 F.3d at 818 (citing Campbell, 510 U.S. at\\n579). If, as the Second Circuit suggested, Google’s use\\n“tests the boundaries of fair use,” Zillow’s efforts push Digs\\ninto the outer space of fair use. Authors Guild, 804 F.3d at\\n206. So while Google Books may inch across the boundary\\nof fair use, Zillow’s use does not approach the line.\\n\\n Our decisions in Amazon and Kelly provide a roadmap\\nfor analyzing the second factor, which focuses on the nature\\nof the copyrighted work. In those cases, we held that\\nphotographers’ images are creative, especially when they are\\ncreated for public viewing. Amazon, 508 F.3d at 1167;\\nKelly, 336 F.3d at 820. “Works that are creative in nature\\nare ‘closer to the core of intended copyright protection’ than\\nare more fact-based works.” Napster, 239 F.3d at 1016\\n(quoting Campbell, 510 U.S. at 586).\\n\\n So too here. VHT’s photos are aesthetically and\\ncreatively shot and edited by professional photographers.\\nThat Zillow’s curators select the most creative photos for the\\nDigs searchable set underscores the creative nature of the\\nworks. But, as the district court properly noted, this factor\\noperates “with less force” in favor of VHT because the\\nphotos had already been published on the Listing Platform.\\nSee Kelly, 336 F.3d at 820 (“The fact that a work is published\\nor unpublished also is a critical element of its nature.\\nPublished works are more likely to qualify as fair use\\nbecause the first appearance of the artist’s expression has\\nalready occurred.”). Ultimately, this factor only slightly\\nfavors VHT, further cutting against finding fair use.\\n\\n The third factor evaluates the amount and substantiality\\nof the copyrighted work that was used. “[C]opying an entire\\n\\f34 VHT, INC. V. ZILLOW GROUP\\n\\nwork militates against a finding of fair use.” Worldwide\\nChurch of God v. Phila. Church of God, Inc., 227 F.3d 1110,\\n1118 (9th Cir. 2000) (quotation and citation omitted).\\nHowever, this analysis is informed by the “purpose of the\\ncopying.” Campbell, 510 U.S. at 586. In that spirit, we have\\nfound that copying full works qualifies as fair use where “[i]t\\nwas necessary . . . to copy the entire image to allow users to\\nrecognize the image and decide whether to pursue more\\ninformation about the image or the originating web site.”\\nKelly, 336 F.2d at 821. In contrast to Amazon and Kelly,\\nnothing justifies Zillow’s full copy display of VHT’s photos\\non Digs.\\n\\n Finally, the fourth factor considers “the effect of the use\\nupon the potential market for or value of the copyrighted\\nwork.” 17 U.S.C. § 107(4). To defeat a fair use defense,\\n“one need only show that if the challenged use should\\nbecome widespread, it would adversely affect the potential\\nmarket for the copyrighted work.” Harper & Row\\nPublishers, 471 U.S. at 568 (citation and quotation omitted).\\n\\n Although VHT had licensed only a handful of photos for\\nsecondary uses (and none on a searchable database), that\\nmarket was more than “hypothetical.” See Amazon, 508\\nF.3d at 1168. Significantly, VHT was “actively exploring”\\nthe market for licensing its photos to home design websites\\nlike Digs—including with Zillow itself. This factor favors\\nVHT.\\n\\n Taken together, the nature of Zillow’s use, when\\nintegrated with the four factors, cuts against finding fair use\\nby Zillow. We affirm the district court’s grant of summary\\njudgment to VHT with respect to fair use.\\n\\f VHT, INC. V. ZILLOW GROUP 35\\n\\n II. Secondary Infringement—Digs\\n\\n Turning to VHT’s claim for secondary infringement, the\\ndistrict court correctly concluded that Zillow did not\\nsecondarily infringe VHT’s exclusive rights in the 28,125\\nphotos used on Digs, aside from 114 images created on Digs\\nafter VHT specifically identified them, which are not on\\nappeal. As noted before with regard to the district court’s\\nruling on direct infringement, “we apply the same standard\\nused by the district court in evaluating the jury’s verdict” and\\nuphold the verdict unless “the evidence permits only one\\nreasonable conclusion, and that conclusion is contrary to the\\njury’s verdict.” Wallace, 479 F.3d at 624; Fed. R. Civ. P.\\n50(a). The jury’s verdict here did not meet this standard.\\nWe affirm the district court’s grant of Zillow’s motion for\\njudgment notwithstanding the verdict with respect to\\nsecondary infringement, both contributory and vicarious\\ninfringement.\\n\\n A. Contributory Liability\\n\\n Contributory liability requires that a party “(1) has\\nknowledge of another’s infringement and (2) either\\n(a) materially contributes to or (b) induces that\\ninfringement.” Perfect 10, Inc. v. Visa Int’l Serv., Ass’n, 494\\nF.3d 788, 795 (9th Cir. 2007). Zillow’s actions do not satisfy\\nthe second prong—material contribution or inducement—so\\nwe do not address the first prong. See Giganews, 847 F.3d\\nat 671.\\n\\n In Giganews, we outlined the means of material\\ncontribution to infringement:\\n\\n In the online context, . . . a “computer system\\n operator” is liable under a material\\n contribution theory of infringement “if it has\\n\\f36 VHT, INC. V. ZILLOW GROUP\\n\\n actual knowledge that specific infringing\\n material is available using its system, and can\\n take simple measures to prevent further\\n damage to copyrighted works, yet continues\\n to provide access to infringing works.”\\n\\nId. at 671 (quoting Amazon, 508 F.3d at 1172); see also\\nEllison v. Robertson, 357 F.3d 1072, 1078 (9th Cir. 2004)\\n(applying this standard in the online context); Napster, 239\\nF.3d at 1022 (same); Religious Tech. Ctr. v. Netcom On-Line\\nComm’n Servs., Inc., 907 F. Supp. 1361, 1375 (N.D. Cal.\\n1995) (same). There is insufficient evidence to support the\\ncontributory infringement verdict under the “simple\\nmeasures” standard.\\n\\n VHT’s position that “the jury could have reasonably\\ndecided that Zillow in fact had the means to identify and\\nremove” the allegedly infringing images that VHT identified\\nby property address, as opposed to their website designation\\nor Uniform Resource Locator (“URL”), fails. Zillow\\ntestified throughout trial that, in order to systematically or\\nswiftly take down a large number of photos, it required the\\nZillow Image ID–a number that is in the URL for each\\nimage. This stands to reason, because “Zillow receives\\nmultiple copies of the same photograph, depicting the same\\nproperty, with the same listing agent, from different feeds.”\\nMerely identifying the physical property address in no way\\nidentified the proper feed or the correct photo. Thus, Zillow\\ndid not have appropriately “specific” information necessary\\nto take “simple measures” to remedy the violation.\\n\\n VHT’s argument that Zillow is liable for failing to ask\\nfor the URLs of the allegedly infringing photos also fails.\\nAsking for the URLs was not Zillow’s duty under the\\ncontributory liability standard: Zillow must have “actual\\n\\f VHT, INC. V. ZILLOW GROUP 37\\n\\nknowledge that specific infringing material is available using\\nits system.” Amazon, 508 F.3d at 1172 (citation and\\nquotation omitted). Zillow first reasonably asked to see the\\nlicenses between VHT and the feed providers; otherwise,\\nZillow could not assess ownership and rights in the\\nundefined images. That Zillow did not proactively request a\\nlist of URLs before VHT filed suit does not make Zillow\\ncontributorily liable.\\n\\n Additionally, Zillow’s failure to systematically use\\nwatermarking technology does not show there was a “simple\\nmeasure” available that it failed to use. Even assuming there\\nwere “reasonable and feasible means” for Zillow to employ\\nwatermark detection technology, in practice VHT rarely\\nwatermarked its photos. See Giganews, 847 F.3d at 672\\n(citation and quotation omitted).\\n\\n Nor did Zillow induce infringement. Inducement\\nliability requires evidence of “active steps . . . taken to\\nencourage direct infringement.” Metro-Goldwyn-Mayer\\nStudios Inc. v. Grokster, Ltd., 545 U.S. 913, 936 (2005)\\n(citation and quotation omitted)). Evidence of active steps\\nincludes “advertising an infringing use or instructing how to\\nengage in an infringing use,” because such evidence\\n“show[s] an affirmative intent that the product be used to\\ninfringe.” Id.; see also Columbia Pictures Industries, Inc. v.\\nFung, 710 F.3d 1020, 1032 (9th Cir. 2013) (inducement\\nliability requires “an object of promoting [the product’s] use\\nto infringe copyright”). The “improper object” of\\ninfringement “must be plain and must be affirmatively\\ncommunicated through words or actions[.]” Fung, 710 F.3d\\nat 1034.\\n\\n In view of the evidence, “no reasonable juror could\\nconclude [Zillow] distributed its product with the object of\\n\\f38 VHT, INC. V. ZILLOW GROUP\\n\\npromoting its use to infringe copyright.” Giganews, 847\\nF.3d at 672 (quotation and citation omitted). For example,\\nZillow’s generally applicable tools and messages for users\\nto save more photos from the Listing Platform to Digs does\\nnot “promote[] the use of [Digs] specifically to infringe\\ncopyrights.” Amazon, 508 F.3d at 1170 n.11. Nor does\\nevidence that Zillow sometimes makes mistakes about the\\ndisplay rights in a feed plainly communicate an improper\\nobject of infringement. Zillow corrects these inadvertent\\nerrors when it learns of them. Because a “failure to take\\naffirmative steps to prevent infringement” alone cannot\\ntrigger inducement liability, the inducement claim is a\\nparticularly poor fit for Zillow’s real estate and home design\\nwebsites, which have “substantial noninfringing uses.”\\nGrokster, 545 U.S. at 939 n.12.\\n\\n B. Vicarious Liability\\n\\n Neither does the vicarious liability theory fit the Zillow\\nplatform. To prevail on a vicarious liability claim, “[VHT]\\nmust prove ‘[Zillow] has (1) the right and ability to supervise\\nthe infringing conduct and (2) a direct financial interest in\\nthe infringing activity.’” Giganews, 847 F.3d at 673\\n(quoting Visa, 494 F.3d at 802). The first element requires\\n“both a legal right to stop or limit the directly infringing\\nconduct, as well as the practical ability to do so.” Amazon,\\n508 F.3d at 1173. VHT’s vicarious infringement argument\\nfails because, as the district court found, “Zillow ‘lack[ed]\\nthe practical ability to police’ its users’ infringing conduct”\\non Digs.\\n\\n As discussed with respect to contributory infringement,\\nthere was insufficient evidence that Zillow had the technical\\nability to screen out or identify infringing VHT photos\\namong the many photos that users saved or uploaded daily\\n\\f VHT, INC. V. ZILLOW GROUP 39\\n\\nto Digs. Regardless, such allegations do not fall under the\\nvicarious liability rubric: Zillow’s “failure to change its\\noperations to avoid assisting [users] to distribute . . .\\ninfringing content . . . is not the same as declining to exercise\\na right and ability to make [third parties] stop their direct\\ninfringement.” Amazon, 508 F.3d at 1175.\\n\\n Our conclusion is consistent with earlier dicta that “the\\nvicarious liability standard applied in Napster can be met by\\nmerely having the general ability to locate infringing\\nmaterial and terminate users’ access.” UMG Recordings,\\nInc. v. Shelter Capital Partners LLC, 718 F.3d 1006, 1030\\n(9th Cir. 2013) (emphasis added). Once VHT photos were\\nuploaded to the Listing Platform with appropriate\\ncertification of rights, ferreting out claimed infringement\\nthrough use on Digs was beyond hunting for a needle in a\\nhaystack. As the district court concluded, “the trial record\\nlacks substantial evidence of a practical ability to limit direct\\ninfringement for the same reasons it lacks substantial\\nevidence of simple measures to remove infringing material.”\\nAnd linking a claimed infringement to a feed provider was\\neven more of an impossibility.\\n\\n We affirm the district court’s judgment notwithstanding\\nthe verdict concluding that substantial evidence did not\\nsupport the claim that Zillow secondarily infringed VHT’s\\nexclusive rights in its photos.\\n\\n III. Damages\\n\\n A. Compilation\\n\\n The size of the damages award hinges on whether VHT’s\\nphotos used on Digs are part of a “compilation” or if they are\\nindividual photos. This distinction makes a difference. If\\nthe VHT photo database is a “compilation,” and therefore\\n\\f40 VHT, INC. V. ZILLOW GROUP\\n\\none “work” for the purposes of the Copyright Act, then VHT\\nwould be limited to a single award of statutory damages for\\nZillow’s use of thousands of photos on Digs. 17 U.S.C.\\n§ 504(c)(1). But if the database is not a compilation, then\\nVHT could seek damages for each photo that Zillow used.\\n\\n In lieu of actual damages, a copyright holder may elect\\nto receive statutory damages under the Copyright Act. 17\\nU.S.C. § 504(c)(1). VHT did so. The Act provides for “an\\naward of statutory damages for all infringements involved in\\nthe action, with respect to any one work . . . in a sum of not\\nless than $750 or more than $30,000.” Id. This provision\\nties statutory damages to the term “work,” which is\\nundefined, except in a circular manner: copyright law\\nprotects “original works of authorship.” 17 U.S.C. § 102(a).\\nFortunately, there is a definition of compilation: “a work\\nformed by the collection and assembling of preexisting\\nmaterials or of data that are selected, coordinated, or\\narranged in such a way that the resulting work as a whole\\nconstitutes an original work of authorship.” 17 U.S.C.\\n§ 101. For purposes of statutory damages, “all the parts of a\\ncompilation . . . constitute one work.” 17 U.S.C.\\n§ 504(c)(1). “The question of whether a work constitutes a\\n‘compilation’ for the purposes of statutory damages pursuant\\nto Section 504(c)(1) of the Copyright Act is a mixed question\\nof law and fact.” Bryant v. Media Right Prods., Inc., 603\\nF.3d 135, 140 (2d Cir. 2010).\\n\\n Whether various VHT photo collections comprise one or\\nmore compilations is a threshold damages question. Before\\ntrial, Zillow asked for a legal determination on the\\ncompilation issue. That motion was denied. However, the\\ndistrict court did not make an explicit determination about\\ncompilation and the specifics of compilation were not put\\nbefore the jury. In fairness to the district court, we might\\n\\f VHT, INC. V. ZILLOW GROUP 41\\n\\ninfer from the transcript that there was an implicit\\ndetermination as to whether VHT’s photos are part of a\\ncompilation, but we are left in doubt. Instead, the jury was\\ninstructed that “[e]ach VHT Photo that has independent\\neconomic value constitutes a separate work.” On the verdict\\nform, the jury was asked which “photographs have\\nindependent economic value.”\\n\\n The notion of “independent economic value” derives not\\nfrom the statute, but from case law. In the Ninth Circuit, the\\nquestion of whether something—like a photo, television\\nepisode, or so forth—has “independent economic value”\\ninforms our analysis of whether the photo or episode is a\\nwork, though it is not a dispositive factor. See Columbia\\nPictures Television, Inc. v. Krypton Broad. of Birmingham,\\nInc., 259 F.3d 1186, 1193 (9th Cir. 2001). But consideration\\nof the independent economic value factor does not answer\\nthe question whether something is a compilation. That\\nquestion remains unanswered here.\\n\\n On appeal, the parties have polar opposite views on\\nwhether there was a compilation (Zillow’s position) or\\nwhether each photo is entitled to a separate damages award\\n(VHT’s position). VHT registered thousands of photos as\\ncompilations. But the Copyright Office warns that such a\\nregistration “may” limit the copyright holder “to claim only\\none award of statutory damages in an infringement action,\\neven if the defendant infringed all of the component works\\ncovered by the registration.” U.S. COPYRIGHT OFFICE,\\nCOMPENDIUM OF U.S. COPYRIGHT OFFICE PRACTICES\\n§ 1104.5 (Sept. 2017), https://www.copyright.gov/comp3/\\ndocs/compendium.pdf. Though the registration label is not\\ncontrolling, it may be considered by the court when\\nassessing whether a work is a compilation. See Yellow\\nPages Photos, Inc. v. Ziplocal, LP, 795 F.3d 1255, 1277\\n\\f42 VHT, INC. V. ZILLOW GROUP\\n\\n(11th Cir. 2015) (“Although the manner of copyright\\nregistration is not dispositive of the works issue, this Court\\nhas previously considered it to be at least a relevant factor.”).\\nUltimately, what counts is the statutory definition.\\n\\n Because there were at least ten different copyright\\nregistrations, thousands of photos, and no explicit\\ndetermination on compilation, we decline to sort out the\\ncompilation issue on appeal. We remand to the district court\\nfor further proceedings as to whether the VHT photos\\nremaining at issue were a compilation.\\n\\n B. Willfulness\\n\\n The jury found that Zillow willfully infringed exclusive\\nrights to 3,373 searchable VHT photos that were eligible for\\nstatutory damages. The district court largely upheld the\\nwillfulness finding in its post-trial motions order. However,\\nthe court granted Zillow judgment notwithstanding the\\nverdict on 673 images that were not displayed, so the court’s\\nfinal willfulness judgment applied to 2,700 searchable\\nphotos on Digs.\\n\\n To uphold a jury’s willfulness finding, there must be\\nsubstantial evidence “(1) that the [the infringing party] was\\nactually aware of the infringing activity, or (2) that the\\n[infringing party’s] actions were the result of reckless\\ndisregard for, or willful blindness to, the copyright holder’s\\nrights.” Unicolors, 853 F.3d at 991 (citation and quotation\\nomitted). Under the second prong, willful blindness requires\\nthat the infringing party “(1) subjectively believed that\\ninfringement was likely occurring on their networks and that\\nthey (2) took deliberate actions to avoid learning about the\\ninfringement.” Luvdarts, LLC v. AT&T Mobility, LLC, 710\\nF.3d 1068, 1073 (9th Cir. 2013). Reckless disregard can be\\ndemonstrated, for example, when a party “refus[es], as a\\n\\f VHT, INC. V. ZILLOW GROUP 43\\n\\nmatter of policy, to even investigate or attempt to determine\\nwhether particular [photos] are subject to copyright\\nprotections.” Unicolors, 853 F.3d at 992. A finding of\\nwillfulness has significant financial consequences—the jury\\nmay increase damages up to $150,000 per violation. See 17\\nU.S.C. § 504(c)(2).\\n\\n As noted with respect to infringement, we do not take\\nlightly the decision to reverse a jury verdict, nor do we\\ncavalierly set aside the district court’s thoughtful analysis.\\nBut here, we are compelled to disagree with both because\\nsubstantial evidence does not support willfulness as to the\\n2,700 photos.\\n\\n The test for willfulness is in the alternative: either actual\\nnotice or recklessness shown by reckless disregard or\\nturning a blind eye to infringement. We turn first to actual\\nnotice. That VHT provided Zillow with minimal notice of\\ninfringement does not itself establish that any subsequent\\ninfringement was willful. Rather, “[c]ontinued use of a work\\neven after one has been notified of his or her alleged\\ninfringement does not constitute willfulness so long as one\\nbelieves reasonably, and in good faith, that he or she is not\\ninfringing.” Evergreen Safety Council v. RSA Network Inc.,\\n697 F.3d 1221, 1228 (9th Cir. 2012); see also Danjaq LLC\\nv. Sony Corp., 263 F.3d 942, 959 (9th Cir. 2001) (“It would\\nseem to follow that one who has been notified that his\\nconduct constitutes copyright infringement, but who\\nreasonably and in good faith believes the contrary, is not\\n‘willful’ for these purposes.”) (quoting 4 NIMMER ON\\nCOPYRIGHT § 14.04[B][3]).\\n\\n Such is the case here. Zillow’s agreements with its feed\\nproviders grant it an express license to use, copy, distribute,\\npublicly display, and create derivative works for each photo,\\n\\f44 VHT, INC. V. ZILLOW GROUP\\n\\nand the agreements include unambiguous representations by\\nthe feed providers that they have the authority to assign such\\nrights. Zillow developed procedures to identify ex ante the\\nscope of its license for each uploaded photo and employed\\nautomated protocols to manage the use of each photo\\nconsistent with its evergreen or deciduous designation. At\\nno point during their year of communications prior to\\nissuance of the notice letter did VHT raise the specter of\\ninfringement. Notably, VHT’s eventual notice was minimal:\\none letter with a list of allegedly infringing photos,\\ndesignated by residential street address, not web address.\\n\\n The notion that Zillow failed to take appropriate\\nresponsive measures after receiving this notice is belied by\\nthe record. Zillow immediately requested information to\\nconfirm VHT’s copyright ownership and cross-reference the\\nphotos with licensing information. VHT was not\\nforthcoming with that information. Rather, in response,\\nVHT offered merely an unsigned form contract. Instead of\\nproviding helpful information, VHT then filed suit. Given\\nthe limited information provided by VHT, Zillow could not\\nreasonably be expected to have promptly and unilaterally\\nremoved each flagged photo. As the district court noted,\\nVHT failed to demonstrate there were simple measures\\navailable for the removal of infringing photos or that Zillow\\nhad any “practical ability to independently identify\\ninfringing images.”\\n\\n Thus, we are compelled to conclude that substantial\\nevidence does not show Zillow was “actually aware” of its\\ninfringing activity. See Evergreen Safety Council, 697 F.3d\\nat 1228. Zillow’s belief that feed providers had properly\\nlicensed its uses and that its system effectively respected\\nthose rights was reasonable. And, as the district court\\nobserved, “[t]he record suggests no reason to conclude that\\n\\f VHT, INC. V. ZILLOW GROUP 45\\n\\nZillow maintained that position in bad faith, and Zillow’s\\nnon-infringement contention proved accurate as to most of\\nthe images at issue in this lawsuit.”\\n\\n We reach the same conclusion as to whether Zillow\\nrecklessly disregarded or willfully blinded itself to its\\ninfringement. In reaching an opposite conclusion, the\\ndistrict court observed that Zillow did not “perform[] further\\ninvestigation into the rights each [feed provider] possesses,”\\nnor did it “t[ake] responsive measures to obtain further\\ninformation” after VHT provided the minimal notice of\\npotential infringement. That conclusion is at odds with the\\nevidence, for the reasons outlined above.\\n\\n VHT’s argument that Zillow, a sophisticated business\\nwith a robust legal team, should have known that its feed\\nprovider license agreements were invalid is unavailing.\\nVHT argues that when Zillow saw the non-exclusive grant\\nof rights in VHT’s unsigned form contract, showing that the\\nfeed providers did not have a right to sublicense, Zillow\\nshould have known the licenses were invalid. Despite\\nrequests for such information, Zillow did not have access to\\nVHT’s executed licenses with the feed providers who\\nfurnished VHT’s photos to Zillow. Access to a blank form\\ncontract (that the district court earlier found ambiguous as a\\nmatter of law) is not enough. We conclude that substantial\\nevidence does not show Zillow was “reckless or willfully\\nblind” as to its infringement. We reverse the district court\\nand vacate the jury’s finding of willful infringement.\\n\\n IV. Conclusion\\n\\n We affirm the district court’s summary judgment in\\nfavor of Zillow on direct infringement of the Listing\\nPlatform photos. With respect to direct liability on the Digs\\nphotos, we affirm the district court’s grant of judgment\\n\\f46 VHT, INC. V. ZILLOW GROUP\\n\\nnotwithstanding the verdict on the 22,109 non-displayed\\nphotos and the 2,093 displayed but not searchable photos.\\nWe uphold summary judgment in favor of VHT on the 3,921\\ndisplayed, searchable photos.\\n\\n We affirm the district court’s judgment notwithstanding\\nthe verdict on secondary liability, both contributory and\\nvicarious, on the Digs photos.\\n\\n We reverse the district court’s denial of judgment\\nnotwithstanding the verdict on the issue of willfulness and\\nvacate the jury’s finding on willfulness.\\n\\n We remand consideration of the compilation issue to the\\ndistrict court.\\n\\n Affirmed in part, reversed in part and remanded.\\nEach party shall pay its own costs on appeal.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1294,"cells":{"id":{"kind":"number","value":4603415,"string":"4,603,415"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. Kyle Korte"},"decision_date":{"kind":"string","value":"2019-03-15"},"docket_number":{"kind":"string","value":"18-50051"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Dorothy W. Nelson, Consuelo M. Callahan, and John B. Owens, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"OWENS, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 18-50051\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 8:16-cr-00156-\\n JLS-1\\nKYLE JASON KORTE,\\n Defendant-Appellant. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Josephine L. Staton, District Judge, Presiding\\n\\n Argued and Submitted February 11, 2019\\n Pasadena, California\\n\\n Filed March 15, 2019\\n\\n Before: Dorothy W. Nelson, Consuelo M. Callahan, and\\n John B. Owens, Circuit Judges.\\n\\n Opinion by Judge Owens;\\n Concurrence by Judge D.W. Nelson\\n\\n2 UNITED STATES V. KORTE\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n The panel affirmed the district court’s denial of the\\ndefendant’s suppression motions and his convictions for\\nbank robbery in a case in which the defendant, who was on\\nparole during his crime spree, had consented to suspicionless\\nsearches of his person, residence, and any property under his\\ncontrol.\\n\\n Rejecting the defendant’s contention that a lawful parole\\nsearch of his car does not extend to the trunk because the\\ntrunk is not “property under his control,” the panel held that\\nthe parole search of the defendant’s trunk was lawful.\\n\\n Following precedent distinguishing Fourth Amendment\\nrights in the parolee context, the panel held that the\\nwarrantless placement of a GPS tracker on the defendant’s\\ncar did not violate the Fourth Amendment.\\n\\n The panel held that cell site location information (CSLI)\\nacquired before the Supreme Court issued its decision in\\nCarpenter v. United States, 138 S. Ct. 2206 (2018) (holding\\nthat the Government must obtain a warrant to access a\\nperson’s CSLI from a wireless carrier), is admissible under\\nthe good-faith exception to the exclusionary rule so long as\\nthe Government satisfied the Stored Communications Act’s\\nthen-lawful requirements. The panel took no stance on the\\nconstitutionality of acquiring a parolee’s CSLI without a\\nwarrant.\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n UNITED STATES V. KORTE 3\\n\\n Concurring in both the reasoning and the result, Judge\\nNelson wrote separately because she is concerned with the\\never “diminishing” reasonable expectation of privacy\\nafforded to probationers and parolees, especially as it relates\\nto digital privacy.\\n\\n\\n COUNSEL\\n\\nH. Dean Steward (argued), Attorney at Law, San Clemente,\\nCalifornia, for Defendant-Appellant.\\n\\nJulia L. Reese (argued), Assistant United States Attorney,\\nCriminal Appeals Section; Nicola T. Hanna, United States\\nAttorney; Lawrence S. Middleton, Assistant United States\\nAttorney, Chief, Criminal Division; Gregory W. Staples,\\nAssistant United States Attorney, Santa Ana Branch Office;\\nUnited States Attorney’s Office, Los Angeles, California;\\nfor Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nOWENS, Circuit Judge:\\n\\n Defendant Kyle Korte appeals from his convictions for\\nbank robbery. During his crime spree, Korte was on parole\\nand had consented to warrantless, suspicionless searches of\\nhis person, residence, and any property under his control.\\nRelying on this parole condition, officers placed a Global\\nPositioning System (“GPS”) device on his car and later\\nsearched its trunk. Officers also obtained Korte’s historical\\ncell site location information (“CSLI”) by court order. On\\nappeal, Korte primarily challenges the district court’s denial\\n\\n4 UNITED STATES V. KORTE\\n\\nof his suppression motions as to each of these searches. We\\nhave jurisdiction under 28 U.S.C. § 1291, and we affirm.\\n\\nI. FACTUAL AND PROCEDURAL BACKGROUND\\n\\n In a world of cybercrime and identity theft, Korte stole\\nmoney the old-fashioned way — he robbed banks. After\\nserving time in state prison for bank robbery, Korte was\\nparoled in August 2016. As a parolee in California, Korte\\nwas “subject to search or seizure . . . at any time of the day\\nor night, with or without a search warrant or with or without\\ncause.” CAL. PENAL CODE § 3067(b)(3); see also id.\\n§ 3067(a). On October 25, 2016, Korte acknowledged his\\nparole conditions, including that he was now subject to\\nsearches of “[y]ou, your residence, and any property under\\nyour control.”\\n\\n In October 2016, the Los Angeles Sheriff’s Department\\n(“LASD”) began investigating a series of bank robberies.\\nThe first robbery took place on October 7. A masked robber\\nentered a bank and demanded “all your hundreds.” The\\nfrightened teller, protected by bulletproof glass, activated the\\nsilent alarm and retreated to a back office. The robber left\\nwith no money. On October 12, the masked robber targeted\\nanother bank, this time brandishing a toy gun. He was more\\nsuccessful this go-around, escaping with $1,600. He then hit\\ntwo more banks on October 27. Again displaying the toy\\ngun, the robber pocketed $2,200 and $7,000. In total, the\\nmasked robber stole less than $11,000 — not a Neil\\nMcCauley heist by any means.\\n\\n Working with LASD, the Federal Bureau of\\nInvestigation (“FBI”) began to suspect that Korte was the\\nmasked robber. Surveillance video from one of the robberies\\nshowed a car registered to the address that Korte provided to\\nhis parole officer. An LASD officer who saw video of the\\n\\n UNITED STATES V. KORTE 5\\n\\nmasked robber also reported that the individual resembled\\nKorte, who the officer knew was on parole for bank robbery.\\n\\n On November 4, 2016, without a warrant or Korte’s\\nconsent, LASD placed a GPS tracking device on Korte’s car\\nand periodically monitored the vehicle’s movements over\\nthe next six days. That same day, the Government obtained\\na court order under the Stored Communications Act\\n(“SCA”), 18 U.S.C. § 2703(d), to acquire Korte’s CSLI.\\nThis information placed Korte’s cell phone near three of the\\nfour banks at the time of the respective robberies.\\n\\n On November 10, 2016, LASD learned that the FBI had\\nobtained an arrest warrant for Korte. Officers followed\\nKorte as he drove from his home to a bank and parked\\nnearby, seeming to surveil his next target. An officer saw\\nKorte open the car’s trunk and place something inside.\\nShortly thereafter, they arrested Korte and searched the car.\\nThe officers’ search of the trunk revealed the toy gun used\\nduring the three armed robberies and the shirt Korte had been\\nwearing while casing the bank just prior to his arrest.\\n\\n A grand jury indicted Korte for one count of attempted\\nbank robbery in violation of 18 U.S.C. § 2113(a) and three\\ncounts of bank robbery in violation of 18 U.S.C. § 2113(a),\\n(d). Korte pled not guilty and moved to suppress (1) the\\nevidence found in his car’s trunk, (2) the information derived\\nfrom the GPS tracker on his car, and (3) his CSLI. The\\ndistrict court denied all three suppression motions. The court\\nreasoned that Korte’s parole status permitted the warrantless\\nsearch of the trunk and placement of the GPS tracker on his\\ncar. As for the CSLI, it held that even if the acquisition of\\nthis information violated Korte’s Fourth Amendment rights,\\nsee Carpenter v. United States, 138 S. Ct. 2206 (2018), the\\ngood-faith exception clearly applied based on the case law\\nat that time.\\n\\n6 UNITED STATES V. KORTE\\n\\n Korte went to trial. As to Count 1 — the unsuccessful\\nrobbery — the bank teller testified that even though\\nbulletproof glass separated her from the robber and she saw\\nno weapon, she was nevertheless “[k]ind of panicked,”\\n“[s]hocked,” and “[s]cared.” Korte filed a motion for a\\njudgment of acquittal, under Federal Rule of Criminal\\nProcedure 29, focusing on Count 1. The district court\\nrejected his argument that the Government failed to prove\\nthe element of “force and violence, or by intimidation,” as\\n18 U.S.C. § 2113(a) requires. Citing the teller’s “testimony\\nwith regard to a mask and the demand,” the court found\\nsufficient evidence of “intimidation” to send the matter to\\nthe jury.\\n\\n The jury returned a guilty verdict as to all four counts.\\nKorte was sentenced to 210 months in prison.\\n\\nII. DISCUSSION\\n\\n We review de novo the denial of Korte’s suppression\\nmotions. See United States v. Zapien, 861 F.3d 971, 974 (9th\\nCir. 2017) (per curiam). We address each search — of his\\ncar’s trunk, the GPS tracker on his car, and the acquisition\\nof his CSLI — in turn.\\n\\nA. Search of the Trunk\\n\\n We first examine whether Korte’s parole-search\\ncondition authorized the warrantless search of his car’s\\ntrunk. See United States v. Cervantes, 859 F.3d 1175, 1183\\n(9th Cir. 2017) (“A search of a parolee that complies with\\nthe terms of a valid search condition will usually be deemed\\nreasonable under the Fourth Amendment.”).\\n\\n California Penal Code section 3067(b)(3) provides that\\nevery parolee “is subject to search or seizure . . . at any time\\n\\n UNITED STATES V. KORTE 7\\n\\nof the day or night, with or without a search warrant or with\\nor without cause.” In Samson v. California, 547 U.S. 843,\\n846 (2006), the United States Supreme Court reviewed\\nCalifornia’s parole-search condition to determine “whether\\na suspicionless search, conducted under the authority of\\n[section 3067], violates the Constitution.” The Court first\\nreasoned that “parole is more akin to imprisonment.” Id. at\\n850. Because parolees “have severely diminished\\nexpectations of privacy by virtue of their status alone” —\\neven less than probationers — and because “[t]he State’s\\ninterests” in supervising parolees and reducing recidivism\\n“are substantial,” the Court upheld California’s parole-\\nsearch condition. 1 Id. at 852–53.\\n\\n Korte does not — and cannot — argue that officers\\nunlawfully searched his car. He admits that he rented the car\\nand referred to it as “my car.” As “property under [his]\\ncontrol,” California’s parole-search condition authorized the\\n 1\\n There are two limitations on this condition. First, the officer\\nconducting the search must know at the time that the individual is\\ncurrently on parole. See Moreno v. Baca, 431 F.3d 633, 641 (9th Cir.\\n2005). Second, the search cannot be “arbitrary, capricious, or\\nharassing.” See CAL. PENAL CODE § 3067(d) (“It is not the intent of the\\nLegislature to authorize law enforcement officers to conduct searches for\\nthe sole purpose of harassment.”); see also Samson, 547 U.S. at 856\\n(“The concern that California’s suspicionless search system gives\\nofficers unbridled discretion to conduct searches . . . is belied by\\nCalifornia’s prohibition on ‘arbitrary, capricious or harassing’ searches.”\\n(citations omitted)).\\n\\n Here, it is undisputed that the officers knew that Korte was on\\nparole, and there is no evidence the officers engaged in an arbitrary,\\ncapricious, or harassing search. Nor does Korte contend he was unaware\\nof the parole-search condition. See People v. Lilienthal, 587 P.2d 706,\\n711 (Cal. 1978) (“The condition itself provides general notice to a\\ndefendant that he or his belongings may be subjected to warrantless\\nsearches.”).\\n\\n8 UNITED STATES V. KORTE\\n\\nwarrantless search of it. See, e.g., United States v. Caseres,\\n533 F.3d 1064, 1075–76 (9th Cir. 2008) (explaining that a\\nsearch of the parolee’s car would have been lawful if the\\nofficer had known that the defendant was on parole).\\n\\n Yet, Korte contends that a lawful parole search of his car\\ndoes not extend to the trunk because the trunk is not similarly\\n“property under [his] control.” We reject his narrow\\ninterpretation of this condition. Property is subject to search\\nwhen a parolee “exhibit[s] a sufficiently strong connection\\nto [the property in question] to demonstrate ‘control’ over\\nit.” United States v. Grandberry, 730 F.3d 968, 980 (9th Cir.\\n2013) (defining “property under [a parolee’s] control”); see\\nalso Cervantes, 859 F.3d at 1183 (explaining that having a\\nkey to and belongings inside a hotel room sufficiently\\ndemonstrate control, even if the room is co-occupied).\\nConsistent with our understanding of the condition, the\\nSupreme Court of California explained that a parolee\\ncontrols property based on “the nexus between the parolee\\nand the area or items searched,” including the “nature of that\\narea or item” and “how close and accessible the area or item\\nis to the parolee.” People v. Schmitz, 288 P.3d 1259, 1270\\n(Cal. 2012) (holding that a parolee, who is only a passenger\\nin a third-party’s vehicle, is in control of areas within his\\nreach in the passenger compartment).\\n\\n Korte’s uncontested control over the car was sufficient\\nto permit a warrantless search of its trunk. In any event, his\\nconduct also illustrated a sufficiently close nexus to the trunk\\nitself. Officers observed him putting things inside the trunk.\\n\\n Permitting a warrantless search of the trunk of a\\nparolee’s car is also consistent with broader Fourth\\nAmendment precedent. Generally, a lawful search of a fixed\\nspace or premise extends to its entire area, whether or not\\nthat requires opening a confined space: “[N]ice distinctions\\n\\n UNITED STATES V. KORTE 9\\n\\nbetween closets, drawers, and containers, in the case of a\\nhome, or between glove compartments, upholstered seats,\\ntrunks, and wrapped packages, in the case of a vehicle, must\\ngive way to the interest in the prompt and efficient\\ncompletion of the task at hand.” United States v. Ross, 456\\nU.S. 798, 821 (1982). This reasoning applies to a car’s\\ntrunk. See, e.g., id. at 825 (“If probable cause justifies the\\nsearch of a lawfully stopped vehicle, it justifies the search of\\nevery part of the vehicle and its contents that may conceal\\nthe object of the search.”); United States v. Ewing, 638 F.3d\\n1226, 1231 (9th Cir. 2011) (“[I]f a law enforcement officer\\nhas probable cause to search a vehicle, that probable cause\\nextends to all contents in the vehicle that could be connected\\nto the suspected activity.”); United States v. McWeeney, 454\\nF.3d 1030, 1035 (9th Cir. 2006) (holding that consent to\\nsearch a defendant’s car extends to the trunk). We also note\\nthat Korte’s proposed distinction between a car and trunk\\nwas already rejected by the Supreme Court of California in\\nthe context of a probationer – who has more Fourth\\nAmendment protection than Korte as a parolee. See\\nLilienthal, 587 P.2d at 711 (“We conclude that the officers\\nwere justified in searching defendant’s car trunk pursuant to\\ndefendant’s consent to warrantless searches as a condition of\\nhis probation.”).\\n\\n As property under his control, Korte fails to explain how\\nsearching his car’s trunk would offend the Fourth\\nAmendment when a warrantless search of his home — the\\napex of constitutionally protected places — would not. See\\nUnited States v. Lopez, 474 F.3d 1208, 1213–14 (9th Cir.\\n2007) (upholding the warrantless search of a parolee’s\\nhome), overruled on other grounds by United States v. King,\\n687 F.3d 1189 (9th Cir. 2012) (en banc) (per curiam); see\\nalso Florida v. Jardines, 569 U.S. 1, 6 (2013) (“But when it\\ncomes to the Fourth Amendment, the home is first among\\n\\n10 UNITED STATES V. KORTE\\n\\nequals.”). The district court summed it up well: “If an\\nindividual’s residence falls within the scope of an\\nappropriate search under the parolee provisions, how would\\na trunk not?”\\n\\n Accordingly, the district court correctly held that the\\nsearch of Korte’s trunk was a lawful parole search.\\n\\nB. Placement and Use of the GPS Tracker\\n\\n We next decide whether Korte’s parole-search condition\\npermitted the warrantless placement of a GPS device on his\\ncar and the subsequent surveillance of his car’s movements.\\n\\n Installing a GPS tracker on a car constitutes a search,\\ntypically requiring a warrant. See United States v. Jones,\\n565 U.S. 400, 404 (2012). The application of this principle\\nto a parolee’s car, however, is less clear. While the parties\\nagree that placing the device on Korte’s car was a search\\nunder the Fourth Amendment, they disagree on whether\\ndoing so without a warrant offends the Constitution. In light\\nof Samson’s strong pronouncement that parolees in\\nCalifornia have very limited Fourth Amendment rights, 547\\nU.S. at 851–52, we agree with the district court that this was\\na lawful parole search.\\n\\n Our decision in United States v. Johnson, 875 F.3d 1265\\n(9th Cir. 2017), instructs us not to necessarily apply a newly\\nestablished Fourth Amendment protection to parolees. In\\nRiley v. California, 573 U.S. 373, 386 (2014), the United\\nStates Supreme Court held that the warrantless search of an\\narrestee’s cell phone was unlawful. The Court emphasized\\nthe significant privacy intrusion that arose when searching a\\nperson’s cell phone. Id. at 393; see also Johnson, 875 F.3d\\nat 1274 (noting that Riley was premised on “privacy interests\\nimplicated in cell phone searches [being] particularly\\n\\n UNITED STATES V. KORTE 11\\n\\nacute”). Cell phones, the Supreme Court said, are now “a\\npervasive and insistent part of daily life” that “place vast\\nquantities of personal information literally in the hands of\\nindividuals.” Riley, 573 U.S. at 385–86. Because cell\\nphones “collect[] in one place many distinct types of\\ninformation . . . that reveal much more in combination than\\nany isolated record,” searching a cell phone would give law\\nenforcement the unparalleled ability to reconstruct “[t]he\\nsum of an individual’s private life.” Id. at 394.\\n\\n Despite the Court’s cautionary words, we held that Riley\\ndid not apply to parolees. Johnson, 875 F.3d at 1275; but\\nsee United States v. Lara, 815 F.3d 605, 612 (9th Cir. 2016)\\n(applying Riley to the warrantless search of a probationer’s\\ncell phone). Rather, noting that “the balance of privacy\\ninterests and factual circumstances in this context are\\ndifferent,” we permitted the warrantless search of a parolee’s\\ncell phone. Johnson, 875 F.3d at 1273.\\n\\n In light of our ruling in Johnson, we are hard-put to say\\nthat the warrantless placement of a GPS tracker on a\\nparolee’s car is impermissible. If an officer can conduct a\\nwarrantless search of a parolee’s cell phone — an object that\\nis “[t]he sum of an individual’s private life,” Riley, 573 U.S.\\nat 394 — placing a GPS device on a parolee’s car cannot\\nlogically demand more constitutional protection. Although\\na GPS tracker may create a summary of a parolee’s public\\nmovements, it offers none of the “vast quantities of personal\\ninformation” that a cell phone does. Id. at 386.\\n\\n The State’s interest in supervising parolees is also\\nparticularly strong here. See Samson, 547 U.S. at 853\\n(referencing the State’s “substantial” interests in “reducing\\nrecidivism,” “promoting reintegration,” and deterring future\\ncriminal conduct). Tracking a parolee’s movements by car\\ncan be a critical tool for monitoring this group. Its value is\\n\\n12 UNITED STATES V. KORTE\\n\\nwell illustrated here: Korte returned to a life of crime just\\nmonths after his release from prison, but LASD was able to\\ninvestigate Korte and prevented other armed robberies by\\ntracking his movements. In Johnson, we explained that\\nrequiring officers to obtain a warrant before searching a\\nparolee’s cell phone “would often undermine the state’s\\nability to supervise effectively.” 875 F.3d at 1274. We have\\nsimilar concerns with requiring officers to obtain a warrant\\nbefore tracking a parolee’s vehicular movements.\\n\\n Lastly, we note that our decision aligns with another\\ncourt’s interpretation of California’s parole-search\\ncondition. In People v. Zichwic, 114 Cal. Rptr. 2d 733, 738–\\n39 (Ct. App. 2001), the California Court of Appeal\\nconsidered the warrantless placement of an electronic\\nmonitoring device on a parolee’s car. It held that even\\n“assum[ing] that attaching an electronic tracking device to\\nthe undercarriage of defendant’s truck constituted a search,\\nit was authorized by defendant’s parole search condition.”\\nId. at 740. Granted, Zichwic differs in that it considered the\\nuse of a beeper, id. at 738, rather than a GPS device —\\ninvestigatory methods treated differently under the Fourth\\nAmendment. Compare United States v. Karo, 468 U.S. 705,\\n713 (1984) (“[N]o Fourth Amendment interest . . . was\\ninfringed by the installation of the beeper.”), with Jones, 565\\nU.S. at 409 n.6 (noting that GPS tracking is a more intrusive\\nlaw enforcement practice than a beeper).\\n\\n Nonetheless, we believe that Zichwic is, at minimum,\\ninformative. Although it was decided almost two decades\\nbefore Jones, its holding was unaffected because it had\\nassumed that placing a tracking device on a car constituted a\\nsearch. Zichwic, 114 Cal. Rptr. 2d at 740. The Supreme\\nCourt of California has not since interpreted the parole-\\nsearch condition differently. And, while a beeper and GPS\\n\\n UNITED STATES V. KORTE 13\\n\\ndevice might differ in their tracking capabilities, the court in\\nZichwic at least concluded, as we do, that the State’s need\\nfor electronically monitoring a parolee’s movements\\noutweighs the privacy interests at issue. Id. at 739–40.\\n\\n We do not disregard the importance of Jones. We\\nacknowledge that “GPS monitoring generates a precise,\\ncomprehensive record of a person’s public movements.”\\nJones, 565 U.S. at 415 (Sotomayor, J., concurring).\\nHowever, in following precedent distinguishing Fourth\\nAmendment rights in the parolee context, we hold that the\\nwarrantless placement of a GPS tracker on Korte’s car does\\nnot violate the Fourth Amendment.\\n\\nC. Warrantless CSLI Acquisition\\n\\n Finally, we consider whether the district court should\\nhave excluded the CSLI evidence, acquired without a\\nwarrant, as a Fourth Amendment violation. The Supreme\\nCourt granted certiorari to resolve the constitutionality of\\nwarrantless CSLI acquisition before the district court could\\nrule on Korte’s suppression motion. Carpenter v. United\\nStates, 137 S. Ct. 2211 (2017) (mem.). As such, the district\\ncourt did not address the constitutional issue before it, but\\ndenied Korte’s motion under the Fourth Amendment’s good-\\nfaith exception. We, therefore, consider only a narrow issue:\\nwhether the good-faith exception applies to the warrantless\\nacquisition of a defendant’s CSLI before Carpenter v.\\nUnited States, 138 S. Ct. 2206 (2018). 2\\n\\n\\n\\n\\n 2\\n We take no stance on the constitutionality of acquiring a parolee’s\\nCSLI without a warrant.\\n\\n14 UNITED STATES V. KORTE\\n\\n In Carpenter, the Supreme Court summarized CSLI and\\nhow this technology permits almost real-time compilation of\\na person’s location during any given period:\\n\\n Cell phones continuously scan their\\n environment looking for the best signal,\\n which generally comes from the closest cell\\n site. Most modern devices . . . tap into the\\n wireless network several times a minute\\n whenever their signal is on, even if the owner\\n is not using one of the phone’s features. Each\\n time the phone connects to a cell site, it\\n generates a time-stamped record known as\\n cell-site location information (CSLI).\\n\\nId. at 2211. The Court wrestled with how to apply this “new\\nphenomenon” under the Fourth Amendment, id. at 2216,\\nnoting that it “does not fit neatly under existing precedents,”\\nid. at 2214. It ultimately concluded, however, that the\\nGovernment must obtain a warrant to access a person’s CSLI\\nfrom a wireless carrier, and could no longer rely on a court\\norder under § 2703(d) of the SCA. Id. at 2221.\\n\\n Although the Government obtained Korte’s CSLI\\nwithout a warrant, he is not automatically entitled to relief.\\nSee United States v. Leon, 468 U.S. 897, 906–07 (1984)\\n(explaining that whether there was a Fourth Amendment\\nviolation and “[w]hether the exclusionary sanction is\\nappropriately imposed” are separate questions). Illinois v.\\nKrull, 480 U.S. 340, 342, 350 (1987), established an\\nimportant exception to the exclusionary rule: Evidence\\nobtained by the Government, acting in “objectively\\nreasonable reliance upon a statute” that is “ultimately found\\nto violate the Fourth Amendment,” does not require\\nsuppression.\\n\\n UNITED STATES V. KORTE 15\\n\\n Because we find the Government reasonably relied on\\nthe SCA when it obtained Korte’s CSLI, we affirm the\\ndistrict court’s application of the Fourth Amendment’s\\ngood-faith exception. See also United States v. Camou, 773\\nF.3d 932, 944 (9th Cir. 2014) (placing the burden on the\\ngovernment to prove it acted in good faith). Before\\nCarpenter, the SCA authorized a government entity to\\nrequest “a provider of electronic communication service . . .\\nto disclose a record or other information pertaining to a\\nsubscriber to or a customer of such service.” 18 U.S.C.\\n§ 2703(c)(1). The statute explicitly authorized retrieval of\\nthese records by court order if the Government “offer[ed]\\nspecific and articulable facts showing that there are\\nreasonable grounds to believe that . . . the records or other\\ninformation sought, are relevant and material to an ongoing\\ncriminal investigation” — a more lenient standard than\\nprobable cause. Id. § 2703(d).\\n\\n When the Government obtained Korte’s CSLI — before\\nCarpenter was decided — acting by court order was still\\nauthorized. Moreover, we cannot say that the Government\\nhad any reason to doubt the SCA’s constitutionality, such\\nthat it may have been acting in bad faith. See Krull, 480 U.S.\\nat 355 (“[T]he standard of reasonableness . . . is an objective\\none.”). CSLI remained a relatively novel form of evidence.\\nAnd, although we had not yet commented on the\\nconstitutionality of warrantless CSLI acquisition, a number\\nof our sister circuits had. All had affirmed the SCA’s\\nconstitutionality under the Fourth Amendment. See United\\nStates v. Gilton, No. 16-10109, slip op. at 20 (9th Cir. Mar.\\n4, 2019) (explaining that, at least as of 2012, “the prevailing\\nbelief” was “that CSLI data was not protected by the Fourth\\nAmendment”); see also United States v. Graham, 824 F.3d\\n421, 424 (4th Cir. 2016) (en banc); United States v.\\nCarpenter, 819 F.3d 880, 884 (6th Cir. 2016); United States\\n\\n16 UNITED STATES V. KORTE\\n\\nv. Davis, 785 F.3d 498, 511 (11th Cir. 2015) (en banc); In re\\nApplication of the U.S. for Historical Cell Site Data, 724\\nF.3d 600, 615 (5th Cir. 2013); In re Application of the U.S.\\nfor an Order Directing a Provider of Elec. Commc’n Serv.\\nto Disclose Records to the Gov’t, 620 F.3d 304, 313 (3d Cir.\\n2010).\\n\\n As explained in Krull, it is hardly objectively\\nunreasonable to rely on a then-lawful statute when courts\\nwere upholding it or similar legislative schemes. 480 U.S.\\nat 358–59. Officials should not be “expected to question the\\njudgment of the legislature that passed the law,” particularly\\nwhen confronted with the pattern of judicial holdings as\\nexisted here. Id. at 350. Any “defect in the [SCA],”\\ntherefore, “was not sufficiently obvious so as to render [the\\nGovernment’s] reliance upon the statute objectively\\nunreasonable.” Id. at 359. The Supreme Court’s own\\nsharply divided opinion in Carpenter brings this point to\\nbear. See 138 S. Ct. at 2217 (explaining that the CSLI issue\\npresents “novel circumstances”); see also id. at 2223\\n(referring to the majority’s decision as a “stark departure\\nfrom relevant Fourth Amendment precedents”) (Kennedy,\\nJ., dissenting).\\n\\n Moreover, our application of Krull is anything but novel.\\nSeveral other circuits have already invoked this good-faith\\nexception when presented with similar facts. See United\\nStates v. Goldstein, 914 F.3d 200, 203–05 (3d Cir. 2019);\\nUnited States v. Curtis, 901 F.3d 846, 848–49 (7th Cir.\\n2018); see also United States v. Chambers, No. 16-163-cr,\\n2018 WL 4523607, at *2 (2d Cir. Sept. 21, 2018) (explaining\\nthat “the authorities sought information from third parties by\\ncomplying with the SCA” and “[r]eliance on a federal statute\\ngives rise to a presumption of good faith” (citing Krull, 480\\nU.S. at 349)). Others have analyzed the issue similarly,\\n\\n UNITED STATES V. KORTE 17\\n\\nalthough without directly invoking Krull. See United States\\nv. Joyner, 899 F.3d 1199, 1205 (11th Cir. 2018) (per curiam)\\n(noting that “the Government complied with the\\nrequirements of the SCA in obtaining the orders to compel\\ncell site records” while it was still lawful); United States v.\\nChavez, 894 F.3d 593, 608 (4th Cir. 2018) (refusing to apply\\nthe exclusionary rule because “investigators in this case\\nreasonably relied on court orders and the [SCA] in obtaining\\nthe cell site records”). Similarly, we recently refused to\\nexclude CSLI data obtained in good-faith reliance on a\\nwarrant, later deemed to be defective. Gilton, slip op. at 21.\\n\\n Finally, we note that our decision accords with the\\nexclusionary rule’s limited purpose to deter future Fourth\\nAmendment violations, rather than remedy the rights of a\\nsingle aggrieved party. See United States v. Calandra, 414\\nU.S. 338, 347–48 (1974); see also Krull, 480 U.S. at 347.\\nFor us to exclude CSLI, obtained in good faith based on a\\nthen-lawful legislative scheme, would do nothing to prevent\\nfuture Fourth Amendment violations. See Davis v. United\\nStates, 564 U.S. 229, 236–37 (2011) (“The [exclusionary]\\nrule’s sole purpose, we have repeatedly held, is to deter\\nfuture Fourth Amendment violations.”). With the\\nexclusionary rule as “our last resort, not our first impulse,”\\nHudson v. Michigan, 547 U.S. 586, 591 (2006), we hold that\\nCSLI acquired pre-Carpenter is admissible — so long as the\\nGovernment satisfied the SCA’s then-lawful requirements\\n— under Krull’s good-faith exception.\\n\\n18 UNITED STATES V. KORTE\\n\\nIII. CONCLUSION\\n\\n For the reasons provided above, we affirm the district\\ncourt’s denial of Korte’s suppression motions and his\\nconvictions for bank robbery. 3\\n\\n AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4380668/\", \"author_raw\": \"OWENS, Circuit Judge:\"}, {\"author\": \"D.W. NELSON, Circuit Judge, concurring\", \"type\": \"concurrence\", \"text\": \"D.W. NELSON, Circuit Judge, concurring:\\n\\n I concur in both the reasoning and result—the good-faith\\nexception saves the government’s conduct here. I write\\nseparately, however, because I am concerned with the ever\\n“diminishing” reasonable expectation of privacy afforded to\\nprobationers and parolees, especially as it relates to their\\ndigital privacy.\\n\\n In Samson v. California, the United States Supreme\\nCourt seemingly established a ceiling on a parolee’s\\nexpectation of privacy. 547 U.S. 843, 846 (2006). The\\nSupreme Court, however, did not set a floor. Nor did it offer\\na limiting principle. It is not surprising, therefore, that in the\\ndecade or so since Samson, our Court has further diminished\\n\\n 3\\n Korte also argues that his Count 1 conviction for attempted bank\\nrobbery under 18 U.S.C. § 2113(a) cannot stand because there was\\ninsufficient evidence of “intimidation.” Longstanding precedent\\nforecloses this argument. Korte’s demand for money and use of a mask\\nsatisfies “intimidation” under § 2113(a). See, e.g., United States v.\\nHopkins, 703 F.2d 1102, 1103 (9th Cir. 1983) (“[T]he threats implicit in\\n[defendant’s] written and verbal demands for money provide sufficient\\nevidence of intimidation to support the jury’s verdict.”); United States v.\\nBingham, 628 F.2d 548, 549 (9th Cir. 1980) (“[E]xpress threats of bodily\\nharm, threatening body motions, or the physical possibility of concealed\\nweapon[s] . . . have never been held to be requirements for a § 2113(a)\\nconviction.”).\\n\\n UNITED STATES V. KORTE 19\\n\\nprobationers’ and parolees’ expectations of privacy. See,\\ne.g., United States v. Johnson, 875 F.3d 1265, 1275 (9th Cir.\\n2017) (holding that Riley’s prohibition on warrantless phone\\nsearches does not apply to parolees); United States v.\\nCervantes, 859 F.3d 1175, 1183 (9th Cir. 2017) (holding that\\na search of a parolee that complies with the terms of a valid\\nsearch condition is usually reasonable); United States v.\\nGrandberry, 730 F.3d 968, 980 (9th Cir. 2013) (defining the\\ntypes of property under a parolee’s control that are\\nsearchable without a warrant); United States v. Lopez, 474\\nF.3d 1208, 1213–14 (9th Cir. 2007) (upholding the\\nwarrantless search of a parolee’s home).\\n\\n Justice Stevens, in his Samson dissent, stated that the\\nSupreme Court’s precedents did not support “a regime of\\nsuspicionless searches. . . untethered by any procedural\\nsafeguards, by law enforcement personnel who have no\\nspecial interest in the welfare of the parolee or probationer.”\\nSamson, 547 U.S. at 857 (Stevens, J., dissenting). He\\nsuggested an individualized suspicion requirement to\\nprevent what he warned would be an “unprecedented\\ncurtailment of liberty.” Id. at 866. His words proved to be a\\nprescient warning.\\n\\n Our physical and digital spaces continue to merge with\\nthe creation of new technologies. The individual and societal\\nbenefits of these technologies come with increased risks—\\nour devices and third parties store ever more deeply\\nrevealing and private information that is easily accessed by\\nlaw enforcement. The Supreme Court recently reaffirmed\\nthat the “progress of science” should not come at the expense\\nof Fourth Amendment protections. See Carpenter v. United\\nStates, 138 S. Ct. 2206, 2223 (2018) (quoting Olmstead v.\\nUnited States, 277 U.S. 438, 474 (1928)). I believe,\\n\\n20 UNITED STATES V. KORTE\\n\\ntherefore, it is time to revisit the degree of protection the\\nFourth Amendment affords probationers and parolees.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4380668/\", \"author_raw\": \"D.W. NELSON, Circuit Judge, concurring\"}]}"},"judge1":{"kind":"string","value":"DOROTHY W NELSON"},"judge2":{"kind":"string","value":"CONSUELO M CALLAHAN"},"judge3":{"kind":"string","value":"JOHN B OWENS"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":2,"string":"2"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":1,"string":"1"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4603415/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. Kyle Korte\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-15\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: Yes\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 18-50051\n Plaintiff-Appellee,\n D.C. No.\n v. 8:16-cr-00156-\n JLS-1\nKYLE JASON KORTE,\n Defendant-Appellant. OPINION\n\n\n\n Appeal from the United States District Court\n for the Central District of California\n Josephine L. Staton, District Judge, Presiding\n\n Argued and Submitted February 11, 2019\n Pasadena, California\n\n Filed March 15, 2019\n\n Before: Dorothy W. Nelson, Consuelo M. Callahan, and\n John B. Owens, Circuit Judges.\n\n Opinion by Judge Owens;\n Concurrence by Judge D.W. Nelson\n\n2 UNITED STATES V. KORTE\n\n SUMMARY *\n\n\n Criminal Law\n\n The panel affirmed the district court’s denial of the\ndefendant’s suppression motions and his convictions for\nbank robbery in a case in which the defendant, who was on\nparole during his crime spree, had consented to suspicionless\nsearches of his person, residence, and any property under his\ncontrol.\n\n Rejecting the defendant’s contention that a lawful parole\nsearch of his car does not extend to the trunk because the\ntrunk is not “property under his control,” the panel held that\nthe parole search of the defendant’s trunk was lawful.\n\n Following precedent distinguishing Fourth Amendment\nrights in the parolee context, the panel held that the\nwarrantless placement of a GPS tracker on the defendant’s\ncar did not violate the Fourth Amendment.\n\n The panel held that cell site location information (CSLI)\nacquired before the Supreme Court issued its decision in\nCarpenter v. United States, 138 S. Ct. 2206 (2018) (holding\nthat the Government must obtain a warrant to access a\nperson’s CSLI from a wireless carrier), is admissible under\nthe good-faith exception to the exclusionary rule so long as\nthe Government satisfied the Stored Communications Act’s\nthen-lawful requirements. The panel took no stance on the\nconstitutionality of acquiring a parolee’s CSLI without a\nwarrant.\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\n UNITED STATES V. KORTE 3\n\n Concurring in both the reasoning and the result, Judge\nNelson wrote separately because she is concerned with the\never “diminishing” reasonable expectation of privacy\nafforded to probationers and parolees, especially as it relates\nto digital privacy.\n\n\n COUNSEL\n\nH. Dean Steward (argued), Attorney at Law, San Clemente,\nCalifornia, for Defendant-Appellant.\n\nJulia L. Reese (argued), Assistant United States Attorney,\nCriminal Appeals Section; Nicola T. Hanna, United States\nAttorney; Lawrence S. Middleton, Assistant United States\nAttorney, Chief, Criminal Division; Gregory W. Staples,\nAssistant United States Attorney, Santa Ana Branch Office;\nUnited States Attorney’s Office, Los Angeles, California;\nfor Plaintiff-Appellee.\n\n\n OPINION\n\nOWENS, Circuit Judge:\n\n Defendant Kyle Korte appeals from his convictions for\nbank robbery. During his crime spree, Korte was on parole\nand had consented to warrantless, suspicionless searches of\nhis person, residence, and any property under his control.\nRelying on this parole condition, officers placed a Global\nPositioning System (“GPS”) device on his car and later\nsearched its trunk. Officers also obtained Korte’s historical\ncell site location information (“CSLI”) by court order. On\nappeal, Korte primarily challenges the district court’s denial\n\n4 UNITED STATES V. KORTE\n\nof his suppression motions as to each of these searches. We\nhave jurisdiction under 28 U.S.C. § 1291, and we affirm.\n\nI. FACTUAL AND PROCEDURAL BACKGROUND\n\n In a world of cybercrime and identity theft, Korte stole\nmoney the old-fashioned way — he robbed banks. After\nserving time in state prison for bank robbery, Korte was\nparoled in August 2016. As a parolee in California, Korte\nwas “subject to search or seizure . . . at any time of the day\nor night, with or without a search warrant or with or without\ncause.” CAL. PENAL CODE § 3067(b)(3); see also id.\n§ 3067(a). On October 25, 2016, Korte acknowledged his\nparole conditions, including that he was now subject to\nsearches of “[y]ou, your residence, and any property under\nyour control.”\n\n In October 2016, the Los Angeles Sheriff’s Department\n(“LASD”) began investigating a series of bank robberies.\nThe first robbery took place on October 7. A masked robber\nentered a bank and demanded “all your hundreds.” The\nfrightened teller, protected by bulletproof glass, activated the\nsilent alarm and retreated to a back office. The robber left\nwith no money. On October 12, the masked robber targeted\nanother bank, this time brandishing a toy gun. He was more\nsuccessful this go-around, escaping with $1,600. He then hit\ntwo more banks on October 27. Again displaying the toy\ngun, the robber pocketed $2,200 and $7,000. In total, the\nmasked robber stole less than $11,000 — not a Neil\nMcCauley heist by any means.\n\n Working with LASD, the Federal Bureau of\nInvestigation (“FBI”) began to suspect that Korte was the\nmasked robber. Surveillance video from one of the robberies\nshowed a car registered to the address that Korte provided to\nhis parole officer. An LASD officer who saw video of the\n\n UNITED STATES V. KORTE 5\n\nmasked robber also reported that the individual resembled\nKorte, who the officer knew was on parole for bank robbery.\n\n On November 4, 2016, without a warrant or Korte’s\nconsent, LASD placed a GPS tracking device on Korte’s car\nand periodically monitored the vehicle’s movements over\nthe next six days. That same day, the Government obtained\na court order under the Stored Communications Act\n(“SCA”), 18 U.S.C. § 2703(d), to acquire Korte’s CSLI.\nThis information placed Korte’s cell phone near three of the\nfour banks at the time of the respective robberies.\n\n On November 10, 2016, LASD learned that the FBI had\nobtained an arrest warrant for Korte. Officers followed\nKorte as he drove from his home to a bank and parked\nnearby, seeming to surveil his next target. An officer saw\nKorte open the car’s trunk and place something inside.\nShortly thereafter, they arrested Korte and searched the car.\nThe officers’ search of the trunk revealed the toy gun used\nduring the three armed robberies and the shirt Korte had been\nwearing while casing the bank just prior to his arrest.\n\n A grand jury indicted Korte for one count of attempted\nbank robbery in violation of 18 U.S.C. § 2113(a) and three\ncounts of bank robbery in violation of 18 U.S.C. § 2113(a),\n(d). Korte pled not guilty and moved to suppress (1) the\nevidence found in his car’s trunk, (2) the information derived\nfrom the GPS tracker on his car, and (3) his CSLI. The\ndistrict court denied all three suppression motions. The court\nreasoned that Korte’s parole status permitted the warrantless\nsearch of the trunk and placement of the GPS tracker on his\ncar. As for the CSLI, it held that even if the acquisition of\nthis information violated Korte’s Fourth Amendment rights,\nsee Carpenter v. United States, 138 S. Ct. 2206 (2018), the\ngood-faith exception clearly applied based on the case law\nat that time.\n\n6 UNITED STATES V. KORTE\n\n Korte went to trial. As to Count 1 — the unsuccessful\nrobbery — the bank teller testified that even though\nbulletproof glass separated her from the robber and she saw\nno weapon, she was nevertheless “[k]ind of panicked,”\n“[s]hocked,” and “[s]cared.” Korte filed a motion for a\njudgment of acquittal, under Federal Rule of Criminal\nProcedure 29, focusing on Count 1. The district court\nrejected his argument that the Government failed to prove\nthe element of “force and violence, or by intimidation,” as\n18 U.S.C. § 2113(a) requires. Citing the teller’s “testimony\nwith regard to a mask and the demand,” the court found\nsufficient evidence of “intimidation” to send the matter to\nthe jury.\n\n The jury returned a guilty verdict as to all four counts.\nKorte was sentenced to 210 months in prison.\n\nII. DISCUSSION\n\n We review de novo the denial of Korte’s suppression\nmotions. See United States v. Zapien, 861 F.3d 971, 974 (9th\nCir. 2017) (per curiam). We address each search — of his\ncar’s trunk, the GPS tracker on his car, and the acquisition\nof his CSLI — in turn.\n\nA. Search of the Trunk\n\n We first examine whether Korte’s parole-search\ncondition authorized the warrantless search of his car’s\ntrunk. See United States v. Cervantes, 859 F.3d 1175, 1183\n(9th Cir. 2017) (“A search of a parolee that complies with\nthe terms of a valid search condition will usually be deemed\nreasonable under the Fourth Amendment.”).\n\n California Penal Code section 3067(b)(3) provides that\nevery parolee “is subject to search or seizure . . . at any time\n\n UNITED STATES V. KORTE 7\n\nof the day or night, with or without a search warrant or with\nor without cause.” In Samson v. California, 547 U.S. 843,\n846 (2006), the United States Supreme Court reviewed\nCalifornia’s parole-search condition to determine “whether\na suspicionless search, conducted under the authority of\n[section 3067], violates the Constitution.” The Court first\nreasoned that “parole is more akin to imprisonment.” Id. at\n850. Because parolees “have severely diminished\nexpectations of privacy by virtue of their status alone” —\neven less than probationers — and because “[t]he State’s\ninterests” in supervising parolees and reducing recidivism\n“are substantial,” the Court upheld California’s parole-\nsearch condition. 1 Id. at 852–53.\n\n Korte does not — and cannot — argue that officers\nunlawfully searched his car. He admits that he rented the car\nand referred to it as “my car.” As “property under [his]\ncontrol,” California’s parole-search condition authorized the\n 1\n There are two limitations on this condition. First, the officer\nconducting the search must know at the time that the individual is\ncurrently on parole. See Moreno v. Baca, 431 F.3d 633, 641 (9th Cir.\n2005). Second, the search cannot be “arbitrary, capricious, or\nharassing.” See CAL. PENAL CODE § 3067(d) (“It is not the intent of the\nLegislature to authorize law enforcement officers to conduct searches for\nthe sole purpose of harassment.”); see also Samson, 547 U.S. at 856\n(“The concern that California’s suspicionless search system gives\nofficers unbridled discretion to conduct searches . . . is belied by\nCalifornia’s prohibition on ‘arbitrary, capricious or harassing’ searches.”\n(citations omitted)).\n\n Here, it is undisputed that the officers knew that Korte was on\nparole, and there is no evidence the officers engaged in an arbitrary,\ncapricious, or harassing search. Nor does Korte contend he was unaware\nof the parole-search condition. See People v. Lilienthal, 587 P.2d 706,\n711 (Cal. 1978) (“The condition itself provides general notice to a\ndefendant that he or his belongings may be subjected to warrantless\nsearches.”).\n\n8 UNITED STATES V. KORTE\n\nwarrantless search of it. See, e.g., United States v. Caseres,\n533 F.3d 1064, 1075–76 (9th Cir. 2008) (explaining that a\nsearch of the parolee’s car would have been lawful if the\nofficer had known that the defendant was on parole).\n\n Yet, Korte contends that a lawful parole search of his car\ndoes not extend to the trunk because the trunk is not similarly\n“property under [his] control.” We reject his narrow\ninterpretation of this condition. Property is subject to search\nwhen a parolee “exhibit[s] a sufficiently strong connection\nto [the property in question] to demonstrate ‘control’ over\nit.” United States v. Grandberry, 730 F.3d 968, 980 (9th Cir.\n2013) (defining “property under [a parolee’s] control”); see\nalso Cervantes, 859 F.3d at 1183 (explaining that having a\nkey to and belongings inside a hotel room sufficiently\ndemonstrate control, even if the room is co-occupied).\nConsistent with our understanding of the condition, the\nSupreme Court of California explained that a parolee\ncontrols property based on “the nexus between the parolee\nand the area or items searched,” including the “nature of that\narea or item” and “how close and accessible the area or item\nis to the parolee.” People v. Schmitz, 288 P.3d 1259, 1270\n(Cal. 2012) (holding that a parolee, who is only a passenger\nin a third-party’s vehicle, is in control of areas within his\nreach in the passenger compartment).\n\n Korte’s uncontested control over the car was sufficient\nto permit a warrantless search of its trunk. In any event, his\nconduct also illustrated a sufficiently close nexus to the trunk\nitself. Officers observed him putting things inside the trunk.\n\n Permitting a warrantless search of the trunk of a\nparolee’s car is also consistent with broader Fourth\nAmendment precedent. Generally, a lawful search of a fixed\nspace or premise extends to its entire area, whether or not\nthat requires opening a confined space: “[N]ice distinctions\n\n UNITED STATES V. KORTE 9\n\nbetween closets, drawers, and containers, in the case of a\nhome, or between glove compartments, upholstered seats,\ntrunks, and wrapped packages, in the case of a vehicle, must\ngive way to the interest in the prompt and efficient\ncompletion of the task at hand.” United States v. Ross, 456\nU.S. 798, 821 (1982). This reasoning applies to a car’s\ntrunk. See, e.g., id. at 825 (“If probable cause justifies the\nsearch of a lawfully stopped vehicle, it justifies the search of\nevery part of the vehicle and its contents that may conceal\nthe object of the search.”); United States v. Ewing, 638 F.3d\n1226, 1231 (9th Cir. 2011) (“[I]f a law enforcement officer\nhas probable cause to search a vehicle, that probable cause\nextends to all contents in the vehicle that could be connected\nto the suspected activity.”); United States v. McWeeney, 454\nF.3d 1030, 1035 (9th Cir. 2006) (holding that consent to\nsearch a defendant’s car extends to the trunk). We also note\nthat Korte’s proposed distinction between a car and trunk\nwas already rejected by the Supreme Court of California in\nthe context of a probationer – who has more Fourth\nAmendment protection than Korte as a parolee. See\nLilienthal, 587 P.2d at 711 (“We conclude that the officers\nwere justified in searching defendant’s car trunk pursuant to\ndefendant’s consent to warrantless searches as a condition of\nhis probation.”).\n\n As property under his control, Korte fails to explain how\nsearching his car’s trunk would offend the Fourth\nAmendment when a warrantless search of his home — the\napex of constitutionally protected places — would not. See\nUnited States v. Lopez, 474 F.3d 1208, 1213–14 (9th Cir.\n2007) (upholding the warrantless search of a parolee’s\nhome), overruled on other grounds by United States v. King,\n687 F.3d 1189 (9th Cir. 2012) (en banc) (per curiam); see\nalso Florida v. Jardines, 569 U.S. 1, 6 (2013) (“But when it\ncomes to the Fourth Amendment, the home is first among\n\n10 UNITED STATES V. KORTE\n\nequals.”). The district court summed it up well: “If an\nindividual’s residence falls within the scope of an\nappropriate search under the parolee provisions, how would\na trunk not?”\n\n Accordingly, the district court correctly held that the\nsearch of Korte’s trunk was a lawful parole search.\n\nB. Placement and Use of the GPS Tracker\n\n We next decide whether Korte’s parole-search condition\npermitted the warrantless placement of a GPS device on his\ncar and the subsequent surveillance of his car’s movements.\n\n Installing a GPS tracker on a car constitutes a search,\ntypically requiring a warrant. See United States v. Jones,\n565 U.S. 400, 404 (2012). The application of this principle\nto a parolee’s car, however, is less clear. While the parties\nagree that placing the device on Korte’s car was a search\nunder the Fourth Amendment, they disagree on whether\ndoing so without a warrant offends the Constitution. In light\nof Samson’s strong pronouncement that parolees in\nCalifornia have very limited Fourth Amendment rights, 547\nU.S. at 851–52, we agree with the district court that this was\na lawful parole search.\n\n Our decision in United States v. Johnson, 875 F.3d 1265\n(9th Cir. 2017), instructs us not to necessarily apply a newly\nestablished Fourth Amendment protection to parolees. In\nRiley v. California, 573 U.S. 373, 386 (2014), the United\nStates Supreme Court held that the warrantless search of an\narrestee’s cell phone was unlawful. The Court emphasized\nthe significant privacy intrusion that arose when searching a\nperson’s cell phone. Id. at 393; see also Johnson, 875 F.3d\nat 1274 (noting that Riley was premised on “privacy interests\nimplicated in cell phone searches [being] particularly\n\n UNITED STATES V. KORTE 11\n\nacute”). Cell phones, the Supreme Court said, are now “a\npervasive and insistent part of daily life” that “place vast\nquantities of personal information literally in the hands of\nindividuals.” Riley, 573 U.S. at 385–86. Because cell\nphones “collect[] in one place many distinct types of\ninformation . . . that reveal much more in combination than\nany isolated record,” searching a cell phone would give law\nenforcement the unparalleled ability to reconstruct “[t]he\nsum of an individual’s private life.” Id. at 394.\n\n Despite the Court’s cautionary words, we held that Riley\ndid not apply to parolees. Johnson, 875 F.3d at 1275; but\nsee United States v. Lara, 815 F.3d 605, 612 (9th Cir. 2016)\n(applying Riley to the warrantless search of a probationer’s\ncell phone). Rather, noting that “the balance of privacy\ninterests and factual circumstances in this context are\ndifferent,” we permitted the warrantless search of a parolee’s\ncell phone. Johnson, 875 F.3d at 1273.\n\n In light of our ruling in Johnson, we are hard-put to say\nthat the warrantless placement of a GPS tracker on a\nparolee’s car is impermissible. If an officer can conduct a\nwarrantless search of a parolee’s cell phone — an object that\nis “[t]he sum of an individual’s private life,” Riley, 573 U.S.\nat 394 — placing a GPS device on a parolee’s car cannot\nlogically demand more constitutional protection. Although\na GPS tracker may create a summary of a parolee’s public\nmovements, it offers none of the “vast quantities of personal\ninformation” that a cell phone does. Id. at 386.\n\n The State’s interest in supervising parolees is also\nparticularly strong here. See Samson, 547 U.S. at 853\n(referencing the State’s “substantial” interests in “reducing\nrecidivism,” “promoting reintegration,” and deterring future\ncriminal conduct). Tracking a parolee’s movements by car\ncan be a critical tool for monitoring this group. Its value is\n\n12 UNITED STATES V. KORTE\n\nwell illustrated here: Korte returned to a life of crime just\nmonths after his release from prison, but LASD was able to\ninvestigate Korte and prevented other armed robberies by\ntracking his movements. In Johnson, we explained that\nrequiring officers to obtain a warrant before searching a\nparolee’s cell phone “would often undermine the state’s\nability to supervise effectively.” 875 F.3d at 1274. We have\nsimilar concerns with requiring officers to obtain a warrant\nbefore tracking a parolee’s vehicular movements.\n\n Lastly, we note that our decision aligns with another\ncourt’s interpretation of California’s parole-search\ncondition. In People v. Zichwic, 114 Cal. Rptr. 2d 733, 738–\n39 (Ct. App. 2001), the California Court of Appeal\nconsidered the warrantless placement of an electronic\nmonitoring device on a parolee’s car. It held that even\n“assum[ing] that attaching an electronic tracking device to\nthe undercarriage of defendant’s truck constituted a search,\nit was authorized by defendant’s parole search condition.”\nId. at 740. Granted, Zichwic differs in that it considered the\nuse of a beeper, id. at 738, rather than a GPS device —\ninvestigatory methods treated differently under the Fourth\nAmendment. Compare United States v. Karo, 468 U.S. 705,\n713 (1984) (“[N]o Fourth Amendment interest . . . was\ninfringed by the installation of the beeper.”), with Jones, 565\nU.S. at 409 n.6 (noting that GPS tracking is a more intrusive\nlaw enforcement practice than a beeper).\n\n Nonetheless, we believe that Zichwic is, at minimum,\ninformative. Although it was decided almost two decades\nbefore Jones, its holding was unaffected because it had\nassumed that placing a tracking device on a car constituted a\nsearch. Zichwic, 114 Cal. Rptr. 2d at 740. The Supreme\nCourt of California has not since interpreted the parole-\nsearch condition differently. And, while a beeper and GPS\n\n UNITED STATES V. KORTE 13\n\ndevice might differ in their tracking capabilities, the court in\nZichwic at least concluded, as we do, that the State’s need\nfor electronically monitoring a parolee’s movements\noutweighs the privacy interests at issue. Id. at 739–40.\n\n We do not disregard the importance of Jones. We\nacknowledge that “GPS monitoring generates a precise,\ncomprehensive record of a person’s public movements.”\nJones, 565 U.S. at 415 (Sotomayor, J., concurring).\nHowever, in following precedent distinguishing Fourth\nAmendment rights in the parolee context, we hold that the\nwarrantless placement of a GPS tracker on Korte’s car does\nnot violate the Fourth Amendment.\n\nC. Warrantless CSLI Acquisition\n\n Finally, we consider whether the district court should\nhave excluded the CSLI evidence, acquired without a\nwarrant, as a Fourth Amendment violation. The Supreme\nCourt granted certiorari to resolve the constitutionality of\nwarrantless CSLI acquisition before the district court could\nrule on Korte’s suppression motion. Carpenter v. United\nStates, 137 S. Ct. 2211 (2017) (mem.). As such, the district\ncourt did not address the constitutional issue before it, but\ndenied Korte’s motion under the Fourth Amendment’s good-\nfaith exception. We, therefore, consider only a narrow issue:\nwhether the good-faith exception applies to the warrantless\nacquisition of a defendant’s CSLI before Carpenter v.\nUnited States, 138 S. Ct. 2206 (2018). 2\n\n\n\n\n 2\n We take no stance on the constitutionality of acquiring a parolee’s\nCSLI without a warrant.\n\n14 UNITED STATES V. KORTE\n\n In Carpenter, the Supreme Court summarized CSLI and\nhow this technology permits almost real-time compilation of\na person’s location during any given period:\n\n Cell phones continuously scan their\n environment looking for the best signal,\n which generally comes from the closest cell\n site. Most modern devices . . . tap into the\n wireless network several times a minute\n whenever their signal is on, even if the owner\n is not using one of the phone’s features. Each\n time the phone connects to a cell site, it\n generates a time-stamped record known as\n cell-site location information (CSLI).\n\nId. at 2211. The Court wrestled with how to apply this “new\nphenomenon” under the Fourth Amendment, id. at 2216,\nnoting that it “does not fit neatly under existing precedents,”\nid. at 2214. It ultimately concluded, however, that the\nGovernment must obtain a warrant to access a person’s CSLI\nfrom a wireless carrier, and could no longer rely on a court\norder under § 2703(d) of the SCA. Id. at 2221.\n\n Although the Government obtained Korte’s CSLI\nwithout a warrant, he is not automatically entitled to relief.\nSee United States v. Leon, 468 U.S. 897, 906–07 (1984)\n(explaining that whether there was a Fourth Amendment\nviolation and “[w]hether the exclusionary sanction is\nappropriately imposed” are separate questions). Illinois v.\nKrull, 480 U.S. 340, 342, 350 (1987), established an\nimportant exception to the exclusionary rule: Evidence\nobtained by the Government, acting in “objectively\nreasonable reliance upon a statute” that is “ultimately found\nto violate the Fourth Amendment,” does not require\nsuppression.\n\n UNITED STATES V. KORTE 15\n\n Because we find the Government reasonably relied on\nthe SCA when it obtained Korte’s CSLI, we affirm the\ndistrict court’s application of the Fourth Amendment’s\ngood-faith exception. See also United States v. Camou, 773\nF.3d 932, 944 (9th Cir. 2014) (placing the burden on the\ngovernment to prove it acted in good faith). Before\nCarpenter, the SCA authorized a government entity to\nrequest “a provider of electronic communication service . . .\nto disclose a record or other information pertaining to a\nsubscriber to or a customer of such service.” 18 U.S.C.\n§ 2703(c)(1). The statute explicitly authorized retrieval of\nthese records by court order if the Government “offer[ed]\nspecific and articulable facts showing that there are\nreasonable grounds to believe that . . . the records or other\ninformation sought, are relevant and material to an ongoing\ncriminal investigation” — a more lenient standard than\nprobable cause. Id. § 2703(d).\n\n When the Government obtained Korte’s CSLI — before\nCarpenter was decided — acting by court order was still\nauthorized. Moreover, we cannot say that the Government\nhad any reason to doubt the SCA’s constitutionality, such\nthat it may have been acting in bad faith. See Krull, 480 U.S.\nat 355 (“[T]he standard of reasonableness . . . is an objective\none.”). CSLI remained a relatively novel form of evidence.\nAnd, although we had not yet commented on the\nconstitutionality of warrantless CSLI acquisition, a number\nof our sister circuits had. All had affirmed the SCA’s\nconstitutionality under the Fourth Amendment. See United\nStates v. Gilton, No. 16-10109, slip op. at 20 (9th Cir. Mar.\n4, 2019) (explaining that, at least as of 2012, “the prevailing\nbelief” was “that CSLI data was not protected by the Fourth\nAmendment”); see also United States v. Graham, 824 F.3d\n421, 424 (4th Cir. 2016) (en banc); United States v.\nCarpenter, 819 F.3d 880, 884 (6th Cir. 2016); United States\n\n16 UNITED STATES V. KORTE\n\nv. Davis, 785 F.3d 498, 511 (11th Cir. 2015) (en banc); In re\nApplication of the U.S. for Historical Cell Site Data, 724\nF.3d 600, 615 (5th Cir. 2013); In re Application of the U.S.\nfor an Order Directing a Provider of Elec. Commc’n Serv.\nto Disclose Records to the Gov’t, 620 F.3d 304, 313 (3d Cir.\n2010).\n\n As explained in Krull, it is hardly objectively\nunreasonable to rely on a then-lawful statute when courts\nwere upholding it or similar legislative schemes. 480 U.S.\nat 358–59. Officials should not be “expected to question the\njudgment of the legislature that passed the law,” particularly\nwhen confronted with the pattern of judicial holdings as\nexisted here. Id. at 350. Any “defect in the [SCA],”\ntherefore, “was not sufficiently obvious so as to render [the\nGovernment’s] reliance upon the statute objectively\nunreasonable.” Id. at 359. The Supreme Court’s own\nsharply divided opinion in Carpenter brings this point to\nbear. See 138 S. Ct. at 2217 (explaining that the CSLI issue\npresents “novel circumstances”); see also id. at 2223\n(referring to the majority’s decision as a “stark departure\nfrom relevant Fourth Amendment precedents”) (Kennedy,\nJ., dissenting).\n\n Moreover, our application of Krull is anything but novel.\nSeveral other circuits have already invoked this good-faith\nexception when presented with similar facts. See United\nStates v. Goldstein, 914 F.3d 200, 203–05 (3d Cir. 2019);\nUnited States v. Curtis, 901 F.3d 846, 848–49 (7th Cir.\n2018); see also United States v. Chambers, No. 16-163-cr,\n2018 WL 4523607, at *2 (2d Cir. Sept. 21, 2018) (explaining\nthat “the authorities sought information from third parties by\ncomplying with the SCA” and “[r]eliance on a federal statute\ngives rise to a presumption of good faith” (citing Krull, 480\nU.S. at 349)). Others have analyzed the issue similarly,\n\n UNITED STATES V. KORTE 17\n\nalthough without directly invoking Krull. See United States\nv. Joyner, 899 F.3d 1199, 1205 (11th Cir. 2018) (per curiam)\n(noting that “the Government complied with the\nrequirements of the SCA in obtaining the orders to compel\ncell site records” while it was still lawful); United States v.\nChavez, 894 F.3d 593, 608 (4th Cir. 2018) (refusing to apply\nthe exclusionary rule because “investigators in this case\nreasonably relied on court orders and the [SCA] in obtaining\nthe cell site records”). Similarly, we recently refused to\nexclude CSLI data obtained in good-faith reliance on a\nwarrant, later deemed to be defective. Gilton, slip op. at 21.\n\n Finally, we note that our decision accords with the\nexclusionary rule’s limited purpose to deter future Fourth\nAmendment violations, rather than remedy the rights of a\nsingle aggrieved party. See United States v. Calandra, 414\nU.S. 338, 347–48 (1974); see also Krull, 480 U.S. at 347.\nFor us to exclude CSLI, obtained in good faith based on a\nthen-lawful legislative scheme, would do nothing to prevent\nfuture Fourth Amendment violations. See Davis v. United\nStates, 564 U.S. 229, 236–37 (2011) (“The [exclusionary]\nrule’s sole purpose, we have repeatedly held, is to deter\nfuture Fourth Amendment violations.”). With the\nexclusionary rule as “our last resort, not our first impulse,”\nHudson v. Michigan, 547 U.S. 586, 591 (2006), we hold that\nCSLI acquired pre-Carpenter is admissible — so long as the\nGovernment satisfied the SCA’s then-lawful requirements\n— under Krull’s good-faith exception.\n\n18 UNITED STATES V. KORTE\n\nIII. CONCLUSION\n\n For the reasons provided above, we affirm the district\ncourt’s denial of Korte’s suppression motions and his\nconvictions for bank robbery. 3\n\n AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. Kyle Korte\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-15\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: Yes\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 18-50051\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 8:16-cr-00156-\\n JLS-1\\nKYLE JASON KORTE,\\n Defendant-Appellant. OPINION\\n\\n\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Josephine L. Staton, District Judge, Presiding\\n\\n Argued and Submitted February 11, 2019\\n Pasadena, California\\n\\n Filed March 15, 2019\\n\\n Before: Dorothy W. Nelson, Consuelo M. Callahan, and\\n John B. Owens, Circuit Judges.\\n\\n Opinion by Judge Owens;\\n Concurrence by Judge D.W. Nelson\\n\\n2 UNITED STATES V. KORTE\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n The panel affirmed the district court’s denial of the\\ndefendant’s suppression motions and his convictions for\\nbank robbery in a case in which the defendant, who was on\\nparole during his crime spree, had consented to suspicionless\\nsearches of his person, residence, and any property under his\\ncontrol.\\n\\n Rejecting the defendant’s contention that a lawful parole\\nsearch of his car does not extend to the trunk because the\\ntrunk is not “property under his control,” the panel held that\\nthe parole search of the defendant’s trunk was lawful.\\n\\n Following precedent distinguishing Fourth Amendment\\nrights in the parolee context, the panel held that the\\nwarrantless placement of a GPS tracker on the defendant’s\\ncar did not violate the Fourth Amendment.\\n\\n The panel held that cell site location information (CSLI)\\nacquired before the Supreme Court issued its decision in\\nCarpenter v. United States, 138 S. Ct. 2206 (2018) (holding\\nthat the Government must obtain a warrant to access a\\nperson’s CSLI from a wireless carrier), is admissible under\\nthe good-faith exception to the exclusionary rule so long as\\nthe Government satisfied the Stored Communications Act’s\\nthen-lawful requirements. The panel took no stance on the\\nconstitutionality of acquiring a parolee’s CSLI without a\\nwarrant.\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n UNITED STATES V. KORTE 3\\n\\n Concurring in both the reasoning and the result, Judge\\nNelson wrote separately because she is concerned with the\\never “diminishing” reasonable expectation of privacy\\nafforded to probationers and parolees, especially as it relates\\nto digital privacy.\\n\\n\\n COUNSEL\\n\\nH. Dean Steward (argued), Attorney at Law, San Clemente,\\nCalifornia, for Defendant-Appellant.\\n\\nJulia L. Reese (argued), Assistant United States Attorney,\\nCriminal Appeals Section; Nicola T. Hanna, United States\\nAttorney; Lawrence S. Middleton, Assistant United States\\nAttorney, Chief, Criminal Division; Gregory W. Staples,\\nAssistant United States Attorney, Santa Ana Branch Office;\\nUnited States Attorney’s Office, Los Angeles, California;\\nfor Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nOWENS, Circuit Judge:\\n\\n Defendant Kyle Korte appeals from his convictions for\\nbank robbery. During his crime spree, Korte was on parole\\nand had consented to warrantless, suspicionless searches of\\nhis person, residence, and any property under his control.\\nRelying on this parole condition, officers placed a Global\\nPositioning System (“GPS”) device on his car and later\\nsearched its trunk. Officers also obtained Korte’s historical\\ncell site location information (“CSLI”) by court order. On\\nappeal, Korte primarily challenges the district court’s denial\\n\\n4 UNITED STATES V. KORTE\\n\\nof his suppression motions as to each of these searches. We\\nhave jurisdiction under 28 U.S.C. § 1291, and we affirm.\\n\\nI. FACTUAL AND PROCEDURAL BACKGROUND\\n\\n In a world of cybercrime and identity theft, Korte stole\\nmoney the old-fashioned way — he robbed banks. After\\nserving time in state prison for bank robbery, Korte was\\nparoled in August 2016. As a parolee in California, Korte\\nwas “subject to search or seizure . . . at any time of the day\\nor night, with or without a search warrant or with or without\\ncause.” CAL. PENAL CODE § 3067(b)(3); see also id.\\n§ 3067(a). On October 25, 2016, Korte acknowledged his\\nparole conditions, including that he was now subject to\\nsearches of “[y]ou, your residence, and any property under\\nyour control.”\\n\\n In October 2016, the Los Angeles Sheriff’s Department\\n(“LASD”) began investigating a series of bank robberies.\\nThe first robbery took place on October 7. A masked robber\\nentered a bank and demanded “all your hundreds.” The\\nfrightened teller, protected by bulletproof glass, activated the\\nsilent alarm and retreated to a back office. The robber left\\nwith no money. On October 12, the masked robber targeted\\nanother bank, this time brandishing a toy gun. He was more\\nsuccessful this go-around, escaping with $1,600. He then hit\\ntwo more banks on October 27. Again displaying the toy\\ngun, the robber pocketed $2,200 and $7,000. In total, the\\nmasked robber stole less than $11,000 — not a Neil\\nMcCauley heist by any means.\\n\\n Working with LASD, the Federal Bureau of\\nInvestigation (“FBI”) began to suspect that Korte was the\\nmasked robber. Surveillance video from one of the robberies\\nshowed a car registered to the address that Korte provided to\\nhis parole officer. An LASD officer who saw video of the\\n\\n UNITED STATES V. KORTE 5\\n\\nmasked robber also reported that the individual resembled\\nKorte, who the officer knew was on parole for bank robbery.\\n\\n On November 4, 2016, without a warrant or Korte’s\\nconsent, LASD placed a GPS tracking device on Korte’s car\\nand periodically monitored the vehicle’s movements over\\nthe next six days. That same day, the Government obtained\\na court order under the Stored Communications Act\\n(“SCA”), 18 U.S.C. § 2703(d), to acquire Korte’s CSLI.\\nThis information placed Korte’s cell phone near three of the\\nfour banks at the time of the respective robberies.\\n\\n On November 10, 2016, LASD learned that the FBI had\\nobtained an arrest warrant for Korte. Officers followed\\nKorte as he drove from his home to a bank and parked\\nnearby, seeming to surveil his next target. An officer saw\\nKorte open the car’s trunk and place something inside.\\nShortly thereafter, they arrested Korte and searched the car.\\nThe officers’ search of the trunk revealed the toy gun used\\nduring the three armed robberies and the shirt Korte had been\\nwearing while casing the bank just prior to his arrest.\\n\\n A grand jury indicted Korte for one count of attempted\\nbank robbery in violation of 18 U.S.C. § 2113(a) and three\\ncounts of bank robbery in violation of 18 U.S.C. § 2113(a),\\n(d). Korte pled not guilty and moved to suppress (1) the\\nevidence found in his car’s trunk, (2) the information derived\\nfrom the GPS tracker on his car, and (3) his CSLI. The\\ndistrict court denied all three suppression motions. The court\\nreasoned that Korte’s parole status permitted the warrantless\\nsearch of the trunk and placement of the GPS tracker on his\\ncar. As for the CSLI, it held that even if the acquisition of\\nthis information violated Korte’s Fourth Amendment rights,\\nsee Carpenter v. United States, 138 S. Ct. 2206 (2018), the\\ngood-faith exception clearly applied based on the case law\\nat that time.\\n\\n6 UNITED STATES V. KORTE\\n\\n Korte went to trial. As to Count 1 — the unsuccessful\\nrobbery — the bank teller testified that even though\\nbulletproof glass separated her from the robber and she saw\\nno weapon, she was nevertheless “[k]ind of panicked,”\\n“[s]hocked,” and “[s]cared.” Korte filed a motion for a\\njudgment of acquittal, under Federal Rule of Criminal\\nProcedure 29, focusing on Count 1. The district court\\nrejected his argument that the Government failed to prove\\nthe element of “force and violence, or by intimidation,” as\\n18 U.S.C. § 2113(a) requires. Citing the teller’s “testimony\\nwith regard to a mask and the demand,” the court found\\nsufficient evidence of “intimidation” to send the matter to\\nthe jury.\\n\\n The jury returned a guilty verdict as to all four counts.\\nKorte was sentenced to 210 months in prison.\\n\\nII. DISCUSSION\\n\\n We review de novo the denial of Korte’s suppression\\nmotions. See United States v. Zapien, 861 F.3d 971, 974 (9th\\nCir. 2017) (per curiam). We address each search — of his\\ncar’s trunk, the GPS tracker on his car, and the acquisition\\nof his CSLI — in turn.\\n\\nA. Search of the Trunk\\n\\n We first examine whether Korte’s parole-search\\ncondition authorized the warrantless search of his car’s\\ntrunk. See United States v. Cervantes, 859 F.3d 1175, 1183\\n(9th Cir. 2017) (“A search of a parolee that complies with\\nthe terms of a valid search condition will usually be deemed\\nreasonable under the Fourth Amendment.”).\\n\\n California Penal Code section 3067(b)(3) provides that\\nevery parolee “is subject to search or seizure . . . at any time\\n\\n UNITED STATES V. KORTE 7\\n\\nof the day or night, with or without a search warrant or with\\nor without cause.” In Samson v. California, 547 U.S. 843,\\n846 (2006), the United States Supreme Court reviewed\\nCalifornia’s parole-search condition to determine “whether\\na suspicionless search, conducted under the authority of\\n[section 3067], violates the Constitution.” The Court first\\nreasoned that “parole is more akin to imprisonment.” Id. at\\n850. Because parolees “have severely diminished\\nexpectations of privacy by virtue of their status alone” —\\neven less than probationers — and because “[t]he State’s\\ninterests” in supervising parolees and reducing recidivism\\n“are substantial,” the Court upheld California’s parole-\\nsearch condition. 1 Id. at 852–53.\\n\\n Korte does not — and cannot — argue that officers\\nunlawfully searched his car. He admits that he rented the car\\nand referred to it as “my car.” As “property under [his]\\ncontrol,” California’s parole-search condition authorized the\\n 1\\n There are two limitations on this condition. First, the officer\\nconducting the search must know at the time that the individual is\\ncurrently on parole. See Moreno v. Baca, 431 F.3d 633, 641 (9th Cir.\\n2005). Second, the search cannot be “arbitrary, capricious, or\\nharassing.” See CAL. PENAL CODE § 3067(d) (“It is not the intent of the\\nLegislature to authorize law enforcement officers to conduct searches for\\nthe sole purpose of harassment.”); see also Samson, 547 U.S. at 856\\n(“The concern that California’s suspicionless search system gives\\nofficers unbridled discretion to conduct searches . . . is belied by\\nCalifornia’s prohibition on ‘arbitrary, capricious or harassing’ searches.”\\n(citations omitted)).\\n\\n Here, it is undisputed that the officers knew that Korte was on\\nparole, and there is no evidence the officers engaged in an arbitrary,\\ncapricious, or harassing search. Nor does Korte contend he was unaware\\nof the parole-search condition. See People v. Lilienthal, 587 P.2d 706,\\n711 (Cal. 1978) (“The condition itself provides general notice to a\\ndefendant that he or his belongings may be subjected to warrantless\\nsearches.”).\\n\\n8 UNITED STATES V. KORTE\\n\\nwarrantless search of it. See, e.g., United States v. Caseres,\\n533 F.3d 1064, 1075–76 (9th Cir. 2008) (explaining that a\\nsearch of the parolee’s car would have been lawful if the\\nofficer had known that the defendant was on parole).\\n\\n Yet, Korte contends that a lawful parole search of his car\\ndoes not extend to the trunk because the trunk is not similarly\\n“property under [his] control.” We reject his narrow\\ninterpretation of this condition. Property is subject to search\\nwhen a parolee “exhibit[s] a sufficiently strong connection\\nto [the property in question] to demonstrate ‘control’ over\\nit.” United States v. Grandberry, 730 F.3d 968, 980 (9th Cir.\\n2013) (defining “property under [a parolee’s] control”); see\\nalso Cervantes, 859 F.3d at 1183 (explaining that having a\\nkey to and belongings inside a hotel room sufficiently\\ndemonstrate control, even if the room is co-occupied).\\nConsistent with our understanding of the condition, the\\nSupreme Court of California explained that a parolee\\ncontrols property based on “the nexus between the parolee\\nand the area or items searched,” including the “nature of that\\narea or item” and “how close and accessible the area or item\\nis to the parolee.” People v. Schmitz, 288 P.3d 1259, 1270\\n(Cal. 2012) (holding that a parolee, who is only a passenger\\nin a third-party’s vehicle, is in control of areas within his\\nreach in the passenger compartment).\\n\\n Korte’s uncontested control over the car was sufficient\\nto permit a warrantless search of its trunk. In any event, his\\nconduct also illustrated a sufficiently close nexus to the trunk\\nitself. Officers observed him putting things inside the trunk.\\n\\n Permitting a warrantless search of the trunk of a\\nparolee’s car is also consistent with broader Fourth\\nAmendment precedent. Generally, a lawful search of a fixed\\nspace or premise extends to its entire area, whether or not\\nthat requires opening a confined space: “[N]ice distinctions\\n\\n UNITED STATES V. KORTE 9\\n\\nbetween closets, drawers, and containers, in the case of a\\nhome, or between glove compartments, upholstered seats,\\ntrunks, and wrapped packages, in the case of a vehicle, must\\ngive way to the interest in the prompt and efficient\\ncompletion of the task at hand.” United States v. Ross, 456\\nU.S. 798, 821 (1982). This reasoning applies to a car’s\\ntrunk. See, e.g., id. at 825 (“If probable cause justifies the\\nsearch of a lawfully stopped vehicle, it justifies the search of\\nevery part of the vehicle and its contents that may conceal\\nthe object of the search.”); United States v. Ewing, 638 F.3d\\n1226, 1231 (9th Cir. 2011) (“[I]f a law enforcement officer\\nhas probable cause to search a vehicle, that probable cause\\nextends to all contents in the vehicle that could be connected\\nto the suspected activity.”); United States v. McWeeney, 454\\nF.3d 1030, 1035 (9th Cir. 2006) (holding that consent to\\nsearch a defendant’s car extends to the trunk). We also note\\nthat Korte’s proposed distinction between a car and trunk\\nwas already rejected by the Supreme Court of California in\\nthe context of a probationer – who has more Fourth\\nAmendment protection than Korte as a parolee. See\\nLilienthal, 587 P.2d at 711 (“We conclude that the officers\\nwere justified in searching defendant’s car trunk pursuant to\\ndefendant’s consent to warrantless searches as a condition of\\nhis probation.”).\\n\\n As property under his control, Korte fails to explain how\\nsearching his car’s trunk would offend the Fourth\\nAmendment when a warrantless search of his home — the\\napex of constitutionally protected places — would not. See\\nUnited States v. Lopez, 474 F.3d 1208, 1213–14 (9th Cir.\\n2007) (upholding the warrantless search of a parolee’s\\nhome), overruled on other grounds by United States v. King,\\n687 F.3d 1189 (9th Cir. 2012) (en banc) (per curiam); see\\nalso Florida v. Jardines, 569 U.S. 1, 6 (2013) (“But when it\\ncomes to the Fourth Amendment, the home is first among\\n\\n10 UNITED STATES V. KORTE\\n\\nequals.”). The district court summed it up well: “If an\\nindividual’s residence falls within the scope of an\\nappropriate search under the parolee provisions, how would\\na trunk not?”\\n\\n Accordingly, the district court correctly held that the\\nsearch of Korte’s trunk was a lawful parole search.\\n\\nB. Placement and Use of the GPS Tracker\\n\\n We next decide whether Korte’s parole-search condition\\npermitted the warrantless placement of a GPS device on his\\ncar and the subsequent surveillance of his car’s movements.\\n\\n Installing a GPS tracker on a car constitutes a search,\\ntypically requiring a warrant. See United States v. Jones,\\n565 U.S. 400, 404 (2012). The application of this principle\\nto a parolee’s car, however, is less clear. While the parties\\nagree that placing the device on Korte’s car was a search\\nunder the Fourth Amendment, they disagree on whether\\ndoing so without a warrant offends the Constitution. In light\\nof Samson’s strong pronouncement that parolees in\\nCalifornia have very limited Fourth Amendment rights, 547\\nU.S. at 851–52, we agree with the district court that this was\\na lawful parole search.\\n\\n Our decision in United States v. Johnson, 875 F.3d 1265\\n(9th Cir. 2017), instructs us not to necessarily apply a newly\\nestablished Fourth Amendment protection to parolees. In\\nRiley v. California, 573 U.S. 373, 386 (2014), the United\\nStates Supreme Court held that the warrantless search of an\\narrestee’s cell phone was unlawful. The Court emphasized\\nthe significant privacy intrusion that arose when searching a\\nperson’s cell phone. Id. at 393; see also Johnson, 875 F.3d\\nat 1274 (noting that Riley was premised on “privacy interests\\nimplicated in cell phone searches [being] particularly\\n\\n UNITED STATES V. KORTE 11\\n\\nacute”). Cell phones, the Supreme Court said, are now “a\\npervasive and insistent part of daily life” that “place vast\\nquantities of personal information literally in the hands of\\nindividuals.” Riley, 573 U.S. at 385–86. Because cell\\nphones “collect[] in one place many distinct types of\\ninformation . . . that reveal much more in combination than\\nany isolated record,” searching a cell phone would give law\\nenforcement the unparalleled ability to reconstruct “[t]he\\nsum of an individual’s private life.” Id. at 394.\\n\\n Despite the Court’s cautionary words, we held that Riley\\ndid not apply to parolees. Johnson, 875 F.3d at 1275; but\\nsee United States v. Lara, 815 F.3d 605, 612 (9th Cir. 2016)\\n(applying Riley to the warrantless search of a probationer’s\\ncell phone). Rather, noting that “the balance of privacy\\ninterests and factual circumstances in this context are\\ndifferent,” we permitted the warrantless search of a parolee’s\\ncell phone. Johnson, 875 F.3d at 1273.\\n\\n In light of our ruling in Johnson, we are hard-put to say\\nthat the warrantless placement of a GPS tracker on a\\nparolee’s car is impermissible. If an officer can conduct a\\nwarrantless search of a parolee’s cell phone — an object that\\nis “[t]he sum of an individual’s private life,” Riley, 573 U.S.\\nat 394 — placing a GPS device on a parolee’s car cannot\\nlogically demand more constitutional protection. Although\\na GPS tracker may create a summary of a parolee’s public\\nmovements, it offers none of the “vast quantities of personal\\ninformation” that a cell phone does. Id. at 386.\\n\\n The State’s interest in supervising parolees is also\\nparticularly strong here. See Samson, 547 U.S. at 853\\n(referencing the State’s “substantial” interests in “reducing\\nrecidivism,” “promoting reintegration,” and deterring future\\ncriminal conduct). Tracking a parolee’s movements by car\\ncan be a critical tool for monitoring this group. Its value is\\n\\n12 UNITED STATES V. KORTE\\n\\nwell illustrated here: Korte returned to a life of crime just\\nmonths after his release from prison, but LASD was able to\\ninvestigate Korte and prevented other armed robberies by\\ntracking his movements. In Johnson, we explained that\\nrequiring officers to obtain a warrant before searching a\\nparolee’s cell phone “would often undermine the state’s\\nability to supervise effectively.” 875 F.3d at 1274. We have\\nsimilar concerns with requiring officers to obtain a warrant\\nbefore tracking a parolee’s vehicular movements.\\n\\n Lastly, we note that our decision aligns with another\\ncourt’s interpretation of California’s parole-search\\ncondition. In People v. Zichwic, 114 Cal. Rptr. 2d 733, 738–\\n39 (Ct. App. 2001), the California Court of Appeal\\nconsidered the warrantless placement of an electronic\\nmonitoring device on a parolee’s car. It held that even\\n“assum[ing] that attaching an electronic tracking device to\\nthe undercarriage of defendant’s truck constituted a search,\\nit was authorized by defendant’s parole search condition.”\\nId. at 740. Granted, Zichwic differs in that it considered the\\nuse of a beeper, id. at 738, rather than a GPS device —\\ninvestigatory methods treated differently under the Fourth\\nAmendment. Compare United States v. Karo, 468 U.S. 705,\\n713 (1984) (“[N]o Fourth Amendment interest . . . was\\ninfringed by the installation of the beeper.”), with Jones, 565\\nU.S. at 409 n.6 (noting that GPS tracking is a more intrusive\\nlaw enforcement practice than a beeper).\\n\\n Nonetheless, we believe that Zichwic is, at minimum,\\ninformative. Although it was decided almost two decades\\nbefore Jones, its holding was unaffected because it had\\nassumed that placing a tracking device on a car constituted a\\nsearch. Zichwic, 114 Cal. Rptr. 2d at 740. The Supreme\\nCourt of California has not since interpreted the parole-\\nsearch condition differently. And, while a beeper and GPS\\n\\n UNITED STATES V. KORTE 13\\n\\ndevice might differ in their tracking capabilities, the court in\\nZichwic at least concluded, as we do, that the State’s need\\nfor electronically monitoring a parolee’s movements\\noutweighs the privacy interests at issue. Id. at 739–40.\\n\\n We do not disregard the importance of Jones. We\\nacknowledge that “GPS monitoring generates a precise,\\ncomprehensive record of a person’s public movements.”\\nJones, 565 U.S. at 415 (Sotomayor, J., concurring).\\nHowever, in following precedent distinguishing Fourth\\nAmendment rights in the parolee context, we hold that the\\nwarrantless placement of a GPS tracker on Korte’s car does\\nnot violate the Fourth Amendment.\\n\\nC. Warrantless CSLI Acquisition\\n\\n Finally, we consider whether the district court should\\nhave excluded the CSLI evidence, acquired without a\\nwarrant, as a Fourth Amendment violation. The Supreme\\nCourt granted certiorari to resolve the constitutionality of\\nwarrantless CSLI acquisition before the district court could\\nrule on Korte’s suppression motion. Carpenter v. United\\nStates, 137 S. Ct. 2211 (2017) (mem.). As such, the district\\ncourt did not address the constitutional issue before it, but\\ndenied Korte’s motion under the Fourth Amendment’s good-\\nfaith exception. We, therefore, consider only a narrow issue:\\nwhether the good-faith exception applies to the warrantless\\nacquisition of a defendant’s CSLI before Carpenter v.\\nUnited States, 138 S. Ct. 2206 (2018). 2\\n\\n\\n\\n\\n 2\\n We take no stance on the constitutionality of acquiring a parolee’s\\nCSLI without a warrant.\\n\\n14 UNITED STATES V. KORTE\\n\\n In Carpenter, the Supreme Court summarized CSLI and\\nhow this technology permits almost real-time compilation of\\na person’s location during any given period:\\n\\n Cell phones continuously scan their\\n environment looking for the best signal,\\n which generally comes from the closest cell\\n site. Most modern devices . . . tap into the\\n wireless network several times a minute\\n whenever their signal is on, even if the owner\\n is not using one of the phone’s features. Each\\n time the phone connects to a cell site, it\\n generates a time-stamped record known as\\n cell-site location information (CSLI).\\n\\nId. at 2211. The Court wrestled with how to apply this “new\\nphenomenon” under the Fourth Amendment, id. at 2216,\\nnoting that it “does not fit neatly under existing precedents,”\\nid. at 2214. It ultimately concluded, however, that the\\nGovernment must obtain a warrant to access a person’s CSLI\\nfrom a wireless carrier, and could no longer rely on a court\\norder under § 2703(d) of the SCA. Id. at 2221.\\n\\n Although the Government obtained Korte’s CSLI\\nwithout a warrant, he is not automatically entitled to relief.\\nSee United States v. Leon, 468 U.S. 897, 906–07 (1984)\\n(explaining that whether there was a Fourth Amendment\\nviolation and “[w]hether the exclusionary sanction is\\nappropriately imposed” are separate questions). Illinois v.\\nKrull, 480 U.S. 340, 342, 350 (1987), established an\\nimportant exception to the exclusionary rule: Evidence\\nobtained by the Government, acting in “objectively\\nreasonable reliance upon a statute” that is “ultimately found\\nto violate the Fourth Amendment,” does not require\\nsuppression.\\n\\n UNITED STATES V. KORTE 15\\n\\n Because we find the Government reasonably relied on\\nthe SCA when it obtained Korte’s CSLI, we affirm the\\ndistrict court’s application of the Fourth Amendment’s\\ngood-faith exception. See also United States v. Camou, 773\\nF.3d 932, 944 (9th Cir. 2014) (placing the burden on the\\ngovernment to prove it acted in good faith). Before\\nCarpenter, the SCA authorized a government entity to\\nrequest “a provider of electronic communication service . . .\\nto disclose a record or other information pertaining to a\\nsubscriber to or a customer of such service.” 18 U.S.C.\\n§ 2703(c)(1). The statute explicitly authorized retrieval of\\nthese records by court order if the Government “offer[ed]\\nspecific and articulable facts showing that there are\\nreasonable grounds to believe that . . . the records or other\\ninformation sought, are relevant and material to an ongoing\\ncriminal investigation” — a more lenient standard than\\nprobable cause. Id. § 2703(d).\\n\\n When the Government obtained Korte’s CSLI — before\\nCarpenter was decided — acting by court order was still\\nauthorized. Moreover, we cannot say that the Government\\nhad any reason to doubt the SCA’s constitutionality, such\\nthat it may have been acting in bad faith. See Krull, 480 U.S.\\nat 355 (“[T]he standard of reasonableness . . . is an objective\\none.”). CSLI remained a relatively novel form of evidence.\\nAnd, although we had not yet commented on the\\nconstitutionality of warrantless CSLI acquisition, a number\\nof our sister circuits had. All had affirmed the SCA’s\\nconstitutionality under the Fourth Amendment. See United\\nStates v. Gilton, No. 16-10109, slip op. at 20 (9th Cir. Mar.\\n4, 2019) (explaining that, at least as of 2012, “the prevailing\\nbelief” was “that CSLI data was not protected by the Fourth\\nAmendment”); see also United States v. Graham, 824 F.3d\\n421, 424 (4th Cir. 2016) (en banc); United States v.\\nCarpenter, 819 F.3d 880, 884 (6th Cir. 2016); United States\\n\\n16 UNITED STATES V. KORTE\\n\\nv. Davis, 785 F.3d 498, 511 (11th Cir. 2015) (en banc); In re\\nApplication of the U.S. for Historical Cell Site Data, 724\\nF.3d 600, 615 (5th Cir. 2013); In re Application of the U.S.\\nfor an Order Directing a Provider of Elec. Commc’n Serv.\\nto Disclose Records to the Gov’t, 620 F.3d 304, 313 (3d Cir.\\n2010).\\n\\n As explained in Krull, it is hardly objectively\\nunreasonable to rely on a then-lawful statute when courts\\nwere upholding it or similar legislative schemes. 480 U.S.\\nat 358–59. Officials should not be “expected to question the\\njudgment of the legislature that passed the law,” particularly\\nwhen confronted with the pattern of judicial holdings as\\nexisted here. Id. at 350. Any “defect in the [SCA],”\\ntherefore, “was not sufficiently obvious so as to render [the\\nGovernment’s] reliance upon the statute objectively\\nunreasonable.” Id. at 359. The Supreme Court’s own\\nsharply divided opinion in Carpenter brings this point to\\nbear. See 138 S. Ct. at 2217 (explaining that the CSLI issue\\npresents “novel circumstances”); see also id. at 2223\\n(referring to the majority’s decision as a “stark departure\\nfrom relevant Fourth Amendment precedents”) (Kennedy,\\nJ., dissenting).\\n\\n Moreover, our application of Krull is anything but novel.\\nSeveral other circuits have already invoked this good-faith\\nexception when presented with similar facts. See United\\nStates v. Goldstein, 914 F.3d 200, 203–05 (3d Cir. 2019);\\nUnited States v. Curtis, 901 F.3d 846, 848–49 (7th Cir.\\n2018); see also United States v. Chambers, No. 16-163-cr,\\n2018 WL 4523607, at *2 (2d Cir. Sept. 21, 2018) (explaining\\nthat “the authorities sought information from third parties by\\ncomplying with the SCA” and “[r]eliance on a federal statute\\ngives rise to a presumption of good faith” (citing Krull, 480\\nU.S. at 349)). Others have analyzed the issue similarly,\\n\\n UNITED STATES V. KORTE 17\\n\\nalthough without directly invoking Krull. See United States\\nv. Joyner, 899 F.3d 1199, 1205 (11th Cir. 2018) (per curiam)\\n(noting that “the Government complied with the\\nrequirements of the SCA in obtaining the orders to compel\\ncell site records” while it was still lawful); United States v.\\nChavez, 894 F.3d 593, 608 (4th Cir. 2018) (refusing to apply\\nthe exclusionary rule because “investigators in this case\\nreasonably relied on court orders and the [SCA] in obtaining\\nthe cell site records”). Similarly, we recently refused to\\nexclude CSLI data obtained in good-faith reliance on a\\nwarrant, later deemed to be defective. Gilton, slip op. at 21.\\n\\n Finally, we note that our decision accords with the\\nexclusionary rule’s limited purpose to deter future Fourth\\nAmendment violations, rather than remedy the rights of a\\nsingle aggrieved party. See United States v. Calandra, 414\\nU.S. 338, 347–48 (1974); see also Krull, 480 U.S. at 347.\\nFor us to exclude CSLI, obtained in good faith based on a\\nthen-lawful legislative scheme, would do nothing to prevent\\nfuture Fourth Amendment violations. See Davis v. United\\nStates, 564 U.S. 229, 236–37 (2011) (“The [exclusionary]\\nrule’s sole purpose, we have repeatedly held, is to deter\\nfuture Fourth Amendment violations.”). With the\\nexclusionary rule as “our last resort, not our first impulse,”\\nHudson v. Michigan, 547 U.S. 586, 591 (2006), we hold that\\nCSLI acquired pre-Carpenter is admissible — so long as the\\nGovernment satisfied the SCA’s then-lawful requirements\\n— under Krull’s good-faith exception.\\n\\n18 UNITED STATES V. KORTE\\n\\nIII. CONCLUSION\\n\\n For the reasons provided above, we affirm the district\\ncourt’s denial of Korte’s suppression motions and his\\nconvictions for bank robbery. 3\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1295,"cells":{"id":{"kind":"number","value":4603411,"string":"4,603,411"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Rebecca Morris v. California Physicians' Service"},"decision_date":{"kind":"string","value":"2019-03-18"},"docket_number":{"kind":"string","value":"17-55878"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Mary M. Schroeder and Jacqueline H. Nguyen, Circuit Judges, and Thomas J. Whelan,* District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"SCHROEDER, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nREBECCA MORRIS, individually No. 17-55878\\nand on behalf of all others\\nsimilarly situated; BECKY D.C. No.\\nEBENKAMP, individually and on 2:16-cv-05914-JAK-\\nbehalf of all others similarly JPR\\nsituated,\\n Plaintiffs-Appellants,\\n OPINION\\n v.\\n\\nCALIFORNIA PHYSICIANS’\\nSERVICE, DBA Blue Shield of\\nCalifornia; DOES, 1–10, inclusive,\\n Defendants-Appellees.\\n\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n John A. Kronstadt, District Judge, Presiding\\n\\n Argued and Submitted October 10, 2018\\n Pasadena, California\\n\\n Filed March 18, 2019\\n\\f2 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\n Before: Mary M. Schroeder and Jacqueline H. Nguyen,\\n Circuit Judges, and Thomas J. Whelan,* District Judge.\\n\\n Opinion by Judge Schroeder\\n\\n\\n SUMMARY**\\n\\n\\n Patient Protection and Affordable Care Act\\n\\n The panel affirmed the district court’s dismissal of a\\nclaim that plaintiffs’ insurer violated the Patient Protection\\nand Affordable Care Act’s “Medical Loss Ratio” provision.\\n\\n This provision of the ACA requires an insurer to pay a\\nrebate to enrollees if the ratio between what it pays out in\\nclaims for medical services is less than 80% of the revenue it\\ntakes in. The panel held that, in determining its Medical Loss\\nRatio under 42 U.S.C. § 300gg-18, the defendant insurer\\nproperly included as part of its payout the payments it made\\nin settling a dispute with some of its enrollees, and there was\\nno basis for excluding payments for services rendered by out-\\nof-network physicians.\\n\\n\\n\\n\\n *\\n The Honorable Thomas J. Whelan, United States District Judge for\\nthe Southern District of California, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 3\\n\\n COUNSEL\\n\\nJay Angoff (argued) and Christine H. Monahan, Mehri &\\nSkalet PLLC, Washington, D.C.; Dan Stormer, Randy\\nRenick, and Cornelia Dai, Hadsell Stormer & Renick LLP,\\nPasadena, California; for Plaintiffs-Appellants.\\n\\nGregory N. Pimstone (argued), Craig S. Bloomgarden, and\\nJoanna S. McCallum, Manatt, Phelps & Phillips, LLP, Los\\nAngeles, California; Michael S. Kolber, Manatt, Phelps &\\nPhillips, LLP, New York, New York; for Defendant-Appellee\\nCalifornia Physicians’ Service.\\n\\n\\n OPINION\\n\\nSCHROEDER, Circuit Judge:\\n\\n In this appeal, Plaintiffs-Appellants contend that their\\ninsurer, Blue Shield of California (“Blue Shield”), violated an\\nintegral provision of the Patient Protection and Affordable\\nCare Act (“ACA”), the Medical Loss Ratio (“MLR”), 42\\nU.S.C. § 300gg-18, a provision that has not yet been\\ninterpreted by our Court or our sister circuits. The MLR is\\nthe ratio between what an insurer pays out in claims for\\nmedical services and the revenue it takes in. Id.\\n§ 300gg-18(a). The insurer must pay a rebate to enrollees if\\nthe payout is less than 80% of the revenue. Id.\\n§ 300gg-18(b)(1). Plaintiffs in this case, enrollees seeking a\\nlarger rebate, argued that Blue Shield improperly included as\\npart of its payout the payments it made in settling a dispute\\nwith some of its enrollees. The district court dismissed\\nPlaintiffs’ action, ruling that pursuant to the settlement, the\\n\\f4 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\npayments had been made, whether earlier disputed or not, and\\nwere therefore properly included. We affirm.\\n\\n INTRODUCTION\\n\\nA. The Medical Loss Ratio (“MLR”) Defined\\n\\n Congress enacted the ACA in 2010 to decrease the cost of\\nhealth care and increase the number of Americans with health\\ninsurance. See Nat’l Fed. of Indep. Bus. v. Sebelius, 567 U.S.\\n519, 538 (2012). The MLR plays a key role in furthering\\nCongress’ plan to decrease health care costs by requiring\\nhealth insurance companies to spend at least 80 percent of\\ntheir premium income on health care claims and health\\nquality improvement. 42 U.S.C. § 300gg-18(a), (b). Health\\ninsurance companies that do not meet the 80 percent spending\\nrequirement must refund to their enrollees the difference\\nbetween the amount actually spent and the 80 percent figure.\\nId. § 300gg-18(b)(1). For example, if a health insurance\\ncompany has spent only 70 percent of premiums on clinical\\nservices and health improvements, its enrollees are entitled to\\na 10 percent rebate of premium revenue. This rule is intended\\nto ensure that spending is focused on health care expenses,\\nas opposed to administrative costs such as salaries\\nor marketing. Id. § 300gg-18(b)(2); 45 C.F.R.\\n§ 158.140(b)(3)(iii).\\n\\n The statute spells out the enforcement scheme. Health\\ninsurance companies must calculate and annually report their\\nMLR to the Department of Health and Human Services\\n(HHS), the agency tasked with enforcing the provision.\\n42 U.S.C. § 300gg-18(a). The instructions for calculating the\\nratio are provided under the MLR provision of the ACA and\\nfurther explained in federal regulations. The statute defines\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 5\\n\\nthe MLR as “the ratio of the incurred loss (or incurred claims)\\nplus the loss adjustment expense (or change in contract\\nreserves) to earned premiums.” Id. The MLR regulations\\ndefine “incurred claims” to include payments made by an\\nissuer for “clinical services” and to exclude administrative\\nexpenses or work unrelated to clinical services. 45 C.F.R.\\n§ 158.140(a), (b)(3)(iii). Thus, the MLR compares what the\\nhealth insurance company has spent on clinical services to the\\npremium revenue the insurance company received from its\\nenrollees. The statute requires a rebate when reported\\namounts paid out for actual clinical and related services are\\nless than 80% of reported premium revenue. It thereby\\nencourages insurers to use premium revenue to reimburse the\\ncosts of enrollees’ medical treatment rather than to use it on\\nadministrative expenses. Indeed, the statutory section\\ncontaining the MLR is entitled “Bringing down the cost of\\nhealth care coverage.” 42 U.S.C. § 300gg-18.\\n\\nB. This Dispute\\n\\n Blue Shield was selected by the State of California in\\n2010 to provide affordable health care plans on the state’s\\nhealth insurance exchange, “Covered California.” Under\\nBlue Shield’s plans, “participating” or “in-network” providers\\naccepted Blue Shield Covered California patients and charged\\nfor medical services at Blue Shield’s participating provider\\nbenefit level. “Out-of-network” providers billed for medical\\nservices at higher rates than the in-network participating\\nprovider benefit level.\\n\\n Shortly after Blue Shield was selected as an insurance\\ncompany on California’s health insurance exchange,\\nhowever, enrollees who had purchased Blue Shield Covered\\nCalifornia plans began complaining of difficulty finding in-\\n\\f6 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\nnetwork providers charging in-network rates. The enrollee\\ncomplaints led to the discovery that Blue Shield had\\nerroneously listed out-of-network physicians in its network\\ndirectory who had not in fact agreed to accept Blue Shield\\nCovered California patients. As a result of Blue Shield’s\\nnetwork directory mistake, some enrollees were charged the\\nhigher, out-of-network rates for medical services they\\nreceived. To remedy the problem, Blue Shield executed a\\nsettlement agreement to reprocess the out-of-network claims\\nat in-network rates, and to reimburse enrollees for the extra\\nmoney they were charged on account of using an out-of-\\nnetwork provider. Blue Shield then included these settlement\\npayments in the numerator of the MLR as “incurred claims,”\\nor payments for clinical services. According to the settlement\\nagreement, Blue Shield reimbursements already totaled more\\nthan 35 million dollars.\\n\\n Two Blue Shield enrollees, Becky Ebenkamp and\\nRebecca Morris, on behalf of a class of more than 446,000\\nenrollees (“Plaintiffs”), filed an action in Los Angeles County\\nSuperior Court in 2016, alleging Blue Shield violated the\\nACA by including the settlement payments in the MLR\\nnumerator. The named plaintiffs are Blue Shield enrollees\\nwho had received a MLR rebate for 2014, but they asserted\\nthat Blue Shield’s settlement payments improperly inflated\\nBlue Shield’s MLR, thereby unfairly decreasing their rebate.\\nIn Plaintiffs’ view, under the terms of the statute and of the\\nsettlement agreement, Blue Shield should have included in\\nthe MLR numerator only reimbursements for services\\nprovided by in-network providers. Plaintiffs asserted that\\nincluding the payments for services rendered by providers\\noutside those in the policy’s network violated the federal\\nstatute and regulations and resulted in the insurer’s having\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 7\\n\\nbeen unjustly enriched through unfair competition. This\\nallegedly entitled Plaintiffs to damages under state law.\\n\\n Blue Shield timely removed the case to the Central\\nDistrict of California and filed a motion to dismiss pursuant\\nto Federal Rule of Civil Procedure 12(b)(6). The district\\ncourt reviewed the MLR statute, regulations, and the\\nsettlement agreement. It concluded that the settlement\\npayments compensated enrollees for clinical services covered\\nby their policies, regardless of whether the providers who\\nfurnished the services were outside of Blue Shield’s network.\\nThe district court reasoned that because the payments\\ncompensated enrollees for clinical services, Blue Shield\\ncorrectly included those payments under the MLR numerator\\nas “incurred claims.” We affirm the district court’s judgment\\nof dismissal.\\n\\n Plaintiffs are understandably desirous of receiving a\\nlarger rebate, but the MLR is an integral part of the ACA’s\\nintended purpose to broaden access to health care and\\ndecrease health care costs. The purpose of the MLR is thus\\nto encourage use of premium income to provide benefits to\\ninsureds and discourage its use to offset administrative costs,\\nthus serving the primary goal of expanding affordable care.\\nSee 42 U.S.C. § 300gg-18(b)(2); King v. Burwell, 135 S. Ct.\\n2480, 2485 (2015). The statute does not make a distinction\\nbetween in-network and out-of-network providers. We must\\nconclude that the MLR can act as an effective incentive to\\ninsurers to provide benefits only if disputes are discouraged\\nand all benefit payments are included in the ratio, whether\\nthey go to providers in-network or out-of-network.\\n\\n This is best illustrated by looking to the history of the\\nACA and how its key provisions relevant to this dispute are\\n\\f8 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\ndesigned to operate. We therefore turn first to the statute\\nitself.\\n\\n STATUTORY BACKGROUND\\n\\n The ACA was enacted in 2010 after “a long history of\\nfailed health insurance reform.” See King, 135 S. Ct. at 2485.\\nPast experiences in many states had led to what the Supreme\\nCourt termed “an economic ‘death spiral.’” Id. at 2486. As\\nthe price of premiums increased, the number of people\\npurchasing health insurance dropped dramatically. Id.\\n\\n The outcome was very different in Massachusetts,\\nhowever, and Congress turned to it as a model. Id. Adopting\\nsome of the key reforms that made the Massachusetts\\ninsurance system successful, the ACA represented the most\\nsignificant regulatory overhaul and national expansion of\\nhealth care coverage since Medicare and Medicaid in 1965.\\nSee id.; see also Sebelius, 567 U.S. at 538–39, 541. The ACA\\nwas signed into law in 2010, and many of its private health\\ninsurance provisions went into effect in 2014. See, e.g., 42\\nU.S.C. §§ 18022(c)(1), 18041(c)(1); see King, 135 S. Ct. at\\n2487; Sebelius, 567 U.S. at 538. As the Supreme Court\\nemphasized, its purpose is to provide affordable, quality\\nhealth insurance for all Americans and decrease the cost of\\nhealth care. Sebelius, 567 U.S. at 538.\\n\\n Composed of ten titles spanning over 900 pages and\\nhundreds of provisions, the ACA brought sweeping reforms\\nto our health care system. Id. at 538–39. Under the ACA,\\nhealth insurance companies must accept every individual who\\napplies for coverage. 42 U.S.C. § 300gg-1(a). These insurers\\nare prohibited from charging higher premiums based on the\\ncondition of an applicant’s health. See id. § 300gg(a).\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 9\\n\\n To facilitate the purchasing of affordable health plans, the\\nACA created government-regulated health insurance\\n“exchanges.” Id. § 18031(b)(1). Exchanges are essentially\\nonline marketplaces where people can shop for insurance.\\nKing, 135 S. Ct. at 2487. The ACA directs each state to\\nestablish an exchange. 42 U.S.C. § 18031(b)(1). If a state\\ndecides not to establish an exchange, the ACA provides that\\nthe exchange shall be established and operated by the\\nSecretary of Health and Human Services. Id. § 18041(c)(1).\\nFurther, the federal government has published a website\\n(http://www.healthcare.gov) that explains the ACA to\\nindividual health insurance consumers and provides\\ninstructions on how to obtain coverage.\\n\\n While policymakers continue to debate the ACA’s future,\\nthere is research demonstrating that coverage expansion has\\nparticularly benefitted two historically vulnerable populations\\n— low-income adults and those with chronic conditions.\\nBenjamin D. Sommers, Bethany Maylone, Robert J. Blendon,\\nE. John Orav, and Arnold M. Epstein, Three-Year Impacts of\\nthe Affordable Care Act: Improved Medical Care and Health\\nAmong Low-Income Adults, 36 Health Affairs 1119 (2017).\\nAccording to at least one report, the number of uninsured\\nAmericans under the age of 65 has decreased from 44 million\\nin 2013 to less than 27 million at the end of 2016. Key Facts\\nAbout the Uninsured Population, Henry J Kaiser Family\\nFoundation (Dec. 7, 2018), https://www.kff.org/uninsured/\\nfact-sheet/key-facts-about-the-uninsured-population/.\\n\\n As we have seen, this appeal concerns an important\\nprovision of the ACA commonly referred to as the Medical\\nLoss Ratio or “MLR.” 42 U.S.C. § 300gg-18(a). Although\\nthe basic formula for calculating the MLR ratio is defined by\\nthe ACA and federal regulations, the parties here dispute how\\n\\f10 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\nthe ratio should be calculated in this case. The Centers for\\nMedicare & Medicaid Services (CMS), a division within\\nHHS, promulgated the standards for MLR calculation and\\nreporting. 45 C.F.R. § 158.120. CMS requires a health\\ninsurer to submit, for each state in which it provides\\ncoverage, data on the aggregate premiums, claims experience,\\nquality improvement expenditures, and non-claims costs it\\nincurs in connection with the policies it issues. See id.\\n§§ 158.120, 158.150, 150.160. The MLR report must detail\\nthe issuer’s spending on medical benefits, including clinical\\nservices and activities that improve health care, compared to\\nthe revenue the issuer receives from enrollees. 42 U.S.C.\\n§ 300gg-18(a). The required reporting does not distinguish\\nbetween in-network and out-of-network providers.\\n\\n The applicable regulations explain the MLR is to act as an\\nincentive to provide medical services and reduce\\nadministrative costs. 45 C.F.R. § 158.140(b)(3)(iii). The\\nMLR is thus intended to further the ACA’s goal of decreasing\\nhealth care costs by providing greater transparency on how\\nconsumers’ premium dollars are used and incentivizing\\nissuers to maximize spending on health care and activities\\nthat improve health care quality, thereby promoting greater\\nefficiency in health insurance markets. Medical Loss Ratio\\nRequirements Under the Patient Protection and Affordable\\nCare Act, 77 Fed. Reg. 28,790, 28,791, 28,793 (May 16,\\n2012). HHS has further explained that the MLR is intended\\nto ensure that “policyholders receive value for their premium\\ndollars” because the higher the MLR, the more value a\\npolicyholder receives for each premium dollar. Health\\nInsurance Issuers Implementing Medical Loss Ratio (MLR)\\nRequirements Under the Patient Protection and Affordable\\nCare Act, 75 Fed. Reg. 74864-01, 74865 (Dec. 1, 2010). The\\npurpose of the ACA, as demonstrated by the content of its\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 11\\n\\nprovisions and the implementing regulations, as well as its\\nhistory, is to broaden access to health care. This strongly\\nindicates that the ACA should be interpreted and applied in\\na manner that encourages insurers to spend premium revenue\\non medical services.\\n\\n FACTUAL BACKGROUND\\n\\n This dispute began when consumers purchased Blue\\nShield insurance plans through the State of California’s health\\ninsurance exchange. The ACA established health insurance\\nexchanges in all 50 states. The exchanges are regulated,\\nonline marketplaces administered by either the federal or\\nstate government, where individuals and small businesses can\\npurchase affordable, private health insurance plans. These\\nexchanges commenced operation on October 1, 2013, with\\nhealth insurance coverage beginning on January 1, 2014.\\n\\n To implement the ACA’s requirements, California created\\na state-run health exchange, known as Covered California. It\\nselected Blue Shield as one of the insurance companies to\\noffer health care plans on the Covered California\\nmarketplace. Individuals who enrolled in a Blue Shield plan\\nthrough Covered California were able to look up physicians\\naccepting Covered California patients in an online directory\\nposted on Blue Shield’s website. Blue Shield Covered\\nCalifornia enrollees could also select a doctor by calling Blue\\nShield’s customer service representatives, who in turn shared\\nparticipating provider information from the directory.\\n\\n Shortly after the online Covered California marketplace\\nwent live in late 2013, the California Department of Managed\\nHeath Care (“DMHC”), the state regulatory body that\\ngoverns managed health care plans, began receiving\\n\\f12 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\ncomplaints from Covered California consumers who had\\nsigned up for a Blue Shield plan. These enrollees reported\\ndifficulty finding in-network physicians. In response, DMHC\\nconducted an informal telephone survey of randomly selected\\nproviders that Blue Shield had identified as contracted\\nphysicians for the Blue Shield Covered California plans.\\nDuring the course of this survey, several offices informed\\nDMHC that they were not in fact accepting Covered\\nCalifornia enrollees.\\n\\n DMHC then began a formal survey of the entire Blue\\nShield provider directory. On November 17, 2014, DHMC\\nissued its final report which revealed that approximately nine\\npercent of the physicians listed in the Blue Shield online\\nprovider directory were not accepting Covered California\\npatients. Blue Shield acknowledged it had mistakenly listed\\nout-of-network providers in its directory, and also\\nacknowledged the confusion it caused among enrollees and\\nproviders. To resolve the problem, Blue Shield agreed to\\nupdate and correct the provider list and to reprocess\\nout-of-network claims at in-network rates. In October 2015,\\nDMHC and Blue Shield executed a settlement agreement\\ndescribing the provider directory mistake and outlining Blue\\nShield’s obligations to remedy it.\\n\\n Under Exhibit A of the settlement agreement, entitled\\n“Corrective Action,” Blue Shield agreed to adjust claims for\\nenrollees who had paid out-of-network rates for services of\\nproviders mistakenly listed as in-network. They were defined\\nas enrollees who “paid a provider in excess of the amount\\nthey would have paid if the claim had been processed at the\\nparticipating provider benefit level as a result of inaccuracies\\nin its [p]rovider [d]irectory during the 2014 and 2015 benefit\\nyears.” To be reimbursed, enrollees were required to attest\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 13\\n\\nthat they received “covered services.” Paragraph six of\\nExhibit A describes the process that required enrollees to\\nidentify the specific services they received from a provider\\noutside the network:\\n\\n The [Blue Shield] notice shall advise\\n recipients that if they received covered\\n services from a provider outside of their\\n plan’s network as a result of misinformation\\n from Blue Shield regarding the provider’s\\n participation status, and paid the provider out\\n of pocket for care, they may submit a written\\n claims submission form to a designated\\n address within 30 days to request reprocessing\\n of the claim. The claims submission form\\n shall require members to attest: a. that they\\n received covered services from a specific\\n provider who was represented in Blue\\n Shield’s Provider Directory or by Blue Shield\\n as participating in the network. . . .\\n\\nEnrollees were also required to list the date of their visit, and\\nconfirm that they paid the provider for the services at issue or\\nthat the provider was actively seeking payment. The\\nsubmission form included an explanation of benefits or claim\\nnumber so that Blue Shield could identify and reprocess the\\ndisputed claim. If an enrollee submitted all of the required\\ninformation identified in paragraph six of Exhibit A, and Blue\\nShield confirmed that the provider directory mistake led to\\nthe claim or alternatively that the attestation was acceptable,\\nBlue Shield was required to reprocess the claim at the\\nparticipating provider benefit level. Blue Shield paid over\\n$38 million in claims adjustments to enrollees who had\\nreceived clinical services from providers that did not accept\\n\\f14 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\nCovered California patients. Blue Shield included the claims\\nadjustments in the MLR report for 2014, but nevertheless fell\\nshort of the 80% requirement and paid rebates to enrollees.\\nPlaintiffs in this case are Blue Shield enrollees who received\\na MLR rebate for 2014, but now claim they are entitled to a\\nlarger rebate because Blue Shield improperly included in the\\nMLR the claim adjustments paid out under the settlement\\nagreement.\\n\\n PROCEDURAL HISTORY\\n\\n This case began in state court, because Plaintiffs claimed\\nthe alleged MLR miscalculation amounted to unfair\\ncompetition and unjust enrichment under California law. On\\nJuly 1, 2016, Plaintiffs filed a class action in Los Angeles\\nCounty Superior Court, alleging Blue Shield improperly\\nincluded the settlement payments as “incurred claims” in the\\nMLR numerator, so that enrollees did not receive the rebate\\nthey should have received. Blue Shield removed the case to\\nthe Central District of California and moved to dismiss the\\ncomplaint. Rather than opposing the motion, Plaintiffs filed\\na First Amended Complaint (“FAC”). The FAC alleged the\\nclaim adjustments should not have been included in the MLR\\nbecause they were restitution payments pursuant to the\\nsettlement agreement for payments to out-of-network\\nproviders and not to providers covered by the policy.\\n\\n Blue Shield moved to dismiss Plaintiffs’ FAC, arguing\\nthe restitution payments fall within the definition of “incurred\\nclaims” because they were for covered services, and that it\\ndid not matter for purposes of the MLR whether the providers\\nwere in or out-of-network. Plaintiffs opposed the motion,\\nasserting that because the providers were not in-network, the\\nservices were not covered under the policy, and therefore the\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 15\\n\\npayments did not fall within the definition of “incurred\\nclaims” or payments for services “covered by the policy.” 45\\nC.F.R. § 158.140(a). The district court granted Blue Shield’s\\nmotion to dismiss without leave to amend. It concluded that\\nbecause the settlement agreement provided reimbursement\\nfor services that were covered by the policy, these payments\\nwere properly included in the MLR numerator. The district\\ncourt explained, “[a]lthough the providers were not covered\\nby the policies, their services were.” This appeal followed.\\n\\n DISCUSSION\\n\\n The dispositive issue in this appeal is whether the MLR\\nwas properly calculated under federal law by including the\\nsettlement reimbursements for medical services by non-\\nnetwork providers. If there was no violation of the applicable\\nfederal law, we need not decide whether such an alleged\\nviolation could constitute a violation of California consumer\\nprotection laws.\\n\\n Plaintiffs’ position is that the reimbursements do not\\ncount because they were for services from out-of-network\\nproviders. We look first to the provisions of the statute to see\\nif it supports such a distinction. It does not. The statute\\nrequires annual reporting of the ratio of claims\\nreimbursements to premium revenue. 42 U.S.C. § 300gg-\\n18(a). Reimbursements for clinical services must be included\\nin the MLR numerator as “incurred claims.” Id. § 300gg-\\n18(a)(1). In relevant part it provides:\\n\\n (a) Clear accounting for costs\\n\\n A health insurance issuer offering group or\\n individual health insurance coverage\\n\\f16 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\n (including a grandfathered health plan) shall,\\n with respect to each plan year, submit to the\\n Secretary a report concerning the ratio of the\\n incurred loss (or incurred claims) plus the loss\\n adjustment expense (or change in contract\\n reserves) to earned premiums. Such report\\n shall include the percentage of total premium\\n revenue, after accounting for collections or\\n receipts for risk adjustment and risk corridors\\n and payments of reinsurance, that such\\n coverage expends--\\n\\n (1) on reimbursement for clinical services\\n provided to enrollees under such coverage;\\n\\n (2) for activities that improve health care\\n quality; and\\n\\n (3) on all other non-claims costs, including an\\n explanation of the nature of such costs, and\\n excluding Federal and State taxes and\\n licensing or regulatory fees. . . .\\n\\nId. The critical statutory focus is on health insurers providing\\ncoverage for health services. The statute measures total\\nexpenditures on health services. See id. § 300gg-\\n18(a)(1)–(3). For purposes of the statutory provisions, which\\nthe parties agree are controlling, the relative amounts required\\nto be paid to in-network as opposed to out-of-network\\nproviders is irrelevant. Only the total matters.\\n\\n Plaintiffs point out that the numerator’s reimbursement\\nfor clinical services under the statute has some limitation, in\\nthat to be included in the numerator, the services must fall\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 17\\n\\nwithin the policy’s coverage. The statute refers to\\nreimbursement “for clinical services . . . under such\\ncoverage.” Id. § 300gg-18(a)(1). This limitation, however,\\nis on the nature of the services that qualify for inclusion, not\\non any characteristic or qualification of the provider.\\n\\n Plaintiffs therefore go on to focus their argument on the\\ntext of the applicable regulations and specifically the\\ndefinition of “incurred claims” to be included in the\\nnumerator. These must include payments to providers\\n“whose services are covered by the policy.” 45 C.F.R.\\n§ 158.140(a).\\n\\n (a) General requirements. The report required\\n in § 158.110 must include direct claims paid\\n to or received by providers, including under\\n capitation contracts with physicians, whose\\n services are covered by the policy for clinical\\n services or supplies covered by the policy. In\\n addition, the report must include claim\\n reserves associated with claims incurred\\n during the MLR reporting year, the change in\\n contract reserves, reserves for contingent\\n benefits and the medical claim portion of\\n lawsuits, and any incurred experience rating\\n refunds. Reimbursement for clinical services,\\n as defined in this section, is referred to as\\n “incurred claims.” All components of and\\n adjustments to incurred claims, with the\\n exception of contract reserves, must be\\n calculated based on claims incurred only\\n during the MLR reporting year and paid\\n through March 31st of the following year.\\n\\f18 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\n Contract reserves must be calculated as of\\n December 31st of the applicable year.\\n\\nId. Plaintiffs contend that the condition that the services be\\ncovered by the policy means that the providers must be in-\\nnetwork. Yet the regulation does not say that. Like the\\nstatute, the regulation makes no distinction between in-\\nnetwork and out-of-network providers.\\n\\n Nor does the settlement agreement assist Plaintiffs. It is\\nintended to compensate enrollees who paid out-of-network\\nrates because Blue Shield mistakenly listed some providers as\\nbeing in-network. It does no more than establish that Blue\\nShield agreed to provide the compensation in satisfaction of\\nclaims, and to that extent, it supports Blue Shield’s position\\nthat such amounts should be included in the MLR numerator.\\nThe settlement agreement does not expressly refer to the\\nMLR and in no way suggests that the settlement payments\\nshould be excluded from the MLR numerator.\\n\\n We do not minimize the differences in the way various\\nplans treat services performed by in-network as opposed to\\nout-of-network providers. Whether and to what extent a\\nprovider’s services are covered depends on the type of\\ninsurance plan. A government website is helpful in\\nexplaining the differences. Health Insurance Plan &\\nNetwork Types: HMOs, PPOs, and More, Healthcare.gov,\\nhttps://www.healthcare.gov/choose-a-plan/plan-types/ (last\\nvisited February 4, 2019). The insurance universe can be\\nbewildering. For the purposes of this case, however, none of\\nthe relevant documents or provisions of federal law contain\\nany language that supports Plaintiffs’ contention that the\\nMLR numerator must exclude payments to out-of-network\\nproviders.\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 19\\n\\n Nor would there be any point in straining the language in\\nthe manner Plaintiffs suggest in order to narrow the definition\\nof claims that should be included. This is because the\\npurpose of the ACA was to broaden access to healthcare and\\nreduce costs. The MLR was enacted to further that purpose\\nby incentivizing insurers to pay for health services. The more\\nthe percentage of premium income paid out in benefits falls\\nbelow 80%, the greater the rebate the insurer must pay to the\\nenrollees.\\n\\n Congress passed the ACA to expand health insurance\\ncoverage. See King, 135 S. Ct. at 2485, 2493. The MLR 80\\npercent spending requirement on incurred claims furthers this\\ngoal by encouraging health insurance companies to pay for\\nclinical services. Should that spending requirement on\\nclinical services not be met, the MLR rebate serves the\\nACA’s twin goal of lowering health care costs by\\nreimbursing to consumers some of the premium money paid\\nto the health insurance company. See Sebelius, 567 U.S. at\\n538; 42 U.S.C. § 300gg-18. The 80 percent spending\\nrequirement on incurred claims and the rebate work together\\nto create an incentive for health insurance companies to\\nmaximize the amount of money spent on health care and\\nactivities that promote health care. See Medical Loss Ratio\\nRequirements Under the Patient Protection and Affordable\\nCare Act, 77 Fed. Reg. 28,790, 28,791, 28,793 (May 16,\\n2012).\\n\\n Thus, Congress was quite clearly concerned with\\nexpanding benefits and reducing costs. The MLR is\\ninclusively defined with respect to payouts and contains an\\nexpress exclusion for administrative costs. The structure is\\nmeant to encourage insurers to pay benefits and to discourage\\ndisputes. Plaintiffs ask us to turn the system on its head by\\n\\f20 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\nessentially contending that the goal should be to provide\\nrebates to enrollees rather than insurance benefits. Insureds\\npay premiums in order to be eligible for reimbursement of\\nhealth care costs, not for rebates. Plaintiffs seek to draw a\\ntype of distinction that would discourage the payment of\\nbenefits and encourage costly disputes like the one that is at\\nthe root of this case.\\n\\n For all of these reasons, we conclude there is no basis in\\nthe language, history, intent or spirit of the ACA to narrow\\nthe MLR by excluding payments for services rendered by\\nout-of-network physicians. As the district court correctly\\nrecognized, the services were covered by the plan and the\\npayments were made.\\n\\n The district court did not err in dismissing the action\\nwithout leave to amend because Plaintiffs never asked for\\nleave to amend the complaint. See Alaska v. United States,\\n201 F.3d 1154, 1163 (9th Cir. 2000). Plaintiffs belatedly ask\\nus to remand so that they may amend the complaint to claim\\nthat some of the reimbursements were for services outside the\\nplan. The request is untimely. See id. at 1163–64. It is also\\ninconsistent with the history of the underlying dispute.\\nPlaintiffs have never asserted that the services for which\\nreimbursement was made were not covered by the plan. This\\ndispute has always been over the increased costs of out-of-\\nnetwork providers rendering covered services, and that is the\\ndispute the settlement agreement resolved. Plaintiffs chose\\nto litigate how that resolution should be treated in the MLR,\\nand can not now go back and try to change the terms of the\\nsettlement itself. Those terms were clear: Blue Shield agreed\\nto reprocess claims for clinical services at in-network rates\\nonly if the enrollee submitted a written form attesting he or\\nshe received covered services from a provider who was\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 21\\n\\nerroneously listed in the Blue Shield provider directory.\\nEnrollees were not eligible for a claims adjustment if the\\nservices they received were not covered by their plan. Thus,\\nthe settlement agreement sought to remedy the provider\\ndirectory mistake as to the cost of covered services. There is\\nno valid reason to reopen it.\\n\\n The district court’s judgment of dismissal must be\\naffirmed.\\n\\n AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4380664/\", \"author_raw\": \"SCHROEDER, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"MARY M SCHROEDER"},"judge2":{"kind":"string","value":"JACQUELINE H NGUYEN"},"judge3":{"kind":"string","value":"THOMAS J WHELAN"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4603411/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Rebecca Morris v. California Physicians' Service\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-18\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nREBECCA MORRIS, individually No. 17-55878\nand on behalf of all others\nsimilarly situated; BECKY D.C. No.\nEBENKAMP, individually and on 2:16-cv-05914-JAK-\nbehalf of all others similarly JPR\nsituated,\n Plaintiffs-Appellants,\n OPINION\n v.\n\nCALIFORNIA PHYSICIANS’\nSERVICE, DBA Blue Shield of\nCalifornia; DOES, 1–10, inclusive,\n Defendants-Appellees.\n\n\n Appeal from the United States District Court\n for the Central District of California\n John A. Kronstadt, District Judge, Presiding\n\n Argued and Submitted October 10, 2018\n Pasadena, California\n\n Filed March 18, 2019\n\f2 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\n\n Before: Mary M. Schroeder and Jacqueline H. Nguyen,\n Circuit Judges, and Thomas J. Whelan,* District Judge.\n\n Opinion by Judge Schroeder\n\n\n SUMMARY**\n\n\n Patient Protection and Affordable Care Act\n\n The panel affirmed the district court’s dismissal of a\nclaim that plaintiffs’ insurer violated the Patient Protection\nand Affordable Care Act’s “Medical Loss Ratio” provision.\n\n This provision of the ACA requires an insurer to pay a\nrebate to enrollees if the ratio between what it pays out in\nclaims for medical services is less than 80% of the revenue it\ntakes in. The panel held that, in determining its Medical Loss\nRatio under 42 U.S.C. § 300gg-18, the defendant insurer\nproperly included as part of its payout the payments it made\nin settling a dispute with some of its enrollees, and there was\nno basis for excluding payments for services rendered by out-\nof-network physicians.\n\n\n\n\n *\n The Honorable Thomas J. Whelan, United States District Judge for\nthe Southern District of California, sitting by designation.\n **\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 3\n\n COUNSEL\n\nJay Angoff (argued) and Christine H. Monahan, Mehri &\nSkalet PLLC, Washington, D.C.; Dan Stormer, Randy\nRenick, and Cornelia Dai, Hadsell Stormer & Renick LLP,\nPasadena, California; for Plaintiffs-Appellants.\n\nGregory N. Pimstone (argued), Craig S. Bloomgarden, and\nJoanna S. McCallum, Manatt, Phelps & Phillips, LLP, Los\nAngeles, California; Michael S. Kolber, Manatt, Phelps &\nPhillips, LLP, New York, New York; for Defendant-Appellee\nCalifornia Physicians’ Service.\n\n\n OPINION\n\nSCHROEDER, Circuit Judge:\n\n In this appeal, Plaintiffs-Appellants contend that their\ninsurer, Blue Shield of California (“Blue Shield”), violated an\nintegral provision of the Patient Protection and Affordable\nCare Act (“ACA”), the Medical Loss Ratio (“MLR”), 42\nU.S.C. § 300gg-18, a provision that has not yet been\ninterpreted by our Court or our sister circuits. The MLR is\nthe ratio between what an insurer pays out in claims for\nmedical services and the revenue it takes in. Id.\n§ 300gg-18(a). The insurer must pay a rebate to enrollees if\nthe payout is less than 80% of the revenue. Id.\n§ 300gg-18(b)(1). Plaintiffs in this case, enrollees seeking a\nlarger rebate, argued that Blue Shield improperly included as\npart of its payout the payments it made in settling a dispute\nwith some of its enrollees. The district court dismissed\nPlaintiffs’ action, ruling that pursuant to the settlement, the\n\f4 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\n\npayments had been made, whether earlier disputed or not, and\nwere therefore properly included. We affirm.\n\n INTRODUCTION\n\nA. The Medical Loss Ratio (“MLR”) Defined\n\n Congress enacted the ACA in 2010 to decrease the cost of\nhealth care and increase the number of Americans with health\ninsurance. See Nat’l Fed. of Indep. Bus. v. Sebelius, 567 U.S.\n519, 538 (2012). The MLR plays a key role in furthering\nCongress’ plan to decrease health care costs by requiring\nhealth insurance companies to spend at least 80 percent of\ntheir premium income on health care claims and health\nquality improvement. 42 U.S.C. § 300gg-18(a), (b). Health\ninsurance companies that do not meet the 80 percent spending\nrequirement must refund to their enrollees the difference\nbetween the amount actually spent and the 80 percent figure.\nId. § 300gg-18(b)(1). For example, if a health insurance\ncompany has spent only 70 percent of premiums on clinical\nservices and health improvements, its enrollees are entitled to\na 10 percent rebate of premium revenue. This rule is intended\nto ensure that spending is focused on health care expenses,\nas opposed to administrative costs such as salaries\nor marketing. Id. § 300gg-18(b)(2); 45 C.F.R.\n§ 158.140(b)(3)(iii).\n\n The statute spells out the enforcement scheme. Health\ninsurance companies must calculate and annually report their\nMLR to the Department of Health and Human Services\n(HHS), the agency tasked with enforcing the provision.\n42 U.S.C. § 300gg-18(a). The instructions for calculating the\nratio are provided under the MLR provision of the ACA and\nfurther explained in federal regulations. The statute defines\n\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 5\n\nthe MLR as “the ratio of the incurred loss (or incurred claims)\nplus the loss adjustment expense (or change in contract\nreserves) to earned premiums.” Id. The MLR regulations\ndefine “incurred claims” to include payments made by an\nissuer for “clinical services” and to exclude administrative\nexpenses or work unrelated to clinical services. 45 C.F.R.\n§ 158.140(a), (b)(3)(iii). Thus, the MLR compares what the\nhealth insurance company has spent on clinical services to the\npremium revenue the insurance company received from its\nenrollees. The statute requires a rebate when reported\namounts paid out for actual clinical and related services are\nless than 80% of reported premium revenue. It thereby\nencourages insurers to use premium revenue to reimburse the\ncosts of enrollees’ medical treatment rather than to use it on\nadministrative expenses. Indeed, the statutory section\ncontaining the MLR is entitled “Bringing down the cost of\nhealth care coverage.” 42 U.S.C. § 300gg-18.\n\nB. This Dispute\n\n Blue Shield was selected by the State of California in\n2010 to provide affordable health care plans on the state’s\nhealth insurance exchange, “Covered California.” Under\nBlue Shield’s plans, “participating” or “in-network” providers\naccepted Blue Shield Covered California patients and charged\nfor medical services at Blue Shield’s participating provider\nbenefit level. “Out-of-network” providers billed for medical\nservices at higher rates than the in-network participating\nprovider benefit level.\n\n Shortly after Blue Shield was selected as an insurance\ncompany on California’s health insurance exchange,\nhowever, enrollees who had purchased Blue Shield Covered\nCalifornia plans began complaining of difficulty finding in-\n\f6 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\n\nnetwork providers charging in-network rates. The enrollee\ncomplaints led to the discovery that Blue Shield had\nerroneously listed out-of-network physicians in its network\ndirectory who had not in fact agreed to accept Blue Shield\nCovered California patients. As a result of Blue Shield’s\nnetwork directory mistake, some enrollees were charged the\nhigher, out-of-network rates for medical services they\nreceived. To remedy the problem, Blue Shield executed a\nsettlement agreement to reprocess the out-of-network claims\nat in-network rates, and to reimburse enrollees for the extra\nmoney they were charged on account of using an out-of-\nnetwork provider. Blue Shield then included these settlement\npayments in the numerator of the MLR as “incurred claims,”\nor payments for clinical services. According to the settlement\nagreement, Blue Shield reimbursements already totaled more\nthan 35 million dollars.\n\n Two Blue Shield enrollees, Becky Ebenkamp and\nRebecca Morris, on behalf of a class of more than 446,000\nenrollees (“Plaintiffs”), filed an action in Los Angeles County\nSuperior Court in 2016, alleging Blue Shield violated the\nACA by including the settlement payments in the MLR\nnumerator. The named plaintiffs are Blue Shield enrollees\nwho had received a MLR rebate for 2014, but they asserted\nthat Blue Shield’s settlement payments improperly inflated\nBlue Shield’s MLR, thereby unfairly decreasing their rebate.\nIn Plaintiffs’ view, under the terms of the statute and of the\nsettlement agreement, Blue Shield should have included in\nthe MLR numerator only reimbursements for services\nprovided by in-network providers. Plaintiffs asserted that\nincluding the payments for services rendered by providers\noutside those in the policy’s network violated the federal\nstatute and regulations and resulted in the insurer’s having\n\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 7\n\nbeen unjustly enriched through unfair competition. This\nallegedly entitled Plaintiffs to damages under state law.\n\n Blue Shield timely removed the case to the Central\nDistrict of California and filed a motion to dismiss pursuant\nto Federal Rule of Civil Procedure 12(b)(6). The district\ncourt reviewed the MLR statute, regulations, and the\nsettlement agreement. It concluded that the settlement\npayments compensated enrollees for clinical services covered\nby their policies, regardless of whether the providers who\nfurnished the services were outside of Blue Shield’s network.\nThe district court reasoned that because the payments\ncompensated enrollees for clinical services, Blue Shield\ncorrectly included those payments under the MLR numerator\nas “incurred claims.” We affirm the district court’s judgment\nof dismissal.\n\n Plaintiffs are understandably desirous of receiving a\nlarger rebate, but the MLR is an integral part of the ACA’s\nintended purpose to broaden access to health care and\ndecrease health care costs. The purpose of the MLR is thus\nto encourage use of premium income to provide benefits to\ninsureds and discourage its use to offset administrative costs,\nthus serving the primary goal of expanding affordable care.\nSee 42 U.S.C. § 300gg-18(b)(2); King v. Burwell, 135 S. Ct.\n2480, 2485 (2015). The statute does not make a distinction\nbetween in-network and out-of-network providers. We must\nconclude that the MLR can act as an effective incentive to\ninsurers to provide benefits only if disputes are discouraged\nand all benefit payments are included in the ratio, whether\nthey go to providers in-network or out-of-network.\n\n This is best illustrated by looking to the history of the\nACA and how its key provisions relevant to this dispute are\n\f8 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\n\ndesigned to operate. We therefore turn first to the statute\nitself.\n\n STATUTORY BACKGROUND\n\n The ACA was enacted in 2010 after “a long history of\nfailed health insurance reform.” See King, 135 S. Ct. at 2485.\nPast experiences in many states had led to what the Supreme\nCourt termed “an economic ‘death spiral.’” Id. at 2486. As\nthe price of premiums increased, the number of people\npurchasing health insurance dropped dramatically. Id.\n\n The outcome was very different in Massachusetts,\nhowever, and Congress turned to it as a model. Id. Adopting\nsome of the key reforms that made the Massachusetts\ninsurance system successful, the ACA represented the most\nsignificant regulatory overhaul and national expansion of\nhealth care coverage since Medicare and Medicaid in 1965.\nSee id.; see also Sebelius, 567 U.S. at 538–39, 541. The ACA\nwas signed into law in 2010, and many of its private health\ninsurance provisions went into effect in 2014. See, e.g., 42\nU.S.C. §§ 18022(c)(1), 18041(c)(1); see King, 135 S. Ct. at\n2487; Sebelius, 567 U.S. at 538. As the Supreme Court\nemphasized, its purpose is to provide affordable, quality\nhealth insurance for all Americans and decrease the cost of\nhealth care. Sebelius, 567 U.S. at 538.\n\n Composed of ten titles spanning over 900 pages and\nhundreds of provisions, the ACA brought sweeping reforms\nto our health care system. Id. at 538–39. Under the ACA,\nhealth insurance companies must accept every individual who\napplies for coverage. 42 U.S.C. § 300gg-1(a). These insurers\nare prohibited from charging higher premiums based on the\ncondition of an applicant’s health. See id. § 300gg(a).\n\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 9\n\n To facilitate the purchasing of affordable health plans, the\nACA created government-regulated health insurance\n“exchanges.” Id. § 18031(b)(1). Exchanges are essentially\nonline marketplaces where people can shop for insurance.\nKing, 135 S. Ct. at 2487. The ACA directs each state to\nestablish an exchange. 42 U.S.C. § 18031(b)(1). If a state\ndecides not to establish an exchange, the ACA provides that\nthe exchange shall be established and operated by the\nSecretary of Health and Human Services. Id. § 18041(c)(1).\nFurther, the federal government has published a website\n(http://www.healthcare.gov) that explains the ACA to\nindividual health insurance consumers and provides\ninstructions on how to obtain coverage.\n\n While policymakers continue to debate the ACA’s future,\nthere is research demonstrating that coverage expansion has\nparticularly benefitted two historically vulnerable populations\n— low-income adults and those with chronic conditions.\nBenjamin D. Sommers, Bethany Maylone, Robert J. Blendon,\nE. John Orav, and Arnold M. Epstein, Three-Year Impacts of\nthe Affordable Care Act: Improved Medical Care and Health\nAmong Low-Income Adults, 36 Health Affairs 1119 (2017).\nAccording to at least one report, the number of uninsured\nAmericans under the age of 65 has decreased from 44 million\nin 2013 to less than 27 million at the end of 2016. Key Facts\nAbout the Uninsured Population, Henry J Kaiser Family\nFoundation (Dec. 7, 2018), https://www.kff.org/uninsured/\nfact-sheet/key-facts-about-the-uninsured-population/.\n\n As we have seen, this appeal concerns an important\nprovision of the ACA commonly referred to as the Medical\nLoss Ratio or “MLR.” 42 U.S.C. § 300gg-18(a). Although\nthe basic formula for calculating the MLR ratio is defined by\nthe ACA and federal regulations, the parties here dispute how\n\f10 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\n\nthe ratio should be calculated in this case. The Centers for\nMedicare & Medicaid Services (CMS), a division within\nHHS, promulgated the standards for MLR calculation and\nreporting. 45 C.F.R. § 158.120. CMS requires a health\ninsurer to submit, for each state in which it provides\ncoverage, data on the aggregate premiums, claims experience,\nquality improvement expenditures, and non-claims costs it\nincurs in connection with the policies it issues. See id.\n§§ 158.120, 158.150, 150.160. The MLR report must detail\nthe issuer’s spending on medical benefits, including clinical\nservices and activities that improve health care, compared to\nthe revenue the issuer receives from enrollees. 42 U.S.C.\n§ 300gg-18(a). The required reporting does not distinguish\nbetween in-network and out-of-network providers.\n\n The applicable regulations explain the MLR is to act as an\nincentive to provide medical services and reduce\nadministrative costs. 45 C.F.R. § 158.140(b)(3)(iii). The\nMLR is thus intended to further the ACA’s goal of decreasing\nhealth care costs by providing greater transparency on how\nconsumers’ premium dollars are used and incentivizing\nissuers to maximize spending on health care and activities\nthat improve health care quality, thereby promoting greater\nefficiency in health insurance markets. Medical Loss Ratio\nRequirements Under the Patient Protection and Affordable\nCare Act, 77 Fed. Reg. 28,790, 28,791, 28,793 (May 16,\n2012). HHS has further explained that the MLR is intended\nto ensure that “policyholders receive value for their premium\ndollars” because the higher the MLR, the more value a\npolicyholder receives for each premium dollar. Health\nInsurance Issuers Implementing Medical Loss Ratio (MLR)\nRequirements Under the Patient Protection and Affordable\nCare Act, 75 Fed. Reg. 74864-01, 74865 (Dec. 1, 2010). The\npurpose of the ACA, as demonstrated by the content of its\n\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 11\n\nprovisions and the implementing regulations, as well as its\nhistory, is to broaden access to health care. This strongly\nindicates that the ACA should be interpreted and applied in\na manner that encourages insurers to spend premium revenue\non medical services.\n\n FACTUAL BACKGROUND\n\n This dispute began when consumers purchased Blue\nShield insurance plans through the State of California’s health\ninsurance exchange. The ACA established health insurance\nexchanges in all 50 states. The exchanges are regulated,\nonline marketplaces administered by either the federal or\nstate government, where individuals and small businesses can\npurchase affordable, private health insurance plans. These\nexchanges commenced operation on October 1, 2013, with\nhealth insurance coverage beginning on January 1, 2014.\n\n To implement the ACA’s requirements, California created\na state-run health exchange, known as Covered California. It\nselected Blue Shield as one of the insurance companies to\noffer health care plans on the Covered California\nmarketplace. Individuals who enrolled in a Blue Shield plan\nthrough Covered California were able to look up physicians\naccepting Covered California patients in an online directory\nposted on Blue Shield’s website. Blue Shield Covered\nCalifornia enrollees could also select a doctor by calling Blue\nShield’s customer service representatives, who in turn shared\nparticipating provider information from the directory.\n\n Shortly after the online Covered California marketplace\nwent live in late 2013, the California Department of Managed\nHeath Care (“DMHC”), the state regulatory body that\ngoverns managed health care plans, began receiving\n\f12 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\n\ncomplaints from Covered California consumers who had\nsigned up for a Blue Shield plan. These enrollees reported\ndifficulty finding in-network physicians. In response, DMHC\nconducted an informal telephone survey of randomly selected\nproviders that Blue Shield had identified as contracted\nphysicians for the Blue Shield Covered California plans.\nDuring the course of this survey, several offices informed\nDMHC that they were not in fact accepting Covered\nCalifornia enrollees.\n\n DMHC then began a formal survey of the entire Blue\nShield provider directory. On November 17, 2014, DHMC\nissued its final report which revealed that approximately nine\npercent of the physicians listed in the Blue Shield online\nprovider directory were not accepting Covered California\npatients. Blue Shield acknowledged it had mistakenly listed\nout-of-network providers in its directory, and also\nacknowledged the confusion it caused among enrollees and\nproviders. To resolve the problem, Blue Shield agreed to\nupdate and correct the provider list and to reprocess\nout-of-network claims at in-network rates. In October 2015,\nDMHC and Blue Shield executed a settlement agreement\ndescribing the provider directory mistake and outlining Blue\nShield’s obligations to remedy it.\n\n Under Exhibit A of the settlement agreement, entitled\n“Corrective Action,” Blue Shield agreed to adjust claims for\nenrollees who had paid out-of-network rates for services of\nproviders mistakenly listed as in-network. They were defined\nas enrollees who “paid a provider in excess of the amount\nthey would have paid if the claim had been processed at the\nparticipating provider benefit level as a result of inaccuracies\nin its [p]rovider [d]irectory during the 2014 and 2015 benefit\nyears.” To be reimbursed, enrollees were required to attest\n\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 13\n\nthat they received “covered services.” Paragraph six of\nExhibit A describes the process that required enrollees to\nidentify the specific services they received from a provider\noutside the network:\n\n The [Blue Shield] notice shall advise\n recipients that if they received covered\n services from a provider outside of their\n plan’s network as a result of misinformation\n from Blue Shield regarding the provider’s\n participation status, and paid the provider out\n of pocket for care, they may submit a written\n claims submission form to a designated\n address within 30 days to request reprocessing\n of the claim. The claims submission form\n shall require members to attest: a. that they\n received covered services from a specific\n provider who was represented in Blue\n Shield’s Provider Directory or by Blue Shield\n as participating in the network. . . .\n\nEnrollees were also required to list the date of their visit, and\nconfirm that they paid the provider for the services at issue or\nthat the provider was actively seeking payment. The\nsubmission form included an explanation of benefits or claim\nnumber so that Blue Shield could identify and reprocess the\ndisputed claim. If an enrollee submitted all of the required\ninformation identified in paragraph six of Exhibit A, and Blue\nShield confirmed that the provider directory mistake led to\nthe claim or alternatively that the attestation was acceptable,\nBlue Shield was required to reprocess the claim at the\nparticipating provider benefit level. Blue Shield paid over\n$38 million in claims adjustments to enrollees who had\nreceived clinical services from providers that did not accept\n\f14 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\n\nCovered California patients. Blue Shield included the claims\nadjustments in the MLR report for 2014, but nevertheless fell\nshort of the 80% requirement and paid rebates to enrollees.\nPlaintiffs in this case are Blue Shield enrollees who received\na MLR rebate for 2014, but now claim they are entitled to a\nlarger rebate because Blue Shield improperly included in the\nMLR the claim adjustments paid out under the settlement\nagreement.\n\n PROCEDURAL HISTORY\n\n This case began in state court, because Plaintiffs claimed\nthe alleged MLR miscalculation amounted to unfair\ncompetition and unjust enrichment under California law. On\nJuly 1, 2016, Plaintiffs filed a class action in Los Angeles\nCounty Superior Court, alleging Blue Shield improperly\nincluded the settlement payments as “incurred claims” in the\nMLR numerator, so that enrollees did not receive the rebate\nthey should have received. Blue Shield removed the case to\nthe Central District of California and moved to dismiss the\ncomplaint. Rather than opposing the motion, Plaintiffs filed\na First Amended Complaint (“FAC”). The FAC alleged the\nclaim adjustments should not have been included in the MLR\nbecause they were restitution payments pursuant to the\nsettlement agreement for payments to out-of-network\nproviders and not to providers covered by the policy.\n\n Blue Shield moved to dismiss Plaintiffs’ FAC, arguing\nthe restitution payments fall within the definition of “incurred\nclaims” because they were for covered services, and that it\ndid not matter for purposes of the MLR whether the providers\nwere in or out-of-network. Plaintiffs opposed the motion,\nasserting that because the providers were not in-network, the\nservices were not covered under the policy, and therefore the\n\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 15\n\npayments did not fall within the definition of “incurred\nclaims” or payments for services “covered by the policy.” 45\nC.F.R. § 158.140(a). The district court granted Blue Shield’s\nmotion to dismiss without leave to amend. It concluded that\nbecause the settlement agreement provided reimbursement\nfor services that were covered by the policy, these payments\nwere properly included in the MLR numerator. The district\ncourt explained, “[a]lthough the providers were not covered\nby the policies, their services were.” This appeal followed.\n\n DISCUSSION\n\n The dispositive issue in this appeal is whether the MLR\nwas properly calculated under federal law by including the\nsettlement reimbursements for medical services by non-\nnetwork providers. If there was no violation of the applicable\nfederal law, we need not decide whether such an alleged\nviolation could constitute a violation of California consumer\nprotection laws.\n\n Plaintiffs’ position is that the reimbursements do not\ncount because they were for services from out-of-network\nproviders. We look first to the provisions of the statute to see\nif it supports such a distinction. It does not. The statute\nrequires annual reporting of the ratio of claims\nreimbursements to premium revenue. 42 U.S.C. § 300gg-\n18(a). Reimbursements for clinical services must be included\nin the MLR numerator as “incurred claims.” Id. § 300gg-\n18(a)(1). In relevant part it provides:\n\n (a) Clear accounting for costs\n\n A health insurance issuer offering group or\n individual health insurance coverage\n\f16 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\n\n (including a grandfathered health plan) shall,\n with respect to each plan year, submit to the\n Secretary a report concerning the ratio of the\n incurred loss (or incurred claims) plus the loss\n adjustment expense (or change in contract\n reserves) to earned premiums. Such report\n shall include the percentage of total premium\n revenue, after accounting for collections or\n receipts for risk adjustment and risk corridors\n and payments of reinsurance, that such\n coverage expends--\n\n (1) on reimbursement for clinical services\n provided to enrollees under such coverage;\n\n (2) for activities that improve health care\n quality; and\n\n (3) on all other non-claims costs, including an\n explanation of the nature of such costs, and\n excluding Federal and State taxes and\n licensing or regulatory fees. . . .\n\nId. The critical statutory focus is on health insurers providing\ncoverage for health services. The statute measures total\nexpenditures on health services. See id. § 300gg-\n18(a)(1)–(3). For purposes of the statutory provisions, which\nthe parties agree are controlling, the relative amounts required\nto be paid to in-network as opposed to out-of-network\nproviders is irrelevant. Only the total matters.\n\n Plaintiffs point out that the numerator’s reimbursement\nfor clinical services under the statute has some limitation, in\nthat to be included in the numerator, the services must fall\n\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 17\n\nwithin the policy’s coverage. The statute refers to\nreimbursement “for clinical services . . . under such\ncoverage.” Id. § 300gg-18(a)(1). This limitation, however,\nis on the nature of the services that qualify for inclusion, not\non any characteristic or qualification of the provider.\n\n Plaintiffs therefore go on to focus their argument on the\ntext of the applicable regulations and specifically the\ndefinition of “incurred claims” to be included in the\nnumerator. These must include payments to providers\n“whose services are covered by the policy.” 45 C.F.R.\n§ 158.140(a).\n\n (a) General requirements. The report required\n in § 158.110 must include direct claims paid\n to or received by providers, including under\n capitation contracts with physicians, whose\n services are covered by the policy for clinical\n services or supplies covered by the policy. In\n addition, the report must include claim\n reserves associated with claims incurred\n during the MLR reporting year, the change in\n contract reserves, reserves for contingent\n benefits and the medical claim portion of\n lawsuits, and any incurred experience rating\n refunds. Reimbursement for clinical services,\n as defined in this section, is referred to as\n “incurred claims.” All components of and\n adjustments to incurred claims, with the\n exception of contract reserves, must be\n calculated based on claims incurred only\n during the MLR reporting year and paid\n through March 31st of the following year.\n\f18 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\n\n Contract reserves must be calculated as of\n December 31st of the applicable year.\n\nId. Plaintiffs contend that the condition that the services be\ncovered by the policy means that the providers must be in-\nnetwork. Yet the regulation does not say that. Like the\nstatute, the regulation makes no distinction between in-\nnetwork and out-of-network providers.\n\n Nor does the settlement agreement assist Plaintiffs. It is\nintended to compensate enrollees who paid out-of-network\nrates because Blue Shield mistakenly listed some providers as\nbeing in-network. It does no more than establish that Blue\nShield agreed to provide the compensation in satisfaction of\nclaims, and to that extent, it supports Blue Shield’s position\nthat such amounts should be included in the MLR numerator.\nThe settlement agreement does not expressly refer to the\nMLR and in no way suggests that the settlement payments\nshould be excluded from the MLR numerator.\n\n We do not minimize the differences in the way various\nplans treat services performed by in-network as opposed to\nout-of-network providers. Whether and to what extent a\nprovider’s services are covered depends on the type of\ninsurance plan. A government website is helpful in\nexplaining the differences. Health Insurance Plan &\nNetwork Types: HMOs, PPOs, and More, Healthcare.gov,\nhttps://www.healthcare.gov/choose-a-plan/plan-types/ (last\nvisited February 4, 2019). The insurance universe can be\nbewildering. For the purposes of this case, however, none of\nthe relevant documents or provisions of federal law contain\nany language that supports Plaintiffs’ contention that the\nMLR numerator must exclude payments to out-of-network\nproviders.\n\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 19\n\n Nor would there be any point in straining the language in\nthe manner Plaintiffs suggest in order to narrow the definition\nof claims that should be included. This is because the\npurpose of the ACA was to broaden access to healthcare and\nreduce costs. The MLR was enacted to further that purpose\nby incentivizing insurers to pay for health services. The more\nthe percentage of premium income paid out in benefits falls\nbelow 80%, the greater the rebate the insurer must pay to the\nenrollees.\n\n Congress passed the ACA to expand health insurance\ncoverage. See King, 135 S. Ct. at 2485, 2493. The MLR 80\npercent spending requirement on incurred claims furthers this\ngoal by encouraging health insurance companies to pay for\nclinical services. Should that spending requirement on\nclinical services not be met, the MLR rebate serves the\nACA’s twin goal of lowering health care costs by\nreimbursing to consumers some of the premium money paid\nto the health insurance company. See Sebelius, 567 U.S. at\n538; 42 U.S.C. § 300gg-18. The 80 percent spending\nrequirement on incurred claims and the rebate work together\nto create an incentive for health insurance companies to\nmaximize the amount of money spent on health care and\nactivities that promote health care. See Medical Loss Ratio\nRequirements Under the Patient Protection and Affordable\nCare Act, 77 Fed. Reg. 28,790, 28,791, 28,793 (May 16,\n2012).\n\n Thus, Congress was quite clearly concerned with\nexpanding benefits and reducing costs. The MLR is\ninclusively defined with respect to payouts and contains an\nexpress exclusion for administrative costs. The structure is\nmeant to encourage insurers to pay benefits and to discourage\ndisputes. Plaintiffs ask us to turn the system on its head by\n\f20 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\n\nessentially contending that the goal should be to provide\nrebates to enrollees rather than insurance benefits. Insureds\npay premiums in order to be eligible for reimbursement of\nhealth care costs, not for rebates. Plaintiffs seek to draw a\ntype of distinction that would discourage the payment of\nbenefits and encourage costly disputes like the one that is at\nthe root of this case.\n\n For all of these reasons, we conclude there is no basis in\nthe language, history, intent or spirit of the ACA to narrow\nthe MLR by excluding payments for services rendered by\nout-of-network physicians. As the district court correctly\nrecognized, the services were covered by the plan and the\npayments were made.\n\n The district court did not err in dismissing the action\nwithout leave to amend because Plaintiffs never asked for\nleave to amend the complaint. See Alaska v. United States,\n201 F.3d 1154, 1163 (9th Cir. 2000). Plaintiffs belatedly ask\nus to remand so that they may amend the complaint to claim\nthat some of the reimbursements were for services outside the\nplan. The request is untimely. See id. at 1163–64. It is also\ninconsistent with the history of the underlying dispute.\nPlaintiffs have never asserted that the services for which\nreimbursement was made were not covered by the plan. This\ndispute has always been over the increased costs of out-of-\nnetwork providers rendering covered services, and that is the\ndispute the settlement agreement resolved. Plaintiffs chose\nto litigate how that resolution should be treated in the MLR,\nand can not now go back and try to change the terms of the\nsettlement itself. Those terms were clear: Blue Shield agreed\nto reprocess claims for clinical services at in-network rates\nonly if the enrollee submitted a written form attesting he or\nshe received covered services from a provider who was\n\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 21\n\nerroneously listed in the Blue Shield provider directory.\nEnrollees were not eligible for a claims adjustment if the\nservices they received were not covered by their plan. Thus,\nthe settlement agreement sought to remedy the provider\ndirectory mistake as to the cost of covered services. There is\nno valid reason to reopen it.\n\n The district court’s judgment of dismissal must be\naffirmed.\n\n AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Rebecca Morris v. California Physicians' Service\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-18\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nREBECCA MORRIS, individually No. 17-55878\\nand on behalf of all others\\nsimilarly situated; BECKY D.C. No.\\nEBENKAMP, individually and on 2:16-cv-05914-JAK-\\nbehalf of all others similarly JPR\\nsituated,\\n Plaintiffs-Appellants,\\n OPINION\\n v.\\n\\nCALIFORNIA PHYSICIANS’\\nSERVICE, DBA Blue Shield of\\nCalifornia; DOES, 1–10, inclusive,\\n Defendants-Appellees.\\n\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n John A. Kronstadt, District Judge, Presiding\\n\\n Argued and Submitted October 10, 2018\\n Pasadena, California\\n\\n Filed March 18, 2019\\n\\f2 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\n Before: Mary M. Schroeder and Jacqueline H. Nguyen,\\n Circuit Judges, and Thomas J. Whelan,* District Judge.\\n\\n Opinion by Judge Schroeder\\n\\n\\n SUMMARY**\\n\\n\\n Patient Protection and Affordable Care Act\\n\\n The panel affirmed the district court’s dismissal of a\\nclaim that plaintiffs’ insurer violated the Patient Protection\\nand Affordable Care Act’s “Medical Loss Ratio” provision.\\n\\n This provision of the ACA requires an insurer to pay a\\nrebate to enrollees if the ratio between what it pays out in\\nclaims for medical services is less than 80% of the revenue it\\ntakes in. The panel held that, in determining its Medical Loss\\nRatio under 42 U.S.C. § 300gg-18, the defendant insurer\\nproperly included as part of its payout the payments it made\\nin settling a dispute with some of its enrollees, and there was\\nno basis for excluding payments for services rendered by out-\\nof-network physicians.\\n\\n\\n\\n\\n *\\n The Honorable Thomas J. Whelan, United States District Judge for\\nthe Southern District of California, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 3\\n\\n COUNSEL\\n\\nJay Angoff (argued) and Christine H. Monahan, Mehri &\\nSkalet PLLC, Washington, D.C.; Dan Stormer, Randy\\nRenick, and Cornelia Dai, Hadsell Stormer & Renick LLP,\\nPasadena, California; for Plaintiffs-Appellants.\\n\\nGregory N. Pimstone (argued), Craig S. Bloomgarden, and\\nJoanna S. McCallum, Manatt, Phelps & Phillips, LLP, Los\\nAngeles, California; Michael S. Kolber, Manatt, Phelps &\\nPhillips, LLP, New York, New York; for Defendant-Appellee\\nCalifornia Physicians’ Service.\\n\\n\\n OPINION\\n\\nSCHROEDER, Circuit Judge:\\n\\n In this appeal, Plaintiffs-Appellants contend that their\\ninsurer, Blue Shield of California (“Blue Shield”), violated an\\nintegral provision of the Patient Protection and Affordable\\nCare Act (“ACA”), the Medical Loss Ratio (“MLR”), 42\\nU.S.C. § 300gg-18, a provision that has not yet been\\ninterpreted by our Court or our sister circuits. The MLR is\\nthe ratio between what an insurer pays out in claims for\\nmedical services and the revenue it takes in. Id.\\n§ 300gg-18(a). The insurer must pay a rebate to enrollees if\\nthe payout is less than 80% of the revenue. Id.\\n§ 300gg-18(b)(1). Plaintiffs in this case, enrollees seeking a\\nlarger rebate, argued that Blue Shield improperly included as\\npart of its payout the payments it made in settling a dispute\\nwith some of its enrollees. The district court dismissed\\nPlaintiffs’ action, ruling that pursuant to the settlement, the\\n\\f4 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\npayments had been made, whether earlier disputed or not, and\\nwere therefore properly included. We affirm.\\n\\n INTRODUCTION\\n\\nA. The Medical Loss Ratio (“MLR”) Defined\\n\\n Congress enacted the ACA in 2010 to decrease the cost of\\nhealth care and increase the number of Americans with health\\ninsurance. See Nat’l Fed. of Indep. Bus. v. Sebelius, 567 U.S.\\n519, 538 (2012). The MLR plays a key role in furthering\\nCongress’ plan to decrease health care costs by requiring\\nhealth insurance companies to spend at least 80 percent of\\ntheir premium income on health care claims and health\\nquality improvement. 42 U.S.C. § 300gg-18(a), (b). Health\\ninsurance companies that do not meet the 80 percent spending\\nrequirement must refund to their enrollees the difference\\nbetween the amount actually spent and the 80 percent figure.\\nId. § 300gg-18(b)(1). For example, if a health insurance\\ncompany has spent only 70 percent of premiums on clinical\\nservices and health improvements, its enrollees are entitled to\\na 10 percent rebate of premium revenue. This rule is intended\\nto ensure that spending is focused on health care expenses,\\nas opposed to administrative costs such as salaries\\nor marketing. Id. § 300gg-18(b)(2); 45 C.F.R.\\n§ 158.140(b)(3)(iii).\\n\\n The statute spells out the enforcement scheme. Health\\ninsurance companies must calculate and annually report their\\nMLR to the Department of Health and Human Services\\n(HHS), the agency tasked with enforcing the provision.\\n42 U.S.C. § 300gg-18(a). The instructions for calculating the\\nratio are provided under the MLR provision of the ACA and\\nfurther explained in federal regulations. The statute defines\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 5\\n\\nthe MLR as “the ratio of the incurred loss (or incurred claims)\\nplus the loss adjustment expense (or change in contract\\nreserves) to earned premiums.” Id. The MLR regulations\\ndefine “incurred claims” to include payments made by an\\nissuer for “clinical services” and to exclude administrative\\nexpenses or work unrelated to clinical services. 45 C.F.R.\\n§ 158.140(a), (b)(3)(iii). Thus, the MLR compares what the\\nhealth insurance company has spent on clinical services to the\\npremium revenue the insurance company received from its\\nenrollees. The statute requires a rebate when reported\\namounts paid out for actual clinical and related services are\\nless than 80% of reported premium revenue. It thereby\\nencourages insurers to use premium revenue to reimburse the\\ncosts of enrollees’ medical treatment rather than to use it on\\nadministrative expenses. Indeed, the statutory section\\ncontaining the MLR is entitled “Bringing down the cost of\\nhealth care coverage.” 42 U.S.C. § 300gg-18.\\n\\nB. This Dispute\\n\\n Blue Shield was selected by the State of California in\\n2010 to provide affordable health care plans on the state’s\\nhealth insurance exchange, “Covered California.” Under\\nBlue Shield’s plans, “participating” or “in-network” providers\\naccepted Blue Shield Covered California patients and charged\\nfor medical services at Blue Shield’s participating provider\\nbenefit level. “Out-of-network” providers billed for medical\\nservices at higher rates than the in-network participating\\nprovider benefit level.\\n\\n Shortly after Blue Shield was selected as an insurance\\ncompany on California’s health insurance exchange,\\nhowever, enrollees who had purchased Blue Shield Covered\\nCalifornia plans began complaining of difficulty finding in-\\n\\f6 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\nnetwork providers charging in-network rates. The enrollee\\ncomplaints led to the discovery that Blue Shield had\\nerroneously listed out-of-network physicians in its network\\ndirectory who had not in fact agreed to accept Blue Shield\\nCovered California patients. As a result of Blue Shield’s\\nnetwork directory mistake, some enrollees were charged the\\nhigher, out-of-network rates for medical services they\\nreceived. To remedy the problem, Blue Shield executed a\\nsettlement agreement to reprocess the out-of-network claims\\nat in-network rates, and to reimburse enrollees for the extra\\nmoney they were charged on account of using an out-of-\\nnetwork provider. Blue Shield then included these settlement\\npayments in the numerator of the MLR as “incurred claims,”\\nor payments for clinical services. According to the settlement\\nagreement, Blue Shield reimbursements already totaled more\\nthan 35 million dollars.\\n\\n Two Blue Shield enrollees, Becky Ebenkamp and\\nRebecca Morris, on behalf of a class of more than 446,000\\nenrollees (“Plaintiffs”), filed an action in Los Angeles County\\nSuperior Court in 2016, alleging Blue Shield violated the\\nACA by including the settlement payments in the MLR\\nnumerator. The named plaintiffs are Blue Shield enrollees\\nwho had received a MLR rebate for 2014, but they asserted\\nthat Blue Shield’s settlement payments improperly inflated\\nBlue Shield’s MLR, thereby unfairly decreasing their rebate.\\nIn Plaintiffs’ view, under the terms of the statute and of the\\nsettlement agreement, Blue Shield should have included in\\nthe MLR numerator only reimbursements for services\\nprovided by in-network providers. Plaintiffs asserted that\\nincluding the payments for services rendered by providers\\noutside those in the policy’s network violated the federal\\nstatute and regulations and resulted in the insurer’s having\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 7\\n\\nbeen unjustly enriched through unfair competition. This\\nallegedly entitled Plaintiffs to damages under state law.\\n\\n Blue Shield timely removed the case to the Central\\nDistrict of California and filed a motion to dismiss pursuant\\nto Federal Rule of Civil Procedure 12(b)(6). The district\\ncourt reviewed the MLR statute, regulations, and the\\nsettlement agreement. It concluded that the settlement\\npayments compensated enrollees for clinical services covered\\nby their policies, regardless of whether the providers who\\nfurnished the services were outside of Blue Shield’s network.\\nThe district court reasoned that because the payments\\ncompensated enrollees for clinical services, Blue Shield\\ncorrectly included those payments under the MLR numerator\\nas “incurred claims.” We affirm the district court’s judgment\\nof dismissal.\\n\\n Plaintiffs are understandably desirous of receiving a\\nlarger rebate, but the MLR is an integral part of the ACA’s\\nintended purpose to broaden access to health care and\\ndecrease health care costs. The purpose of the MLR is thus\\nto encourage use of premium income to provide benefits to\\ninsureds and discourage its use to offset administrative costs,\\nthus serving the primary goal of expanding affordable care.\\nSee 42 U.S.C. § 300gg-18(b)(2); King v. Burwell, 135 S. Ct.\\n2480, 2485 (2015). The statute does not make a distinction\\nbetween in-network and out-of-network providers. We must\\nconclude that the MLR can act as an effective incentive to\\ninsurers to provide benefits only if disputes are discouraged\\nand all benefit payments are included in the ratio, whether\\nthey go to providers in-network or out-of-network.\\n\\n This is best illustrated by looking to the history of the\\nACA and how its key provisions relevant to this dispute are\\n\\f8 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\ndesigned to operate. We therefore turn first to the statute\\nitself.\\n\\n STATUTORY BACKGROUND\\n\\n The ACA was enacted in 2010 after “a long history of\\nfailed health insurance reform.” See King, 135 S. Ct. at 2485.\\nPast experiences in many states had led to what the Supreme\\nCourt termed “an economic ‘death spiral.’” Id. at 2486. As\\nthe price of premiums increased, the number of people\\npurchasing health insurance dropped dramatically. Id.\\n\\n The outcome was very different in Massachusetts,\\nhowever, and Congress turned to it as a model. Id. Adopting\\nsome of the key reforms that made the Massachusetts\\ninsurance system successful, the ACA represented the most\\nsignificant regulatory overhaul and national expansion of\\nhealth care coverage since Medicare and Medicaid in 1965.\\nSee id.; see also Sebelius, 567 U.S. at 538–39, 541. The ACA\\nwas signed into law in 2010, and many of its private health\\ninsurance provisions went into effect in 2014. See, e.g., 42\\nU.S.C. §§ 18022(c)(1), 18041(c)(1); see King, 135 S. Ct. at\\n2487; Sebelius, 567 U.S. at 538. As the Supreme Court\\nemphasized, its purpose is to provide affordable, quality\\nhealth insurance for all Americans and decrease the cost of\\nhealth care. Sebelius, 567 U.S. at 538.\\n\\n Composed of ten titles spanning over 900 pages and\\nhundreds of provisions, the ACA brought sweeping reforms\\nto our health care system. Id. at 538–39. Under the ACA,\\nhealth insurance companies must accept every individual who\\napplies for coverage. 42 U.S.C. § 300gg-1(a). These insurers\\nare prohibited from charging higher premiums based on the\\ncondition of an applicant’s health. See id. § 300gg(a).\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 9\\n\\n To facilitate the purchasing of affordable health plans, the\\nACA created government-regulated health insurance\\n“exchanges.” Id. § 18031(b)(1). Exchanges are essentially\\nonline marketplaces where people can shop for insurance.\\nKing, 135 S. Ct. at 2487. The ACA directs each state to\\nestablish an exchange. 42 U.S.C. § 18031(b)(1). If a state\\ndecides not to establish an exchange, the ACA provides that\\nthe exchange shall be established and operated by the\\nSecretary of Health and Human Services. Id. § 18041(c)(1).\\nFurther, the federal government has published a website\\n(http://www.healthcare.gov) that explains the ACA to\\nindividual health insurance consumers and provides\\ninstructions on how to obtain coverage.\\n\\n While policymakers continue to debate the ACA’s future,\\nthere is research demonstrating that coverage expansion has\\nparticularly benefitted two historically vulnerable populations\\n— low-income adults and those with chronic conditions.\\nBenjamin D. Sommers, Bethany Maylone, Robert J. Blendon,\\nE. John Orav, and Arnold M. Epstein, Three-Year Impacts of\\nthe Affordable Care Act: Improved Medical Care and Health\\nAmong Low-Income Adults, 36 Health Affairs 1119 (2017).\\nAccording to at least one report, the number of uninsured\\nAmericans under the age of 65 has decreased from 44 million\\nin 2013 to less than 27 million at the end of 2016. Key Facts\\nAbout the Uninsured Population, Henry J Kaiser Family\\nFoundation (Dec. 7, 2018), https://www.kff.org/uninsured/\\nfact-sheet/key-facts-about-the-uninsured-population/.\\n\\n As we have seen, this appeal concerns an important\\nprovision of the ACA commonly referred to as the Medical\\nLoss Ratio or “MLR.” 42 U.S.C. § 300gg-18(a). Although\\nthe basic formula for calculating the MLR ratio is defined by\\nthe ACA and federal regulations, the parties here dispute how\\n\\f10 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\nthe ratio should be calculated in this case. The Centers for\\nMedicare & Medicaid Services (CMS), a division within\\nHHS, promulgated the standards for MLR calculation and\\nreporting. 45 C.F.R. § 158.120. CMS requires a health\\ninsurer to submit, for each state in which it provides\\ncoverage, data on the aggregate premiums, claims experience,\\nquality improvement expenditures, and non-claims costs it\\nincurs in connection with the policies it issues. See id.\\n§§ 158.120, 158.150, 150.160. The MLR report must detail\\nthe issuer’s spending on medical benefits, including clinical\\nservices and activities that improve health care, compared to\\nthe revenue the issuer receives from enrollees. 42 U.S.C.\\n§ 300gg-18(a). The required reporting does not distinguish\\nbetween in-network and out-of-network providers.\\n\\n The applicable regulations explain the MLR is to act as an\\nincentive to provide medical services and reduce\\nadministrative costs. 45 C.F.R. § 158.140(b)(3)(iii). The\\nMLR is thus intended to further the ACA’s goal of decreasing\\nhealth care costs by providing greater transparency on how\\nconsumers’ premium dollars are used and incentivizing\\nissuers to maximize spending on health care and activities\\nthat improve health care quality, thereby promoting greater\\nefficiency in health insurance markets. Medical Loss Ratio\\nRequirements Under the Patient Protection and Affordable\\nCare Act, 77 Fed. Reg. 28,790, 28,791, 28,793 (May 16,\\n2012). HHS has further explained that the MLR is intended\\nto ensure that “policyholders receive value for their premium\\ndollars” because the higher the MLR, the more value a\\npolicyholder receives for each premium dollar. Health\\nInsurance Issuers Implementing Medical Loss Ratio (MLR)\\nRequirements Under the Patient Protection and Affordable\\nCare Act, 75 Fed. Reg. 74864-01, 74865 (Dec. 1, 2010). The\\npurpose of the ACA, as demonstrated by the content of its\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 11\\n\\nprovisions and the implementing regulations, as well as its\\nhistory, is to broaden access to health care. This strongly\\nindicates that the ACA should be interpreted and applied in\\na manner that encourages insurers to spend premium revenue\\non medical services.\\n\\n FACTUAL BACKGROUND\\n\\n This dispute began when consumers purchased Blue\\nShield insurance plans through the State of California’s health\\ninsurance exchange. The ACA established health insurance\\nexchanges in all 50 states. The exchanges are regulated,\\nonline marketplaces administered by either the federal or\\nstate government, where individuals and small businesses can\\npurchase affordable, private health insurance plans. These\\nexchanges commenced operation on October 1, 2013, with\\nhealth insurance coverage beginning on January 1, 2014.\\n\\n To implement the ACA’s requirements, California created\\na state-run health exchange, known as Covered California. It\\nselected Blue Shield as one of the insurance companies to\\noffer health care plans on the Covered California\\nmarketplace. Individuals who enrolled in a Blue Shield plan\\nthrough Covered California were able to look up physicians\\naccepting Covered California patients in an online directory\\nposted on Blue Shield’s website. Blue Shield Covered\\nCalifornia enrollees could also select a doctor by calling Blue\\nShield’s customer service representatives, who in turn shared\\nparticipating provider information from the directory.\\n\\n Shortly after the online Covered California marketplace\\nwent live in late 2013, the California Department of Managed\\nHeath Care (“DMHC”), the state regulatory body that\\ngoverns managed health care plans, began receiving\\n\\f12 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\ncomplaints from Covered California consumers who had\\nsigned up for a Blue Shield plan. These enrollees reported\\ndifficulty finding in-network physicians. In response, DMHC\\nconducted an informal telephone survey of randomly selected\\nproviders that Blue Shield had identified as contracted\\nphysicians for the Blue Shield Covered California plans.\\nDuring the course of this survey, several offices informed\\nDMHC that they were not in fact accepting Covered\\nCalifornia enrollees.\\n\\n DMHC then began a formal survey of the entire Blue\\nShield provider directory. On November 17, 2014, DHMC\\nissued its final report which revealed that approximately nine\\npercent of the physicians listed in the Blue Shield online\\nprovider directory were not accepting Covered California\\npatients. Blue Shield acknowledged it had mistakenly listed\\nout-of-network providers in its directory, and also\\nacknowledged the confusion it caused among enrollees and\\nproviders. To resolve the problem, Blue Shield agreed to\\nupdate and correct the provider list and to reprocess\\nout-of-network claims at in-network rates. In October 2015,\\nDMHC and Blue Shield executed a settlement agreement\\ndescribing the provider directory mistake and outlining Blue\\nShield’s obligations to remedy it.\\n\\n Under Exhibit A of the settlement agreement, entitled\\n“Corrective Action,” Blue Shield agreed to adjust claims for\\nenrollees who had paid out-of-network rates for services of\\nproviders mistakenly listed as in-network. They were defined\\nas enrollees who “paid a provider in excess of the amount\\nthey would have paid if the claim had been processed at the\\nparticipating provider benefit level as a result of inaccuracies\\nin its [p]rovider [d]irectory during the 2014 and 2015 benefit\\nyears.” To be reimbursed, enrollees were required to attest\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 13\\n\\nthat they received “covered services.” Paragraph six of\\nExhibit A describes the process that required enrollees to\\nidentify the specific services they received from a provider\\noutside the network:\\n\\n The [Blue Shield] notice shall advise\\n recipients that if they received covered\\n services from a provider outside of their\\n plan’s network as a result of misinformation\\n from Blue Shield regarding the provider’s\\n participation status, and paid the provider out\\n of pocket for care, they may submit a written\\n claims submission form to a designated\\n address within 30 days to request reprocessing\\n of the claim. The claims submission form\\n shall require members to attest: a. that they\\n received covered services from a specific\\n provider who was represented in Blue\\n Shield’s Provider Directory or by Blue Shield\\n as participating in the network. . . .\\n\\nEnrollees were also required to list the date of their visit, and\\nconfirm that they paid the provider for the services at issue or\\nthat the provider was actively seeking payment. The\\nsubmission form included an explanation of benefits or claim\\nnumber so that Blue Shield could identify and reprocess the\\ndisputed claim. If an enrollee submitted all of the required\\ninformation identified in paragraph six of Exhibit A, and Blue\\nShield confirmed that the provider directory mistake led to\\nthe claim or alternatively that the attestation was acceptable,\\nBlue Shield was required to reprocess the claim at the\\nparticipating provider benefit level. Blue Shield paid over\\n$38 million in claims adjustments to enrollees who had\\nreceived clinical services from providers that did not accept\\n\\f14 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\nCovered California patients. Blue Shield included the claims\\nadjustments in the MLR report for 2014, but nevertheless fell\\nshort of the 80% requirement and paid rebates to enrollees.\\nPlaintiffs in this case are Blue Shield enrollees who received\\na MLR rebate for 2014, but now claim they are entitled to a\\nlarger rebate because Blue Shield improperly included in the\\nMLR the claim adjustments paid out under the settlement\\nagreement.\\n\\n PROCEDURAL HISTORY\\n\\n This case began in state court, because Plaintiffs claimed\\nthe alleged MLR miscalculation amounted to unfair\\ncompetition and unjust enrichment under California law. On\\nJuly 1, 2016, Plaintiffs filed a class action in Los Angeles\\nCounty Superior Court, alleging Blue Shield improperly\\nincluded the settlement payments as “incurred claims” in the\\nMLR numerator, so that enrollees did not receive the rebate\\nthey should have received. Blue Shield removed the case to\\nthe Central District of California and moved to dismiss the\\ncomplaint. Rather than opposing the motion, Plaintiffs filed\\na First Amended Complaint (“FAC”). The FAC alleged the\\nclaim adjustments should not have been included in the MLR\\nbecause they were restitution payments pursuant to the\\nsettlement agreement for payments to out-of-network\\nproviders and not to providers covered by the policy.\\n\\n Blue Shield moved to dismiss Plaintiffs’ FAC, arguing\\nthe restitution payments fall within the definition of “incurred\\nclaims” because they were for covered services, and that it\\ndid not matter for purposes of the MLR whether the providers\\nwere in or out-of-network. Plaintiffs opposed the motion,\\nasserting that because the providers were not in-network, the\\nservices were not covered under the policy, and therefore the\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 15\\n\\npayments did not fall within the definition of “incurred\\nclaims” or payments for services “covered by the policy.” 45\\nC.F.R. § 158.140(a). The district court granted Blue Shield’s\\nmotion to dismiss without leave to amend. It concluded that\\nbecause the settlement agreement provided reimbursement\\nfor services that were covered by the policy, these payments\\nwere properly included in the MLR numerator. The district\\ncourt explained, “[a]lthough the providers were not covered\\nby the policies, their services were.” This appeal followed.\\n\\n DISCUSSION\\n\\n The dispositive issue in this appeal is whether the MLR\\nwas properly calculated under federal law by including the\\nsettlement reimbursements for medical services by non-\\nnetwork providers. If there was no violation of the applicable\\nfederal law, we need not decide whether such an alleged\\nviolation could constitute a violation of California consumer\\nprotection laws.\\n\\n Plaintiffs’ position is that the reimbursements do not\\ncount because they were for services from out-of-network\\nproviders. We look first to the provisions of the statute to see\\nif it supports such a distinction. It does not. The statute\\nrequires annual reporting of the ratio of claims\\nreimbursements to premium revenue. 42 U.S.C. § 300gg-\\n18(a). Reimbursements for clinical services must be included\\nin the MLR numerator as “incurred claims.” Id. § 300gg-\\n18(a)(1). In relevant part it provides:\\n\\n (a) Clear accounting for costs\\n\\n A health insurance issuer offering group or\\n individual health insurance coverage\\n\\f16 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\n (including a grandfathered health plan) shall,\\n with respect to each plan year, submit to the\\n Secretary a report concerning the ratio of the\\n incurred loss (or incurred claims) plus the loss\\n adjustment expense (or change in contract\\n reserves) to earned premiums. Such report\\n shall include the percentage of total premium\\n revenue, after accounting for collections or\\n receipts for risk adjustment and risk corridors\\n and payments of reinsurance, that such\\n coverage expends--\\n\\n (1) on reimbursement for clinical services\\n provided to enrollees under such coverage;\\n\\n (2) for activities that improve health care\\n quality; and\\n\\n (3) on all other non-claims costs, including an\\n explanation of the nature of such costs, and\\n excluding Federal and State taxes and\\n licensing or regulatory fees. . . .\\n\\nId. The critical statutory focus is on health insurers providing\\ncoverage for health services. The statute measures total\\nexpenditures on health services. See id. § 300gg-\\n18(a)(1)–(3). For purposes of the statutory provisions, which\\nthe parties agree are controlling, the relative amounts required\\nto be paid to in-network as opposed to out-of-network\\nproviders is irrelevant. Only the total matters.\\n\\n Plaintiffs point out that the numerator’s reimbursement\\nfor clinical services under the statute has some limitation, in\\nthat to be included in the numerator, the services must fall\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 17\\n\\nwithin the policy’s coverage. The statute refers to\\nreimbursement “for clinical services . . . under such\\ncoverage.” Id. § 300gg-18(a)(1). This limitation, however,\\nis on the nature of the services that qualify for inclusion, not\\non any characteristic or qualification of the provider.\\n\\n Plaintiffs therefore go on to focus their argument on the\\ntext of the applicable regulations and specifically the\\ndefinition of “incurred claims” to be included in the\\nnumerator. These must include payments to providers\\n“whose services are covered by the policy.” 45 C.F.R.\\n§ 158.140(a).\\n\\n (a) General requirements. The report required\\n in § 158.110 must include direct claims paid\\n to or received by providers, including under\\n capitation contracts with physicians, whose\\n services are covered by the policy for clinical\\n services or supplies covered by the policy. In\\n addition, the report must include claim\\n reserves associated with claims incurred\\n during the MLR reporting year, the change in\\n contract reserves, reserves for contingent\\n benefits and the medical claim portion of\\n lawsuits, and any incurred experience rating\\n refunds. Reimbursement for clinical services,\\n as defined in this section, is referred to as\\n “incurred claims.” All components of and\\n adjustments to incurred claims, with the\\n exception of contract reserves, must be\\n calculated based on claims incurred only\\n during the MLR reporting year and paid\\n through March 31st of the following year.\\n\\f18 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\n Contract reserves must be calculated as of\\n December 31st of the applicable year.\\n\\nId. Plaintiffs contend that the condition that the services be\\ncovered by the policy means that the providers must be in-\\nnetwork. Yet the regulation does not say that. Like the\\nstatute, the regulation makes no distinction between in-\\nnetwork and out-of-network providers.\\n\\n Nor does the settlement agreement assist Plaintiffs. It is\\nintended to compensate enrollees who paid out-of-network\\nrates because Blue Shield mistakenly listed some providers as\\nbeing in-network. It does no more than establish that Blue\\nShield agreed to provide the compensation in satisfaction of\\nclaims, and to that extent, it supports Blue Shield’s position\\nthat such amounts should be included in the MLR numerator.\\nThe settlement agreement does not expressly refer to the\\nMLR and in no way suggests that the settlement payments\\nshould be excluded from the MLR numerator.\\n\\n We do not minimize the differences in the way various\\nplans treat services performed by in-network as opposed to\\nout-of-network providers. Whether and to what extent a\\nprovider’s services are covered depends on the type of\\ninsurance plan. A government website is helpful in\\nexplaining the differences. Health Insurance Plan &\\nNetwork Types: HMOs, PPOs, and More, Healthcare.gov,\\nhttps://www.healthcare.gov/choose-a-plan/plan-types/ (last\\nvisited February 4, 2019). The insurance universe can be\\nbewildering. For the purposes of this case, however, none of\\nthe relevant documents or provisions of federal law contain\\nany language that supports Plaintiffs’ contention that the\\nMLR numerator must exclude payments to out-of-network\\nproviders.\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 19\\n\\n Nor would there be any point in straining the language in\\nthe manner Plaintiffs suggest in order to narrow the definition\\nof claims that should be included. This is because the\\npurpose of the ACA was to broaden access to healthcare and\\nreduce costs. The MLR was enacted to further that purpose\\nby incentivizing insurers to pay for health services. The more\\nthe percentage of premium income paid out in benefits falls\\nbelow 80%, the greater the rebate the insurer must pay to the\\nenrollees.\\n\\n Congress passed the ACA to expand health insurance\\ncoverage. See King, 135 S. Ct. at 2485, 2493. The MLR 80\\npercent spending requirement on incurred claims furthers this\\ngoal by encouraging health insurance companies to pay for\\nclinical services. Should that spending requirement on\\nclinical services not be met, the MLR rebate serves the\\nACA’s twin goal of lowering health care costs by\\nreimbursing to consumers some of the premium money paid\\nto the health insurance company. See Sebelius, 567 U.S. at\\n538; 42 U.S.C. § 300gg-18. The 80 percent spending\\nrequirement on incurred claims and the rebate work together\\nto create an incentive for health insurance companies to\\nmaximize the amount of money spent on health care and\\nactivities that promote health care. See Medical Loss Ratio\\nRequirements Under the Patient Protection and Affordable\\nCare Act, 77 Fed. Reg. 28,790, 28,791, 28,793 (May 16,\\n2012).\\n\\n Thus, Congress was quite clearly concerned with\\nexpanding benefits and reducing costs. The MLR is\\ninclusively defined with respect to payouts and contains an\\nexpress exclusion for administrative costs. The structure is\\nmeant to encourage insurers to pay benefits and to discourage\\ndisputes. Plaintiffs ask us to turn the system on its head by\\n\\f20 MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE\\n\\nessentially contending that the goal should be to provide\\nrebates to enrollees rather than insurance benefits. Insureds\\npay premiums in order to be eligible for reimbursement of\\nhealth care costs, not for rebates. Plaintiffs seek to draw a\\ntype of distinction that would discourage the payment of\\nbenefits and encourage costly disputes like the one that is at\\nthe root of this case.\\n\\n For all of these reasons, we conclude there is no basis in\\nthe language, history, intent or spirit of the ACA to narrow\\nthe MLR by excluding payments for services rendered by\\nout-of-network physicians. As the district court correctly\\nrecognized, the services were covered by the plan and the\\npayments were made.\\n\\n The district court did not err in dismissing the action\\nwithout leave to amend because Plaintiffs never asked for\\nleave to amend the complaint. See Alaska v. United States,\\n201 F.3d 1154, 1163 (9th Cir. 2000). Plaintiffs belatedly ask\\nus to remand so that they may amend the complaint to claim\\nthat some of the reimbursements were for services outside the\\nplan. The request is untimely. See id. at 1163–64. It is also\\ninconsistent with the history of the underlying dispute.\\nPlaintiffs have never asserted that the services for which\\nreimbursement was made were not covered by the plan. This\\ndispute has always been over the increased costs of out-of-\\nnetwork providers rendering covered services, and that is the\\ndispute the settlement agreement resolved. Plaintiffs chose\\nto litigate how that resolution should be treated in the MLR,\\nand can not now go back and try to change the terms of the\\nsettlement itself. Those terms were clear: Blue Shield agreed\\nto reprocess claims for clinical services at in-network rates\\nonly if the enrollee submitted a written form attesting he or\\nshe received covered services from a provider who was\\n\\f MORRIS V. CALIFORNIA PHYSICIANS’ SERVICE 21\\n\\nerroneously listed in the Blue Shield provider directory.\\nEnrollees were not eligible for a claims adjustment if the\\nservices they received were not covered by their plan. Thus,\\nthe settlement agreement sought to remedy the provider\\ndirectory mistake as to the cost of covered services. There is\\nno valid reason to reopen it.\\n\\n The district court’s judgment of dismissal must be\\naffirmed.\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1296,"cells":{"id":{"kind":"number","value":4603412,"string":"4,603,412"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Marcelo Martinez-Cedillo V.william P. Barr"},"decision_date":{"kind":"string","value":"2019-03-18"},"docket_number":{"kind":"string","value":"14-71742"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"\", \"parties\": \"\", \"opinions\": [{\"author\": \"THOMAS, Chief Judge\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n MARCELO MARTINEZ-CEDILLO, No. 14-71742\\n AKA Marcelo Martinez,\\n Petitioner, Agency No.\\n A074-112-169\\n v.\\n\\n WILLIAM P. BARR, Attorney General, ORDER\\n Respondent.\\n\\n\\n\\n Filed March 18, 2019\\n\\n\\n ORDER\\n\\nTHOMAS, Chief Judge:\\n\\n Upon the vote of a majority of nonrecused active judges,\\nit is ordered that this case be reheard en banc pursuant to\\nFederal Rule of Appellate Procedure 35(a) and Circuit\\nRule 35-3. The three-judge panel disposition in this case\\nshall not be cited as precedent by or to any court of the Ninth\\nCircuit.\\n\\n Judge Miller did not participate in the deliberations or\\nvote in this case.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4380665/\"}]}"},"judge1":{"kind":"string","value":""},"judge2":{"kind":"string","value":""},"judge3":{"kind":"string","value":""},"exactly_three_caps":{"kind":"number","value":0,"string":"0"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4603412/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Marcelo Martinez-Cedillo V.william P. Barr\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-18\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n MARCELO MARTINEZ-CEDILLO, No. 14-71742\n AKA Marcelo Martinez,\n Petitioner, Agency No.\n A074-112-169\n v.\n\n WILLIAM P. BARR, Attorney General, ORDER\n Respondent.\n\n\n\n Filed March 18, 2019\n\n\n ORDER\n\nTHOMAS, Chief Judge:\n\n Upon the vote of a majority of nonrecused active judges,\nit is ordered that this case be reheard en banc pursuant to\nFederal Rule of Appellate Procedure 35(a) and Circuit\nRule 35-3. The three-judge panel disposition in this case\nshall not be cited as precedent by or to any court of the Ninth\nCircuit.\n\n Judge Miller did not participate in the deliberations or\nvote in this case.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Marcelo Martinez-Cedillo V.william P. Barr\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-18\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n MARCELO MARTINEZ-CEDILLO, No. 14-71742\\n AKA Marcelo Martinez,\\n Petitioner, Agency No.\\n A074-112-169\\n v.\\n\\n WILLIAM P. BARR, Attorney General, ORDER\\n Respondent.\\n\\n\\n\\n Filed March 18, 2019\\n\\n\\n ORDER\\n\\nTHOMAS, Chief Judge:\\n\\n Upon the vote of a majority of nonrecused active judges,\\nit is ordered that this case be reheard en banc pursuant to\\nFederal Rule of Appellate Procedure 35(a) and Circuit\\nRule 35-3. The three-judge panel disposition in this case\\nshall not be cited as precedent by or to any court of the Ninth\\nCircuit.\\n\\n Judge Miller did not participate in the deliberations or\\nvote in this case.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1297,"cells":{"id":{"kind":"number","value":4603413,"string":"4,603,413"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Ann Garcia v. Salvation Army"},"decision_date":{"kind":"string","value":"2019-03-18"},"docket_number":{"kind":"string","value":"16-16827"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Mary M. Schroeder and Paul J. Watford, Circuit Judges, and Edward R. Korman, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"KORMAN, District Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nANN GARCIA, No. 16-16827\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:14-cv-02225-DGC\\n\\nSALVATION ARMY, OPINION\\n Defendant-Appellee.\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n David G. Campbell, District Judge, Presiding\\n\\n Argued and Submitted November 13, 2018\\n San Francisco, California\\n\\n Filed March 18, 2019\\n\\n Before: Mary M. Schroeder and Paul J. Watford, Circuit\\n Judges, and Edward R. Korman, * District Judge.\\n\\n Opinion by Judge Korman\\n\\n\\n\\n\\n *\\n The Honorable Edward R. Korman, United States District Judge\\nfor the Eastern District of New York, sitting by designation.\\n\\f2 GARCIA V. SALVATION ARMY\\n\\n SUMMARY **\\n\\n\\n Employment Discrimination\\n\\n The panel affirmed the district court’s summary\\njudgment in favor of the Salvation Army, the defendant in\\nan employment discrimination action under Title VII and the\\nAmericans with Disabilities Act.\\n\\n The panel held that Title VII’s religious organization\\nexemption is not jurisdictional and is subject to procedural\\nforfeiture. Absent prejudice resulting from the Salvation\\nArmy’s failure to timely raise the defense, however, the\\nreligious organization exemption foreclosed plaintiff’s Title\\nVII claims because the Salvation Army’s purpose and\\ncharacter were primarily religious. The panel held that the\\nexemption does not apply only to hiring and firing decisions,\\nbut rather extends to both retaliation and hostile work\\nenvironment claims.\\n\\n Affirming the district court’s grant of summary\\njudgment on plaintiff’s ADA claim, the panel held that there\\nwas no triable issue whether the Salvation Army failed to\\nengage in an interactive process in good faith with the\\nplaintiff up to the time she was cleared for work after a\\nperiod of leave. After the clearance for work, the plaintiff\\ncould not show that she was disabled.\\n\\n\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f GARCIA V. SALVATION ARMY 3\\n\\n COUNSEL\\n\\nMenno Goedman (argued), Boies Schiller Flexner LLP,\\nWashington, D.C.; Kathleen Hartnett, Boies Schiller Flexner\\nLLP, Palo Alto, California; for Plaintiff-Appellant.\\n\\nR. Shawn Oller (argued) and Ryan G. Lockner, Littler\\nMendelson, P.C., Phoenix, Arizona; for Defendant-\\nAppellee.\\n\\n\\n OPINION\\n\\nKORMAN, District Judge:\\n\\n The Salvation Army is an evangelical ministry founded\\nin 1865 by William Booth, a former Methodist minister. 1\\nThe Salvation Army’s religious tenets differed from\\ntraditional Methodism in rejecting the importance of\\nsacraments and emphasizing strong central governance. 2 To\\nthat end, Booth—“General” of the Salvation Army—\\n\\n\\n\\n\\n 1\\n Unless otherwise indicated, the following historical context is\\ngleaned from the record. Otherwise—and purely for background\\npurposes—we take judicial notice of certain historical facts that “can be\\naccurately and readily determined from sources whose accuracy cannot\\nreasonably be questioned.” Fed. R. Evid. 201(b); see Singh v. Ashcroft,\\n393 F.3d 903, 905 (9th Cir. 2004).\\n\\n 2\\n Protestantism, ENCYCLOPÆDIA BRITANNICA,\\nhttps://www.britannica.com/topic/Protestantism (published Dec. 13,\\n2018).\\n\\f4 GARCIA V. SALVATION ARMY\\n\\nadopted the military-style hierarchy of the British Army3\\nunder which ranked officers were the equivalent of\\nministers. 4 In keeping with Protestantism’s nineteenth\\ncentury “camp revival,” Booth took his ministry to the\\nstreets 5 and began establishing mission centers catering to\\nLondon’s poor. 6\\n\\n What started as a single ministry in the East End of\\nLondon spread to the shores of the United States in 1880 7\\nand now operates in more than 80 countries through 16,000\\nevangelical centers and 3,000 social welfare institutions\\nworldwide. 8 The Salvation Army describes itself as “an\\nevangelical part of the universal Christian church,” whose\\nprofessed mission is “to preach the gospel of Jesus Christ\\nand to meet human needs in His name without\\ndiscrimination.” Here in the United States, the Salvation\\n\\n\\n\\n 3\\n William Booth, ENCYCLOPÆDIA BRITANNICA,\\nhttps://www.britannica.com/biography/William-Booth (published Aug.\\n16, 2018).\\n 4\\n Salvation Army, ENCYCLOPÆDIA BRITANNICA,\\nhttps://www.britannica.com/topic/Salvation-Army (published May 11,\\n2018).\\n\\n 5\\n Protestantism, supra note 2.\\n\\n 6\\n Salvation Army, supra note 4.\\n\\n 7\\n William Booth, supra note 3.\\n\\n 8\\n Salvation Army, supra note 4.\\n\\f GARCIA V. SALVATION ARMY 5\\n\\nArmy operates through 501(c)(3) nonprofit corporations. 9 In\\n2012 and 2013, direct public donations made up the lion’s\\nshare of the Salvation Army’s total revenue; sales to the\\npublic comprised fifteen percent.\\n\\n Ann Garcia’s relationship with the Salvation Army dates\\nto 1999, when she began attending religious services at the\\nEstrella Mountain Corps in Avondale, Arizona. In 2002, the\\nCorps hired Garcia to work as an assistant to the pastor, a\\nposition she held until July 2010, when Arlene and Dionisio\\nTorres became the new pastors. No longer in need of an\\nassistant, Arlene Torres reassigned Garcia to the position of\\nsocial services coordinator in January 2011. In that role,\\nGarcia aided clients under the supervision of Arlene Torres.\\nIn late 2011, Garcia and her husband “left the Church” and\\nstopped attending the Salvation Army’s religious services,\\nbut Garcia continued her work as social services coordinator.\\nAfterward, her relationship with Torres began to deteriorate.\\n\\n Tensions reached new heights in July 2013, when a client\\nfiled a lengthy complaint letter against Garcia, claiming that\\nshe “refused to provide help to [the client’s] family.” After\\nTorres informed Garcia that a complaint had been lodged,\\nGarcia demanded to see it. Torres refused, claiming that the\\ncomplaint was confidential. Three days later, Garcia filed an\\ninternal grievance of her own against Torres, claiming that\\nshe “fe[lt] discriminated against and excluded and isolated”\\n\\n 9\\n We take judicial notice of the Salvation Army’s nonprofit status,\\nas reflected in the publicly available IRS determination letters at Docket\\nEntry No. 45. See Fed. R. Evid. 201(b); Anderson v. Holder, 673 F.3d\\n1089, 1094 n.1 (9th Cir. 2012) (“We may take judicial notice of records\\nand reports of administrative bodies.” (internal quotation marks\\nomitted)).\\n\\f6 GARCIA V. SALVATION ARMY\\n\\nat work ever since leaving the church. The specter of the\\nundisclosed client grievance continued to disturb Garcia.\\nShe would go on to submit complaints to the EEOC and\\nArizona state authorities for religious discrimination and\\nretaliation.\\n\\n Following a lengthy period of medical leave due to\\nfibromyalgia, the Salvation Army fired Garcia after she\\nfailed to report to work despite being cleared by her doctor.\\nGarcia then filed a second complaint with the EEOC and\\nstate authorities alleging that, by declining to disclose the\\nclient complaint, the Salvation Army failed to accommodate\\nher disability.\\n\\n Garcia’s EEOC charges were dismissed, and right-to-sue\\nletters issued. Garcia subsequently brought two lawsuits\\nagainst the Salvation Army: one under Title VII of the Civil\\nRights Act of 1964 (Title VII), 42 U.S.C. § 2000e, et seq.,\\nand another under the Americans with Disabilities Act\\n(ADA), 42 U.S.C. § 12112, et seq., which were\\nconsolidated. In sum, Garcia alleged that the Salvation Army\\nsubjected her to a hostile work environment because she\\nstopped attending religious services and retaliated against\\nher for filing an internal grievance complaining of religion-\\nbased mistreatment. The resulting stress precipitated health\\nproblems that the Salvation Army failed to accommodate.\\n\\n The district judge (Campbell, J.) granted summary\\njudgment to the Salvation Army, holding that Title VII’s\\nreligious organization exemption (ROE) protects the\\nSalvation Army from suit, even if it failed to timely assert\\nthe defense. Garcia v. Salvation Army, 2016 WL 4732845,\\nat *4 (D. Ariz. Sept. 12, 2016). He reasoned that the ROE is\\n\\f GARCIA V. SALVATION ARMY 7\\n\\njurisdictional—a matter of courts’ Article III power to hear\\ncases and controversies—and cannot be forfeited. Id. The\\ndistrict judge also dismissed Garcia’s ADA claims on the\\nmerits. Id. at *5–6.\\n\\n Garcia appeals, raising two legal questions regarding the\\napplication and scope of the ROE. First, whether the ROE is\\njurisdictional, depriving federal courts of subject matter\\njurisdiction when invoked. And second, whether the ROE\\nextends beyond hiring and firing decisions to hostile work\\nenvironment and retaliation claims. She also challenges the\\ndistrict judge’s dismissal of her ADA claims (to which the\\nROE does not apply). We first address the application of the\\nROE before turning to the merits of the ADA claim.\\n\\n DISCUSSION\\n\\nTitle VII Claims\\n\\n A. The ROE Applies to the Salvation Army\\n\\n The ROE provides that Title VII’s protections against\\ndiscrimination\\n\\n shall not apply to an employer with respect\\n to . . . a religious corporation, association,\\n educational institution, or society with\\n respect to the employment of individuals of a\\n particular religion to perform work connected\\n with the carrying on by such corporation,\\n association, educational institution, or\\n society of its activities.\\n\\f8 GARCIA V. SALVATION ARMY\\n\\n42 U.S.C. § 2000e-1(a). The entity seeking the benefit of the\\nstatute bears the burden of proving it is exempt. EEOC v.\\nKamehameha Schs./Bishop Estate, 990 F.2d 458, 460 (9th\\nCir. 1993).\\n\\n In applying the ROE, we determine whether an\\ninstitution’s “purpose and character are primarily religious”\\nby weighing “[a]ll significant religious and secular\\ncharacteristics.” EEOC v. Townley Eng’g & Mfg. Co., 859\\nF.2d 610, 618 (9th Cir. 1988). It does not suffice that an\\ninstitution be “merely ‘affiliated’ with a religious\\norganization.” Id. at 617. Although we construe the ROE\\nnarrowly, often the organization seeking the exemption is\\n“clearly” religious. Id. at 618.\\n\\n This is such a case. See, e.g., Rev. Rul. 59-129, 1959-1\\nC.B. 58 (noting that the Salvation Army is a “church” under\\nthe Internal Revenue Code); Schleicher v. Salvation Army,\\n518 F.3d 472, 478 (7th Cir. 2008) (“The Salvation Army,\\nwhich has existed in the United States since 1880, is\\nacknowledged to be a completely legitimate church . . . .”);\\nMcClure v. Salvation Army, 460 F.2d 553, 554 (5th Cir.\\n1972) (“The Salvation Army is a church . . . .”). The\\nSalvation Army holds regular religious services. It offers\\nsocial services to customers regardless of their religion “to\\nreach new populations and spread the gospel.” Indeed, the\\nSalvation Army’s mission statement describes it as\\n\\n an evangelical part of the universal Christian\\n church. Its message is based on the Bible. Its\\n ministry is motivated by the love of God. Its\\n mission is to preach the gospel of Jesus Christ\\n\\f GARCIA V. SALVATION ARMY 9\\n\\n and to meet human needs in His name\\n without discrimination. 10\\n\\nThe record establishes that, as a church, the Salvation Army\\nis “clearly” religious. See Townley, 859 F.2d at 618 (“[T]he\\ncentral function of [the ROE] has been to exempt churches,\\nsynagogues, and the like, and organizations closely affiliated\\nwith those entities.”). Accord Lown v. Salvation Army, Inc.,\\n393 F. Supp. 2d 223, 246–54 (S.D.N.Y. 2005) (applying\\nROE to the Salvation Army); Clark v. Salvation Army, LLC,\\n2008 WL 11375384, at *1–2 (N.D. Ala. June 16, 2008)\\n(same).\\n\\n Garcia argues that the Salvation Army does not qualify\\nfor the ROE because it does not satisfy the fourth factor of a\\nfour-part test discussed in Spencer v. World Vision, Inc., 633\\nF.3d 723, 724 (9th Cir. 2011) (per curiam), which asks\\nwhether an entity “engage[s] primarily or substantially in the\\nexchange of goods or services for money beyond nominal\\namounts.” Garcia contends this factor is not satisfied\\nbecause the Salvation Army generates a large-dollar amount\\nof sales revenue, even though that amount constitutes a small\\nportion (fifteen percent) of its total income. Moreover, in\\n2012, 82 cents of every dollar spent by the Salvation Army\\nwent toward its program services. These considerations\\naside, the concurring opinions of Judges O’Scannlain and\\nKleinfeld, who wrote separately on how to evaluate the\\nfourth Spencer factor, support the Salvation Army’s\\n\\n 10\\n The Salvation Army’s mission statement is consistent with the\\nSupreme Court’s various articulations of what constitutes a “religion.”\\nSee generally Note, Ari J. Diaconis, The Religion of Alcoholics\\nAnonymous (AA): Applying the Clergy Privilege to Certain AA\\nCommunications, 99 Cornell L. Rev. 1185, 1213–17 (2014).\\n\\f10 GARCIA V. SALVATION ARMY\\n\\nentitlement to ROE protection. Under Judge O’Scannlain’s\\napproach, the ROE applies if the first three factors identified\\nin the per curiam opinion are satisfied and the organization\\nis a nonprofit. Spencer, 633 F.3d at 734 (O’Scannlain, J.,\\nconcurring). The Salvation Army easily satisfies that test.\\nWhile Judge Kleinfeld would alternatively ask how the\\norganization charges for its services, he cites the Salvation\\nArmy as a group that satisfies this criterion because it “gives\\nits homeless shelter and soup kitchen services away, or\\ncharges nominal fees.” Id. at 747 (Kleinfeld, J., concurring).\\n\\n B. The ROE Reaches Claims for Retaliation and Hostile\\n Work Environment\\n\\n Even assuming the ROE applies to the Salvation Army,\\nGarcia argues that it does not reach her claims for retaliation\\nand hostile work environment. In her view, the ROE applies\\nonly to hiring and firing decisions. Although we have not\\naddressed this question, other courts have held that the ROE\\nextends to both retaliation and hostile work environment\\nclaims. See, e.g., Kennedy v. St. Joseph’s Ministries, Inc.,\\n657 F.3d 189, 192–94 (4th Cir. 2011); Saeemodarae v.\\nMercy Health Servs., 456 F. Supp. 2d 1021, 1040–41 (N.D.\\nIowa 2006); Lown, 393 F. Supp. 2d at 254; Hopkins v.\\nWomen’s Div., Gen. Bd. of Glob. Ministries, 238 F. Supp. 2d\\n174, 180 (D.D.C. 2002); Aguillard v. La. Coll., 341 F. Supp.\\n3d 642, 647–48, 652 (W.D. La. 2018); Clark, 2008 WL\\n11375384, at *1. We agree.\\n\\n First, the ROE’s text reaches beyond hiring and firing.\\nCongress “painted with a broader brush, exempting religious\\norganizations from the entire subchapter of Title VII with\\nrespect to the employment of persons of a particular\\n\\f GARCIA V. SALVATION ARMY 11\\n\\nreligion.” Kennedy, 657 F.3d at 194 (internal quotation\\nmarks omitted) (emphasis added). The “entire subchapter”\\nof Title VII includes protections against retaliation and\\ndiscriminatory harassment amounting to a hostile work\\nenvironment. 11 See 42 U.S.C. §§ 2000e-2(a)(1), 2000e-3(a).\\nAnd “employment” encompasses “the breadth of the\\nrelationship between the employer and employee,” not just\\nhiring and firing. Kennedy, 657 F.3d at 193 (deriving term’s\\ndefinition from settled common-law meaning); see\\nNationwide Mut. Ins. Co. v. Darden, 503 U.S. 318, 323–24\\n(1992) (setting out common-law factors for evaluating an\\nemployment relationship, including all aspects of\\nemployer’s control over employee’s actions); cf. 20 C.F.R.\\n§ 404.1007(a) (“In general, you are a common-law\\nemployee if the person you work for may tell you what to do\\nand how, when, and where to do it.”).\\n\\n Limiting “employment” to hiring and firing decisions is\\nalso inconsistent with the term’s use throughout Title VII.\\nFor example, Section 2000e-2(a)(1) defines “unlawful\\n\\n 11\\n Garcia relies on EEOC v. Pacific Press Publishing Association,\\n676 F.2d 1272 (9th Cir. 1982), abrogated on other grounds as\\nrecognized by Am. Friends Serv. Comm. Corp. v. Thornburg, 951 F.2d\\n957, 960 (9th Cir. 1991), for the proposition that Congress did not intend\\nthe ROE to cover retaliation claims. Pacific Press, however, dealt with\\nsex-based Title VII claims against a religious publisher. Id. at 1274. The\\nplaintiff there claimed that her employer paid her less than her male\\ncoworkers and retaliated against her when she filed charges. Id. The\\nemployer sought to avoid liability on the grounds that the employee’s act\\nof filing a lawsuit violated its religious tenets. See id. at 1275. There, we\\nheld that religious employers facing sex-, race-, and national origin-\\nbased claims are not immune from liability for “retaliatory actions\\nagainst employees who exercise their rights under the statute.” See id. at\\n1276. We did not hold that religion-based retaliation claims are beyond\\nthe reach of the ROE. Id.\\n\\f12 GARCIA V. SALVATION ARMY\\n\\nemployment practice” as “to fail or refuse to hire or\\ndischarge any individual, or otherwise to discriminate\\nagainst any individual with respect to his compensation,\\nterms, conditions, or privileges of employment.” There,\\n“‘employment’ . . . encompasses more than hiring or firing;\\nif the term were so limited, the second clause would be\\nsuperfluous.” Kennedy, 657 F.3d at 193. We agree that\\n“‘[e]mployment,’ as used throughout Title VII, simply\\ncovers a much broader understanding than mere hiring and\\nfiring.” Id. Indeed, Garcia conceded at oral argument that a\\nprotected organization may lawfully demote an employee\\nbased on religious preference. Accepting as much upends\\nher theory that the ROE covers only hiring-and-firing\\ndecisions. 12\\n\\n In a final effort to avoid the ROE’s reach, Garcia\\nmaintains that she has not pleaded a cause of action labeled\\n“religious discrimination.” Even so, the complaint centers\\naround religious discrimination. Specifically, the first\\nsentence of Count 1 (retaliation) alleges that the Salvation\\nArmy subjected Garcia to “discrimination based on\\nreligion.” The complaint then goes on to specify alleged\\n\\n 12\\n Garcia also notes that the 2000 edition of the EEOC Compliance\\nManual suggests that the ROE “only applies to hiring and discharge.”\\nU.S. Equal Emp’t Opportunity Comm’n, Compliance Manual § 2-\\nIII(B)(4)(b)(i) (issued May 12, 2000) (hereinafter “EEOC Compliance\\nManual”)). However, the foregoing analysis demonstrates that there is\\nno statutory ambiguity for agency guidance to resolve. See Chevron,\\nU.S.A., Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 842–43 (1984)\\n(“If the intent of Congress is clear, that is the end of the matter . . . .”). In\\nany event, “this interpretation, which is viewed under Skidmore\\ndeference, contains no attendant rationale, lacks the power to persuade,\\nand does not warrant deference.” Kennedy, 657 F.3d at 194 n.9.\\n\\f GARCIA V. SALVATION ARMY 13\\n\\nincidents of mistreatment and concludes that they all\\noccurred “after [Garcia] engaged in protected activity by\\nfiling an internal grievance from what [she] believe[s] was\\ndiscrimination based on religion.” 13 Count 2 (hostile work\\nenvironment) simply repleads all the allegations in Count 1.\\n\\n It is true, as Garcia notes, that Title VII treats workplace\\ndiscrimination and retaliation claims in separate provisions.\\nSee 42 U.S.C. §§ 2000e-2(a), 2000e-3(a). Yet both types of\\nclaims require a nexus to discrimination. See id. Because the\\nROE permits religious organizations to discriminate based\\non religion, retaliation claims based on religious\\ndiscrimination fail against protected organizations because\\nthe practice “opposed” is not “unlawful.” See id. § 2000e-\\n3(a). Likewise, hostile work environment claims must\\ninvolve status-based harassment, id. § 2000e-2(a)(1);\\nMeritor Sav. Bank, FSB v. Vinson, 477 U.S. 57, 63–64\\n(1986), and fail against protected organizations where the\\nalleged harassment is based on religion.\\n\\n C. The ROE Bars Garcia’s Claims\\n\\n The Salvation Army failed to raise the ROE as an\\naffirmative defense in its responsive pleading, which would\\nnormally result in forfeiture. 14 See Fed. R. Civ. P. 12(h)(1)–\\n\\n 13\\n Indeed, Garcia’s internal grievance took issue with her treatment\\n“ever since [she and her husband] left the church.” And Garcia’s EEOC\\ncharge—for which she checked the form’s “religious discrimination”\\nand “retaliation” boxes—states, “Ever[] since I left the Church my\\nimmediate supervisor . . . has been subjecting me to disparate treatment.”\\n 14\\n “[F]orfeiture is the failure to make the timely assertion of a right[;]\\nwaiver is the intentional relinquishment or abandonment of a known\\n\\f14 GARCIA V. SALVATION ARMY\\n\\n(2). The district judge, however, held that the ROE is\\njurisdictional, and as such, the defense cannot be\\nrelinquished. See Fed. R. Civ. P. 12(h)(3) (“If the court\\ndetermines at any time that it lacks subject-matter\\njurisdiction, the court must dismiss the action.” (emphasis\\nadded)).\\n\\n 1. The ROE is Nonjurisdictional and Subject to\\n Forfeiture\\n\\n In Arbaugh v. Y & H Corp., 546 U.S. 500, 516 (2006),\\nthe Supreme Court established the “bright line” rule that\\n“when Congress does not rank a statutory limitation on\\ncoverage as jurisdictional, courts should treat the restriction\\nas nonjurisdictional in character.” Three factors guide our\\nanalysis. See Leeson v. Transamerica Disability Income\\nPlan, 671 F.3d 969, 976–77 (9th Cir. 2012). First is whether\\nthe provision is “clearly labeled jurisdictional.” Id. at 976\\n(quotation marks omitted). Second is whether the provision\\nis “located in a jurisdiction-granting provision.” Id. at 976–\\n77. Third is whether some “other reasons necessitate[] that\\nthe provision be construed as jurisdictional.” Id. at 977.\\n\\n The ROE is not labeled jurisdictional. See 42 U.S.C.\\n§ 2000e-1(a). It does not use the term “jurisdiction” nor\\nspeak in jurisdictional terms about the power of United\\nStates courts to hear cases under Title VII. Compare id. with\\n28 U.S.C. §§ 1345, 1348, 1350. Rather, it appears in a\\nseparate provision from that establishing federal courts’\\njurisdiction over Title VII claims, which militates against\\n\\nright.” Hamer v. Neighborhood Hous. Servs. of Chi., 138 S. Ct. 13, 17\\nn.1 (2017) (quoting United States v. Olano, 507 U.S. 725, 733 (1993)).\\nThus, the case before us concerns a procedural forfeiture.\\n\\f GARCIA V. SALVATION ARMY 15\\n\\nclassifying it as jurisdictional. See 42 U.S.C. § 2000e-5(f)(3)\\n(conferring jurisdiction over Title VII claims to federal\\ndistrict courts). Put otherwise, the ROE limits entitlement to\\nrelief in a narrow class of cases, not “the authority of federal\\ncourts to adjudicate claims under [Title VII].” Leeson, 671\\nF.3d at 978. The district judge erred by ranking the ROE\\njurisdictional where Congress did not. See Arbaugh, 546\\nU.S. at 516.\\n\\n The Salvation Army maintains that a claim cannot\\npossibly arise under Title VII where Congress has mandated\\nTitle VII “shall not apply.” But that is precisely the\\nconsequence of Arbaugh. Title VII imposes no liability on\\nemployers with fewer than 15 employees. See 42 U.S.C.\\n§ 2000e(b). Yet the defendant in Arbaugh—an employer\\nwith fewer than 15 employees—forfeited this argument by\\nfailing to raise it until after trial. Arbaugh, 546 U.S. at 503–\\n04, 510, 516. The employer was held liable, even though\\nCongress did not intend Title VII to reach it. Id. at 516. Such\\nis the consequence of failing to raise a nonjurisdictional\\ndefense. See id. at 510–11.\\n\\n Our decision in United States v. Hui Hsiung, 778 F.3d\\n738 (9th Cir. 2015), provides a useful analog. There, we held\\nthat the Foreign Trade Antitrust Improvements Act\\n(FTAIA), 15 U.S.C. § 6a, which limits conduct covered by\\nthe Sherman Act, is nonjurisdictional. Hui Hsiung, 778 F.3d\\nat 752–53. Section 4 of the Sherman Act confers upon\\nfederal district courts jurisdiction “to prevent and restrain\\nviolations of sections 1 to 7 of this title.” 15 U.S.C. § 4. The\\nFTAIA, like the ROE, provides that “[s]ections 1 to 7 of this\\ntitle shall not apply,” unless certain conditions are satisfied.\\nId. § 6a (emphasis added). Nevertheless, we held that the\\n\\f16 GARCIA V. SALVATION ARMY\\n\\nimpact of the FTAIA on Sherman Act claims “is a merits\\nquestion, not a jurisdictional one.” Hui Hsiung, 778 F.3d at\\n752. As the Supreme Court has explained, “to ask what\\nconduct [a statute] reaches is to ask what conduct [the\\nstatute] prohibits, which is a merits question.” Morrison v.\\nNat’l Austl. Bank Ltd., 561 U.S. 247, 254 (2010). The ROE,\\nlike the FTAIA, “removes conduct from [Title VII’s] reach,”\\nbearing on the question of “what conduct [Title VII]\\nprohibits,” not the power of courts to hear a case. Hui\\nHsiung, 778 F.3d at 752 (internal quotation marks and\\nalterations omitted).\\n\\n Indeed, the Seventh Circuit has held that Title VII’s\\ndomestic-work requirement, embodied in Section 2000e-\\n1(a) alongside the ROE, is nonjurisdictional, as it “appears\\noutside of the statute’s jurisdictional provision” and “[t]here\\nis no other reason to believe that Congress intended to ‘rank’\\nthe restrictions as jurisdictional.” Rabe v. United Air Lines,\\nInc., 636 F.3d 866, 869 (7th Cir. 2011); see also Maguire v.\\nMarquette Univ., 814 F.2d 1213, 1216 (7th Cir. 1987)\\n(application of 42 U.S.C. § 2000e-2(e)(1), permitting\\neducational institutions “to hire and employ employees of a\\nparticular religion” under certain circumstances, is not a\\nquestion of subject matter jurisdiction). And the only court\\nto expressly address whether the ROE itself is jurisdictional\\nfound that it is not. See Smith v. Angel Food Ministries, Inc.,\\n2008 WL 5115037, at *4 (M.D. Ga. Dec. 4, 2008), on\\nreconsideration 611 F. Supp. 2d 1346 (M.D. Ga. 2009).\\n\\n We recognize that a number of courts have suggested\\nthat the ROE can never be waived or forfeited. See, e.g., Hall\\nv. Baptist Mem’l Health Care Corp., 215 F.3d 618, 625 (6th\\nCir. 2000); Ark Encounter, LLC v. Parkinson, 152 F. Supp.\\n\\f GARCIA V. SALVATION ARMY 17\\n\\n3d 880, 915 n.25 (E.D. Ky. 2016); Saeemodarae, 456 F.\\nSupp. 2d at 1038–39; Siegel v. Truett-McConnell Coll., Inc.,\\n13 F. Supp. 2d 1335, 1345 (N.D. Ga. 1994). These cases,\\nlike the district court’s opinion below, can be traced to the\\nThird Circuit’s decision in Little v. Wuerl, 929 F.2d 944 (3d\\nCir. 1991), where a Catholic school invoked the ROE after\\ndeclining to rehire a Protestant teacher. The teacher argued\\nthat the school “waived” the ROE, because it knew she was\\nProtestant when it hired her. Id. at 951. Little held that the\\nROE was not “a privilege or interest” that could be waived\\nbut rather “a decision by Congress that the government\\ninterest in eliminating religious discrimination by religious\\norganizations is outweighed by the rights of those\\norganizations to be free from government intervention,”\\nmeaning that “no act by Little or the Parish could expand the\\nstatute’s scope.” Id. Other courts subsequently concluded\\nthat the ROE “cannot be waived by either party.” E.g., Hall,\\n215 F.3d at 625.\\n\\n Little is inapposite because it concerned the validity of\\nan implied waiver defense based on an organization’s pre-\\nsuit conduct—not a procedural forfeiture. See Little, 929\\nF.2d at 951. A statutory defense is not unrelinquishable per\\nse simply because it embodies Congress’s intent to limit a\\nclass of claims. To the contrary, even where Congress has\\nmandated that a statute “shall not apply” under certain\\ncircumstances, we have routinely applied forfeiture\\nprinciples. See, e.g., Magana v. Commonwealth of the\\nNorthern Mariana Islands, 107 F.3d 1436, 1445 (9th Cir.\\n1997) (failure to plead FLSA exemption can result in\\nforfeiture); Brennan v. Valley Towing Co., Inc., 515 F.2d\\n100, 104 (9th Cir. 1975) (same). See generally 5 Charles\\nAlan Wright & Arthur R. Miller, Federal Practice and\\n\\f18 GARCIA V. SALVATION ARMY\\n\\nProcedure § 1271 (3d ed. 2004) (collecting cases). We see\\nno reason to treat the ROE differently.\\n\\n 2. Absent Prejudice, the ROE May Be First Raised\\n at Summary Judgment\\n\\n Although statutory exemptions are not among the\\naffirmative defenses enumerated in Rule 8(c), see Fed. R.\\nCiv. P. 8(c), we have explained that they must be raised in a\\nparty’s “initial responsive pleading.” See Magana, 107 F.3d\\nat 1445. Accordingly, the Salvation Army was required to\\nplead the ROE as an affirmative defense. Cf. Oden v.\\nOktibbeha County, 246 F.3d 458, 466–67 (5th Cir. 2001)\\n(holding that Title VII’s personal-staff exception must be\\npleaded as an affirmative defense). This makes sense given\\nthat the defendant “bear[s] the burden of proving [it is\\nexempt],” Kamehameha, 990 F.2d at 460, and the facts\\nnecessary to invoke the exemption are “outside of the\\nplaintiff’s prima facie case,” Wright & Miller, supra,\\n§ 1271.\\n\\n The Salvation Army did not raise the ROE in its answer\\nbut insists that it pleaded enough to put Garcia on notice.\\nFirst, it pleaded that Garcia failed to state a claim upon which\\nrelief can be granted, a defense which may be raised up and\\nuntil the close of trial. Fed. R. Civ. P. 12(h)(2). But simply\\nstating that the plaintiff failed to state a claim is insufficient\\nto provide notice of a specific affirmative defense. See\\nSimmons v. Navajo County, 609 F.3d 1011, 1023 (9th Cir.\\n2010) (“The key to determining the sufficiency of pleading\\nan affirmative defense is whether it gives plaintiff fair notice\\nof the defense.” (quoting Wyshak v. City Nat’l Bank, 607\\nF.2d 824, 827 (9th Cir. 1979))), abrogated in part by Castro\\n\\f GARCIA V. SALVATION ARMY 19\\n\\nv. County of Los Angeles, 833 F.3d 1060 (9th Cir. 2016) (en\\nbanc). And independent of Rule 12(b)(6), Rule 8(c) requires\\nthat “[i]n responding to a pleading, a party must\\naffirmatively state any avoidance or affirmative defense.”\\nFed. R. Civ. P. 8(c).\\n\\n The Salvation Army also pleaded “that any adverse\\nemployment actions taken against plaintiff were taken for\\nlawful, legitimate, non-retaliatory, and non-discriminatory\\nreasons.” On its face, this defense does not refer to the ROE\\nbut rather the standard for rebutting a prima facie claim of\\ndiscrimination on the merits. See Coghlan v. Am. Seafoods\\nCo. LLC, 413 F.3d 1090, 1094 (9th Cir. 2005). Thus, the\\nSalvation Army failed to timely raise the ROE.\\n\\n Even so, “[i]n the absence of a showing of prejudice . . .\\nan affirmative defense may be raised for the first time at\\nsummary judgment.” Camarillo v. McCarthy, 998 F.2d 638,\\n639 (9th Cir. 1993). There is no prejudice to a plaintiff where\\nan “affirmative defense would have been dispositive” if\\nasserted “when the action was filed.” Owens v. Kaiser\\nFound. Health Plan, Inc., 244 F.3d 708, 713 (9th Cir. 2001).\\nRather, a party must point to a “tangible way in which it was\\nprejudiced by the delay.” See Ledo Fin. Corp. v. Summers,\\n122 F.3d 825, 827 (9th Cir. 1997).\\n\\n The only prejudice Garcia asserts is that she was denied\\ndiscovery to test the Salvation Army’s defense. But as a\\nformer member of the Salvation Army’s congregation,\\nGarcia was intimately familiar with its religious focus and\\nmission. See, e.g., ER 133–34 (deposition excerpts\\ndescribing Garcia’s participation in religious services); ER\\n633 (Garcia’s opposition to motion for summary judgment\\n\\f20 GARCIA V. SALVATION ARMY\\n\\nindicating “[s]he believed she would be working in an\\nenvironment that was respectful, [c]haritable and\\nencouraging and above all Christ[-]like”). And the Salvation\\nArmy’s status as a nonprofit corporation is public, along\\nwith its yearly financial reports. Absent prejudice, the\\nSalvation Army permissibly invoked the ROE at summary\\njudgment, and it applies to foreclose Garcia’s Title VII\\nclaims.\\n\\nADA Claim\\n\\n The district court properly granted the Salvation Army\\nsummary judgment on Garcia’s ADA claim. We begin by\\nbriefly recounting the underlying circumstances. Following\\na period of leave under the Family and Medical Leave Act\\nof 1993 (“FMLA”) from October to December 2013, Garcia\\nsought additional personal leave, explaining that she “ha[d]\\nbeen advised not to return to a stressful working\\nenvironment for health reasons.” The Salvation Army\\nresponded by asking her to provide medical documentation\\n“outlining working restrictions and estimated date of return\\nto full duty.” Garcia complied, and her request was granted.\\nThe Salvation Army repeatedly extended Garcia’s leave\\nthrough May 5, 2014.\\n\\n On May 5, 2014, Garcia informed the Salvation Army\\nthat her doctor had cleared her to return to work on May 26,\\n“without restrictions.” But Garcia said she would not return\\nto work because she was “not ready to go back into the exact\\nsame working environment which [her] doctors ha[d]\\nadvised against” and “[t]here seem[ed] to be a mental\\nblock/barrier” regarding the customer complaint filed\\nagainst her in July 2013. She therefore requested the\\n\\f GARCIA V. SALVATION ARMY 21\\n\\n“accommodation” that “a copy or summary of [the client]\\ncomplaint . . . be made available to [her] before [she]\\nreturn[ed] to work.” The Salvation Army declined and\\ndemanded medical evidence supporting her disability and\\nproposed accommodation.\\n\\n This time, Garcia did not submit the requested medical\\ninformation. As a result, the Salvation Army made clear that\\nGarcia’s continued absence was not excused, “jeopardizing\\n[her] continued employment.” Even so, on June 17, 2014,\\nthe Salvation Army provided a summary of the client\\ncomplaint but explained that Garcia would “need a doctor’s\\nnote to allow us to assess your accommodation request and\\nyour continued absence.” Garcia disputed the summary of\\nthe complaint and reiterated her request for a copy. She did\\nnot report to work. On July 10, 2014, the Salvation Army\\nterminated Garcia’s employment due to unexcused absence.\\n\\n “The ADA prohibits discrimination ‘against a qualified\\nindividual on the basis of disability in regard to . . . job\\ntraining[] and other terms, conditions, and privileges of\\nemployment.’” EEOC v. UPS Supply Chain Sols., 620 F.3d\\n1103, 1110 (9th Cir. 2010) (quoting 42 U.S.C. § 12112(a)).\\n“[D]iscrimination includes an employer’s not making\\nreasonable accommodations to the known physical or mental\\nlimitations of an otherwise qualified . . . employee, unless\\n[the employer] can demonstrate that the accommodation\\nwould impose an undue hardship on the operation of [its]\\nbusiness.” U.S. Airways, Inc. v. Barnett, 535 U.S. 391, 396\\n(2002) (emphasis and quotation marks omitted). An\\n“interactive process” is required upon a request for an\\naccommodation. UPS Supply Chain, 620 F.3d at 1110\\n\\f22 GARCIA V. SALVATION ARMY\\n\\n(quoting Zivkovic v. S. Cal. Edison Co., 302 F.3d 1080, 1089\\n(9th Cir. 2002)).\\n\\n “To be entitled to the interactive process that leads to a\\nreasonable accommodation, an employee must have a\\n‘disability’ within the meaning of the ADA.” Becerril v.\\nPima Cty. Assessor’s Office, 587 F.3d 1162, 1164 (9th Cir.\\n2009) (per curiam). A doctor’s release to work without\\nrestrictions supports a finding that a person no longer suffers\\nfrom a “disability.” See, e.g., Rivera v. FedEx Corp., 2013\\nWL 6672401, at *4 (N.D. Cal. Dec. 18, 2013) (plaintiff\\nfailed to demonstrate disability where cleared by doctor\\nwithout restrictions); cf. Stevenson v. Abbott Labs., 639 F.\\nApp’x 473, 474 (9th Cir. 2016) (“[A]fter Plaintiff was\\nreleased to work, she was not disabled.”). Moreover, an\\nemployer “do[es] not have a duty under the ADA . . . to\\nengage in further interactive processes . . . in the absence of”\\nrequested medical evidence. See Allen v. Pac. Bell, 348 F.3d\\n1113, 1115 (9th Cir. 2003) (per curiam).\\n\\n First, Garcia claims that the Salvation Army failed to\\ninteract with her in good faith to arrive at a reasonable\\naccommodation in the months leading up to May 26, 2014,\\nthe day Garcia was slated to return to work. The record does\\nnot support this assertion. To the contrary, each step of the\\nway, the Salvation Army extended Garcia’s leave. This was\\nthe only accommodation requested by Garcia and\\ndocumented by her physician. At no time before her doctor’s\\nunconditional release did Garcia indicate that she desired to\\nreturn to work on a modified basis. Accordingly, no\\nreasonable jury could find that the Salvation Army failed to\\nengage in an interactive process with Garcia from the time\\nof her leave up and until the time she was cleared for work.\\n\\f GARCIA V. SALVATION ARMY 23\\n\\n Garcia next alleges that the Salvation Army refused to\\nnegotiate in good faith after Garcia insisted on receiving a\\ncopy of the secret customer complaint rather than a\\nsummary. This request, however, only came after Garcia\\nwas cleared for work “without restrictions.” Garcia cannot\\nshow that she suffered from a disability at that time, and she\\nfailed to provide supporting medical documentation despite\\nmultiple requests. The Salvation Army was not required to\\ncontinue an interactive process in the absence of medical\\nevidence. See Allen, 348 F.3d at 1115.\\n\\n Moreover, even after the Salvation Army provided a\\nsummary of the client complaint as requested, Garcia\\nprotested and demanded to see the original complaint. This\\nonly underscores that Garcia’s requested “accommodation”\\nis not cognizable under the ADA. A “reasonable\\naccommodation” includes “[m]odifications or adjustments\\nto the work environment, or to the manner or circumstances\\nunder which the position held or desired is customarily\\nperformed, that enable an individual with a disability who is\\nqualified to perform the essential functions of that position.”\\n29 C.F.R. § 1630.2(o)(1)(ii). Obtaining a copy of a year-old\\nclient complaint is unrelated to the “essential functions” of\\nGarcia’s former position. See Gonzagowski v. Widnall, 115\\nF.3d 744, 747–48 (10th Cir. 1997) (“While specific stressors\\nin a work environment may in some cases be legitimate\\ntargets of accommodation, it is unreasonable to require an\\nemployer to create a work environment free of stress and\\ncriticism.”).\\n\\n Finally, Garcia argues that the Salvation Army failed to\\ninteract in good faith because it had decided to eliminate her\\nposition rather than negotiate with her. This allegation,\\n\\f24 GARCIA V. SALVATION ARMY\\n\\nhowever, is derived from a draft letter to Garcia dated May\\n27, 2014, which was never used or sent. To the contrary, the\\nSalvation Army continued Garcia’s leave until July 10,\\n2014, when she was finally terminated. Regardless, as\\ndiscussed, the Salvation Army was under no obligation to\\nengage in an interactive process in the absence of a\\ndisability. For these reasons, Garcia fails to make out a claim\\nunder the ADA.\\n\\n CONCLUSION\\n\\n We hold that the ROE is nonjurisdictional and subject to\\nprocedural forfeiture. Absent prejudice resulting from the\\nSalvation Army’s failure to timely raise the defense,\\nhowever, the ROE forecloses Garcia’s Title VII claims for\\nretaliation and hostile work environment. Garcia’s ADA\\nclaim fails on the merits. The judgment of the district court\\nis therefore AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4380666/\", \"author_raw\": \"KORMAN, District Judge:\"}]}"},"judge1":{"kind":"string","value":"MARY M SCHROEDER"},"judge2":{"kind":"string","value":"PAUL J WATFORD"},"judge3":{"kind":"string","value":"EDWARD R KORMAN"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4603413/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Ann Garcia v. Salvation Army\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-18\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nANN GARCIA, No. 16-16827\n Plaintiff-Appellant,\n D.C. No.\n v. 2:14-cv-02225-DGC\n\nSALVATION ARMY, OPINION\n Defendant-Appellee.\n\n\n Appeal from the United States District Court\n for the District of Arizona\n David G. Campbell, District Judge, Presiding\n\n Argued and Submitted November 13, 2018\n San Francisco, California\n\n Filed March 18, 2019\n\n Before: Mary M. Schroeder and Paul J. Watford, Circuit\n Judges, and Edward R. Korman, * District Judge.\n\n Opinion by Judge Korman\n\n\n\n\n *\n The Honorable Edward R. Korman, United States District Judge\nfor the Eastern District of New York, sitting by designation.\n\f2 GARCIA V. SALVATION ARMY\n\n SUMMARY **\n\n\n Employment Discrimination\n\n The panel affirmed the district court’s summary\njudgment in favor of the Salvation Army, the defendant in\nan employment discrimination action under Title VII and the\nAmericans with Disabilities Act.\n\n The panel held that Title VII’s religious organization\nexemption is not jurisdictional and is subject to procedural\nforfeiture. Absent prejudice resulting from the Salvation\nArmy’s failure to timely raise the defense, however, the\nreligious organization exemption foreclosed plaintiff’s Title\nVII claims because the Salvation Army’s purpose and\ncharacter were primarily religious. The panel held that the\nexemption does not apply only to hiring and firing decisions,\nbut rather extends to both retaliation and hostile work\nenvironment claims.\n\n Affirming the district court’s grant of summary\njudgment on plaintiff’s ADA claim, the panel held that there\nwas no triable issue whether the Salvation Army failed to\nengage in an interactive process in good faith with the\nplaintiff up to the time she was cleared for work after a\nperiod of leave. After the clearance for work, the plaintiff\ncould not show that she was disabled.\n\n\n\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f GARCIA V. SALVATION ARMY 3\n\n COUNSEL\n\nMenno Goedman (argued), Boies Schiller Flexner LLP,\nWashington, D.C.; Kathleen Hartnett, Boies Schiller Flexner\nLLP, Palo Alto, California; for Plaintiff-Appellant.\n\nR. Shawn Oller (argued) and Ryan G. Lockner, Littler\nMendelson, P.C., Phoenix, Arizona; for Defendant-\nAppellee.\n\n\n OPINION\n\nKORMAN, District Judge:\n\n The Salvation Army is an evangelical ministry founded\nin 1865 by William Booth, a former Methodist minister. 1\nThe Salvation Army’s religious tenets differed from\ntraditional Methodism in rejecting the importance of\nsacraments and emphasizing strong central governance. 2 To\nthat end, Booth—“General” of the Salvation Army—\n\n\n\n\n 1\n Unless otherwise indicated, the following historical context is\ngleaned from the record. Otherwise—and purely for background\npurposes—we take judicial notice of certain historical facts that “can be\naccurately and readily determined from sources whose accuracy cannot\nreasonably be questioned.” Fed. R. Evid. 201(b); see Singh v. Ashcroft,\n393 F.3d 903, 905 (9th Cir. 2004).\n\n 2\n Protestantism, ENCYCLOPÆDIA BRITANNICA,\nhttps://www.britannica.com/topic/Protestantism (published Dec. 13,\n2018).\n\f4 GARCIA V. SALVATION ARMY\n\nadopted the military-style hierarchy of the British Army3\nunder which ranked officers were the equivalent of\nministers. 4 In keeping with Protestantism’s nineteenth\ncentury “camp revival,” Booth took his ministry to the\nstreets 5 and began establishing mission centers catering to\nLondon’s poor. 6\n\n What started as a single ministry in the East End of\nLondon spread to the shores of the United States in 1880 7\nand now operates in more than 80 countries through 16,000\nevangelical centers and 3,000 social welfare institutions\nworldwide. 8 The Salvation Army describes itself as “an\nevangelical part of the universal Christian church,” whose\nprofessed mission is “to preach the gospel of Jesus Christ\nand to meet human needs in His name without\ndiscrimination.” Here in the United States, the Salvation\n\n\n\n 3\n William Booth, ENCYCLOPÆDIA BRITANNICA,\nhttps://www.britannica.com/biography/William-Booth (published Aug.\n16, 2018).\n 4\n Salvation Army, ENCYCLOPÆDIA BRITANNICA,\nhttps://www.britannica.com/topic/Salvation-Army (published May 11,\n2018).\n\n 5\n Protestantism, supra note 2.\n\n 6\n Salvation Army, supra note 4.\n\n 7\n William Booth, supra note 3.\n\n 8\n Salvation Army, supra note 4.\n\f GARCIA V. SALVATION ARMY 5\n\nArmy operates through 501(c)(3) nonprofit corporations. 9 In\n2012 and 2013, direct public donations made up the lion’s\nshare of the Salvation Army’s total revenue; sales to the\npublic comprised fifteen percent.\n\n Ann Garcia’s relationship with the Salvation Army dates\nto 1999, when she began attending religious services at the\nEstrella Mountain Corps in Avondale, Arizona. In 2002, the\nCorps hired Garcia to work as an assistant to the pastor, a\nposition she held until July 2010, when Arlene and Dionisio\nTorres became the new pastors. No longer in need of an\nassistant, Arlene Torres reassigned Garcia to the position of\nsocial services coordinator in January 2011. In that role,\nGarcia aided clients under the supervision of Arlene Torres.\nIn late 2011, Garcia and her husband “left the Church” and\nstopped attending the Salvation Army’s religious services,\nbut Garcia continued her work as social services coordinator.\nAfterward, her relationship with Torres began to deteriorate.\n\n Tensions reached new heights in July 2013, when a client\nfiled a lengthy complaint letter against Garcia, claiming that\nshe “refused to provide help to [the client’s] family.” After\nTorres informed Garcia that a complaint had been lodged,\nGarcia demanded to see it. Torres refused, claiming that the\ncomplaint was confidential. Three days later, Garcia filed an\ninternal grievance of her own against Torres, claiming that\nshe “fe[lt] discriminated against and excluded and isolated”\n\n 9\n We take judicial notice of the Salvation Army’s nonprofit status,\nas reflected in the publicly available IRS determination letters at Docket\nEntry No. 45. See Fed. R. Evid. 201(b); Anderson v. Holder, 673 F.3d\n1089, 1094 n.1 (9th Cir. 2012) (“We may take judicial notice of records\nand reports of administrative bodies.” (internal quotation marks\nomitted)).\n\f6 GARCIA V. SALVATION ARMY\n\nat work ever since leaving the church. The specter of the\nundisclosed client grievance continued to disturb Garcia.\nShe would go on to submit complaints to the EEOC and\nArizona state authorities for religious discrimination and\nretaliation.\n\n Following a lengthy period of medical leave due to\nfibromyalgia, the Salvation Army fired Garcia after she\nfailed to report to work despite being cleared by her doctor.\nGarcia then filed a second complaint with the EEOC and\nstate authorities alleging that, by declining to disclose the\nclient complaint, the Salvation Army failed to accommodate\nher disability.\n\n Garcia’s EEOC charges were dismissed, and right-to-sue\nletters issued. Garcia subsequently brought two lawsuits\nagainst the Salvation Army: one under Title VII of the Civil\nRights Act of 1964 (Title VII), 42 U.S.C. § 2000e, et seq.,\nand another under the Americans with Disabilities Act\n(ADA), 42 U.S.C. § 12112, et seq., which were\nconsolidated. In sum, Garcia alleged that the Salvation Army\nsubjected her to a hostile work environment because she\nstopped attending religious services and retaliated against\nher for filing an internal grievance complaining of religion-\nbased mistreatment. The resulting stress precipitated health\nproblems that the Salvation Army failed to accommodate.\n\n The district judge (Campbell, J.) granted summary\njudgment to the Salvation Army, holding that Title VII’s\nreligious organization exemption (ROE) protects the\nSalvation Army from suit, even if it failed to timely assert\nthe defense. Garcia v. Salvation Army, 2016 WL 4732845,\nat *4 (D. Ariz. Sept. 12, 2016). He reasoned that the ROE is\n\f GARCIA V. SALVATION ARMY 7\n\njurisdictional—a matter of courts’ Article III power to hear\ncases and controversies—and cannot be forfeited. Id. The\ndistrict judge also dismissed Garcia’s ADA claims on the\nmerits. Id. at *5–6.\n\n Garcia appeals, raising two legal questions regarding the\napplication and scope of the ROE. First, whether the ROE is\njurisdictional, depriving federal courts of subject matter\njurisdiction when invoked. And second, whether the ROE\nextends beyond hiring and firing decisions to hostile work\nenvironment and retaliation claims. She also challenges the\ndistrict judge’s dismissal of her ADA claims (to which the\nROE does not apply). We first address the application of the\nROE before turning to the merits of the ADA claim.\n\n DISCUSSION\n\nTitle VII Claims\n\n A. The ROE Applies to the Salvation Army\n\n The ROE provides that Title VII’s protections against\ndiscrimination\n\n shall not apply to an employer with respect\n to . . . a religious corporation, association,\n educational institution, or society with\n respect to the employment of individuals of a\n particular religion to perform work connected\n with the carrying on by such corporation,\n association, educational institution, or\n society of its activities.\n\f8 GARCIA V. SALVATION ARMY\n\n42 U.S.C. § 2000e-1(a). The entity seeking the benefit of the\nstatute bears the burden of proving it is exempt. EEOC v.\nKamehameha Schs./Bishop Estate, 990 F.2d 458, 460 (9th\nCir. 1993).\n\n In applying the ROE, we determine whether an\ninstitution’s “purpose and character are primarily religious”\nby weighing “[a]ll significant religious and secular\ncharacteristics.” EEOC v. Townley Eng’g & Mfg. Co., 859\nF.2d 610, 618 (9th Cir. 1988). It does not suffice that an\ninstitution be “merely ‘affiliated’ with a religious\norganization.” Id. at 617. Although we construe the ROE\nnarrowly, often the organization seeking the exemption is\n“clearly” religious. Id. at 618.\n\n This is such a case. See, e.g., Rev. Rul. 59-129, 1959-1\nC.B. 58 (noting that the Salvation Army is a “church” under\nthe Internal Revenue Code); Schleicher v. Salvation Army,\n518 F.3d 472, 478 (7th Cir. 2008) (“The Salvation Army,\nwhich has existed in the United States since 1880, is\nacknowledged to be a completely legitimate church . . . .”);\nMcClure v. Salvation Army, 460 F.2d 553, 554 (5th Cir.\n1972) (“The Salvation Army is a church . . . .”). The\nSalvation Army holds regular religious services. It offers\nsocial services to customers regardless of their religion “to\nreach new populations and spread the gospel.” Indeed, the\nSalvation Army’s mission statement describes it as\n\n an evangelical part of the universal Christian\n church. Its message is based on the Bible. Its\n ministry is motivated by the love of God. Its\n mission is to preach the gospel of Jesus Christ\n\f GARCIA V. SALVATION ARMY 9\n\n and to meet human needs in His name\n without discrimination. 10\n\nThe record establishes that, as a church, the Salvation Army\nis “clearly” religious. See Townley, 859 F.2d at 618 (“[T]he\ncentral function of [the ROE] has been to exempt churches,\nsynagogues, and the like, and organizations closely affiliated\nwith those entities.”). Accord Lown v. Salvation Army, Inc.,\n393 F. Supp. 2d 223, 246–54 (S.D.N.Y. 2005) (applying\nROE to the Salvation Army); Clark v. Salvation Army, LLC,\n2008 WL 11375384, at *1–2 (N.D. Ala. June 16, 2008)\n(same).\n\n Garcia argues that the Salvation Army does not qualify\nfor the ROE because it does not satisfy the fourth factor of a\nfour-part test discussed in Spencer v. World Vision, Inc., 633\nF.3d 723, 724 (9th Cir. 2011) (per curiam), which asks\nwhether an entity “engage[s] primarily or substantially in the\nexchange of goods or services for money beyond nominal\namounts.” Garcia contends this factor is not satisfied\nbecause the Salvation Army generates a large-dollar amount\nof sales revenue, even though that amount constitutes a small\nportion (fifteen percent) of its total income. Moreover, in\n2012, 82 cents of every dollar spent by the Salvation Army\nwent toward its program services. These considerations\naside, the concurring opinions of Judges O’Scannlain and\nKleinfeld, who wrote separately on how to evaluate the\nfourth Spencer factor, support the Salvation Army’s\n\n 10\n The Salvation Army’s mission statement is consistent with the\nSupreme Court’s various articulations of what constitutes a “religion.”\nSee generally Note, Ari J. Diaconis, The Religion of Alcoholics\nAnonymous (AA): Applying the Clergy Privilege to Certain AA\nCommunications, 99 Cornell L. Rev. 1185, 1213–17 (2014).\n\f10 GARCIA V. SALVATION ARMY\n\nentitlement to ROE protection. Under Judge O’Scannlain’s\napproach, the ROE applies if the first three factors identified\nin the per curiam opinion are satisfied and the organization\nis a nonprofit. Spencer, 633 F.3d at 734 (O’Scannlain, J.,\nconcurring). The Salvation Army easily satisfies that test.\nWhile Judge Kleinfeld would alternatively ask how the\norganization charges for its services, he cites the Salvation\nArmy as a group that satisfies this criterion because it “gives\nits homeless shelter and soup kitchen services away, or\ncharges nominal fees.” Id. at 747 (Kleinfeld, J., concurring).\n\n B. The ROE Reaches Claims for Retaliation and Hostile\n Work Environment\n\n Even assuming the ROE applies to the Salvation Army,\nGarcia argues that it does not reach her claims for retaliation\nand hostile work environment. In her view, the ROE applies\nonly to hiring and firing decisions. Although we have not\naddressed this question, other courts have held that the ROE\nextends to both retaliation and hostile work environment\nclaims. See, e.g., Kennedy v. St. Joseph’s Ministries, Inc.,\n657 F.3d 189, 192–94 (4th Cir. 2011); Saeemodarae v.\nMercy Health Servs., 456 F. Supp. 2d 1021, 1040–41 (N.D.\nIowa 2006); Lown, 393 F. Supp. 2d at 254; Hopkins v.\nWomen’s Div., Gen. Bd. of Glob. Ministries, 238 F. Supp. 2d\n174, 180 (D.D.C. 2002); Aguillard v. La. Coll., 341 F. Supp.\n3d 642, 647–48, 652 (W.D. La. 2018); Clark, 2008 WL\n11375384, at *1. We agree.\n\n First, the ROE’s text reaches beyond hiring and firing.\nCongress “painted with a broader brush, exempting religious\norganizations from the entire subchapter of Title VII with\nrespect to the employment of persons of a particular\n\f GARCIA V. SALVATION ARMY 11\n\nreligion.” Kennedy, 657 F.3d at 194 (internal quotation\nmarks omitted) (emphasis added). The “entire subchapter”\nof Title VII includes protections against retaliation and\ndiscriminatory harassment amounting to a hostile work\nenvironment. 11 See 42 U.S.C. §§ 2000e-2(a)(1), 2000e-3(a).\nAnd “employment” encompasses “the breadth of the\nrelationship between the employer and employee,” not just\nhiring and firing. Kennedy, 657 F.3d at 193 (deriving term’s\ndefinition from settled common-law meaning); see\nNationwide Mut. Ins. Co. v. Darden, 503 U.S. 318, 323–24\n(1992) (setting out common-law factors for evaluating an\nemployment relationship, including all aspects of\nemployer’s control over employee’s actions); cf. 20 C.F.R.\n§ 404.1007(a) (“In general, you are a common-law\nemployee if the person you work for may tell you what to do\nand how, when, and where to do it.”).\n\n Limiting “employment” to hiring and firing decisions is\nalso inconsistent with the term’s use throughout Title VII.\nFor example, Section 2000e-2(a)(1) defines “unlawful\n\n 11\n Garcia relies on EEOC v. Pacific Press Publishing Association,\n676 F.2d 1272 (9th Cir. 1982), abrogated on other grounds as\nrecognized by Am. Friends Serv. Comm. Corp. v. Thornburg, 951 F.2d\n957, 960 (9th Cir. 1991), for the proposition that Congress did not intend\nthe ROE to cover retaliation claims. Pacific Press, however, dealt with\nsex-based Title VII claims against a religious publisher. Id. at 1274. The\nplaintiff there claimed that her employer paid her less than her male\ncoworkers and retaliated against her when she filed charges. Id. The\nemployer sought to avoid liability on the grounds that the employee’s act\nof filing a lawsuit violated its religious tenets. See id. at 1275. There, we\nheld that religious employers facing sex-, race-, and national origin-\nbased claims are not immune from liability for “retaliatory actions\nagainst employees who exercise their rights under the statute.” See id. at\n1276. We did not hold that religion-based retaliation claims are beyond\nthe reach of the ROE. Id.\n\f12 GARCIA V. SALVATION ARMY\n\nemployment practice” as “to fail or refuse to hire or\ndischarge any individual, or otherwise to discriminate\nagainst any individual with respect to his compensation,\nterms, conditions, or privileges of employment.” There,\n“‘employment’ . . . encompasses more than hiring or firing;\nif the term were so limited, the second clause would be\nsuperfluous.” Kennedy, 657 F.3d at 193. We agree that\n“‘[e]mployment,’ as used throughout Title VII, simply\ncovers a much broader understanding than mere hiring and\nfiring.” Id. Indeed, Garcia conceded at oral argument that a\nprotected organization may lawfully demote an employee\nbased on religious preference. Accepting as much upends\nher theory that the ROE covers only hiring-and-firing\ndecisions. 12\n\n In a final effort to avoid the ROE’s reach, Garcia\nmaintains that she has not pleaded a cause of action labeled\n“religious discrimination.” Even so, the complaint centers\naround religious discrimination. Specifically, the first\nsentence of Count 1 (retaliation) alleges that the Salvation\nArmy subjected Garcia to “discrimination based on\nreligion.” The complaint then goes on to specify alleged\n\n 12\n Garcia also notes that the 2000 edition of the EEOC Compliance\nManual suggests that the ROE “only applies to hiring and discharge.”\nU.S. Equal Emp’t Opportunity Comm’n, Compliance Manual § 2-\nIII(B)(4)(b)(i) (issued May 12, 2000) (hereinafter “EEOC Compliance\nManual”)). However, the foregoing analysis demonstrates that there is\nno statutory ambiguity for agency guidance to resolve. See Chevron,\nU.S.A., Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 842–43 (1984)\n(“If the intent of Congress is clear, that is the end of the matter . . . .”). In\nany event, “this interpretation, which is viewed under Skidmore\ndeference, contains no attendant rationale, lacks the power to persuade,\nand does not warrant deference.” Kennedy, 657 F.3d at 194 n.9.\n\f GARCIA V. SALVATION ARMY 13\n\nincidents of mistreatment and concludes that they all\noccurred “after [Garcia] engaged in protected activity by\nfiling an internal grievance from what [she] believe[s] was\ndiscrimination based on religion.” 13 Count 2 (hostile work\nenvironment) simply repleads all the allegations in Count 1.\n\n It is true, as Garcia notes, that Title VII treats workplace\ndiscrimination and retaliation claims in separate provisions.\nSee 42 U.S.C. §§ 2000e-2(a), 2000e-3(a). Yet both types of\nclaims require a nexus to discrimination. See id. Because the\nROE permits religious organizations to discriminate based\non religion, retaliation claims based on religious\ndiscrimination fail against protected organizations because\nthe practice “opposed” is not “unlawful.” See id. § 2000e-\n3(a). Likewise, hostile work environment claims must\ninvolve status-based harassment, id. § 2000e-2(a)(1);\nMeritor Sav. Bank, FSB v. Vinson, 477 U.S. 57, 63–64\n(1986), and fail against protected organizations where the\nalleged harassment is based on religion.\n\n C. The ROE Bars Garcia’s Claims\n\n The Salvation Army failed to raise the ROE as an\naffirmative defense in its responsive pleading, which would\nnormally result in forfeiture. 14 See Fed. R. Civ. P. 12(h)(1)–\n\n 13\n Indeed, Garcia’s internal grievance took issue with her treatment\n“ever since [she and her husband] left the church.” And Garcia’s EEOC\ncharge—for which she checked the form’s “religious discrimination”\nand “retaliation” boxes—states, “Ever[] since I left the Church my\nimmediate supervisor . . . has been subjecting me to disparate treatment.”\n 14\n “[F]orfeiture is the failure to make the timely assertion of a right[;]\nwaiver is the intentional relinquishment or abandonment of a known\n\f14 GARCIA V. SALVATION ARMY\n\n(2). The district judge, however, held that the ROE is\njurisdictional, and as such, the defense cannot be\nrelinquished. See Fed. R. Civ. P. 12(h)(3) (“If the court\ndetermines at any time that it lacks subject-matter\njurisdiction, the court must dismiss the action.” (emphasis\nadded)).\n\n 1. The ROE is Nonjurisdictional and Subject to\n Forfeiture\n\n In Arbaugh v. Y & H Corp., 546 U.S. 500, 516 (2006),\nthe Supreme Court established the “bright line” rule that\n“when Congress does not rank a statutory limitation on\ncoverage as jurisdictional, courts should treat the restriction\nas nonjurisdictional in character.” Three factors guide our\nanalysis. See Leeson v. Transamerica Disability Income\nPlan, 671 F.3d 969, 976–77 (9th Cir. 2012). First is whether\nthe provision is “clearly labeled jurisdictional.” Id. at 976\n(quotation marks omitted). Second is whether the provision\nis “located in a jurisdiction-granting provision.” Id. at 976–\n77. Third is whether some “other reasons necessitate[] that\nthe provision be construed as jurisdictional.” Id. at 977.\n\n The ROE is not labeled jurisdictional. See 42 U.S.C.\n§ 2000e-1(a). It does not use the term “jurisdiction” nor\nspeak in jurisdictional terms about the power of United\nStates courts to hear cases under Title VII. Compare id. with\n28 U.S.C. §§ 1345, 1348, 1350. Rather, it appears in a\nseparate provision from that establishing federal courts’\njurisdiction over Title VII claims, which militates against\n\nright.” Hamer v. Neighborhood Hous. Servs. of Chi., 138 S. Ct. 13, 17\nn.1 (2017) (quoting United States v. Olano, 507 U.S. 725, 733 (1993)).\nThus, the case before us concerns a procedural forfeiture.\n\f GARCIA V. SALVATION ARMY 15\n\nclassifying it as jurisdictional. See 42 U.S.C. § 2000e-5(f)(3)\n(conferring jurisdiction over Title VII claims to federal\ndistrict courts). Put otherwise, the ROE limits entitlement to\nrelief in a narrow class of cases, not “the authority of federal\ncourts to adjudicate claims under [Title VII].” Leeson, 671\nF.3d at 978. The district judge erred by ranking the ROE\njurisdictional where Congress did not. See Arbaugh, 546\nU.S. at 516.\n\n The Salvation Army maintains that a claim cannot\npossibly arise under Title VII where Congress has mandated\nTitle VII “shall not apply.” But that is precisely the\nconsequence of Arbaugh. Title VII imposes no liability on\nemployers with fewer than 15 employees. See 42 U.S.C.\n§ 2000e(b). Yet the defendant in Arbaugh—an employer\nwith fewer than 15 employees—forfeited this argument by\nfailing to raise it until after trial. Arbaugh, 546 U.S. at 503–\n04, 510, 516. The employer was held liable, even though\nCongress did not intend Title VII to reach it. Id. at 516. Such\nis the consequence of failing to raise a nonjurisdictional\ndefense. See id. at 510–11.\n\n Our decision in United States v. Hui Hsiung, 778 F.3d\n738 (9th Cir. 2015), provides a useful analog. There, we held\nthat the Foreign Trade Antitrust Improvements Act\n(FTAIA), 15 U.S.C. § 6a, which limits conduct covered by\nthe Sherman Act, is nonjurisdictional. Hui Hsiung, 778 F.3d\nat 752–53. Section 4 of the Sherman Act confers upon\nfederal district courts jurisdiction “to prevent and restrain\nviolations of sections 1 to 7 of this title.” 15 U.S.C. § 4. The\nFTAIA, like the ROE, provides that “[s]ections 1 to 7 of this\ntitle shall not apply,” unless certain conditions are satisfied.\nId. § 6a (emphasis added). Nevertheless, we held that the\n\f16 GARCIA V. SALVATION ARMY\n\nimpact of the FTAIA on Sherman Act claims “is a merits\nquestion, not a jurisdictional one.” Hui Hsiung, 778 F.3d at\n752. As the Supreme Court has explained, “to ask what\nconduct [a statute] reaches is to ask what conduct [the\nstatute] prohibits, which is a merits question.” Morrison v.\nNat’l Austl. Bank Ltd., 561 U.S. 247, 254 (2010). The ROE,\nlike the FTAIA, “removes conduct from [Title VII’s] reach,”\nbearing on the question of “what conduct [Title VII]\nprohibits,” not the power of courts to hear a case. Hui\nHsiung, 778 F.3d at 752 (internal quotation marks and\nalterations omitted).\n\n Indeed, the Seventh Circuit has held that Title VII’s\ndomestic-work requirement, embodied in Section 2000e-\n1(a) alongside the ROE, is nonjurisdictional, as it “appears\noutside of the statute’s jurisdictional provision” and “[t]here\nis no other reason to believe that Congress intended to ‘rank’\nthe restrictions as jurisdictional.” Rabe v. United Air Lines,\nInc., 636 F.3d 866, 869 (7th Cir. 2011); see also Maguire v.\nMarquette Univ., 814 F.2d 1213, 1216 (7th Cir. 1987)\n(application of 42 U.S.C. § 2000e-2(e)(1), permitting\neducational institutions “to hire and employ employees of a\nparticular religion” under certain circumstances, is not a\nquestion of subject matter jurisdiction). And the only court\nto expressly address whether the ROE itself is jurisdictional\nfound that it is not. See Smith v. Angel Food Ministries, Inc.,\n2008 WL 5115037, at *4 (M.D. Ga. Dec. 4, 2008), on\nreconsideration 611 F. Supp. 2d 1346 (M.D. Ga. 2009).\n\n We recognize that a number of courts have suggested\nthat the ROE can never be waived or forfeited. See, e.g., Hall\nv. Baptist Mem’l Health Care Corp., 215 F.3d 618, 625 (6th\nCir. 2000); Ark Encounter, LLC v. Parkinson, 152 F. Supp.\n\f GARCIA V. SALVATION ARMY 17\n\n3d 880, 915 n.25 (E.D. Ky. 2016); Saeemodarae, 456 F.\nSupp. 2d at 1038–39; Siegel v. Truett-McConnell Coll., Inc.,\n13 F. Supp. 2d 1335, 1345 (N.D. Ga. 1994). These cases,\nlike the district court’s opinion below, can be traced to the\nThird Circuit’s decision in Little v. Wuerl, 929 F.2d 944 (3d\nCir. 1991), where a Catholic school invoked the ROE after\ndeclining to rehire a Protestant teacher. The teacher argued\nthat the school “waived” the ROE, because it knew she was\nProtestant when it hired her. Id. at 951. Little held that the\nROE was not “a privilege or interest” that could be waived\nbut rather “a decision by Congress that the government\ninterest in eliminating religious discrimination by religious\norganizations is outweighed by the rights of those\norganizations to be free from government intervention,”\nmeaning that “no act by Little or the Parish could expand the\nstatute’s scope.” Id. Other courts subsequently concluded\nthat the ROE “cannot be waived by either party.” E.g., Hall,\n215 F.3d at 625.\n\n Little is inapposite because it concerned the validity of\nan implied waiver defense based on an organization’s pre-\nsuit conduct—not a procedural forfeiture. See Little, 929\nF.2d at 951. A statutory defense is not unrelinquishable per\nse simply because it embodies Congress’s intent to limit a\nclass of claims. To the contrary, even where Congress has\nmandated that a statute “shall not apply” under certain\ncircumstances, we have routinely applied forfeiture\nprinciples. See, e.g., Magana v. Commonwealth of the\nNorthern Mariana Islands, 107 F.3d 1436, 1445 (9th Cir.\n1997) (failure to plead FLSA exemption can result in\nforfeiture); Brennan v. Valley Towing Co., Inc., 515 F.2d\n100, 104 (9th Cir. 1975) (same). See generally 5 Charles\nAlan Wright & Arthur R. Miller, Federal Practice and\n\f18 GARCIA V. SALVATION ARMY\n\nProcedure § 1271 (3d ed. 2004) (collecting cases). We see\nno reason to treat the ROE differently.\n\n 2. Absent Prejudice, the ROE May Be First Raised\n at Summary Judgment\n\n Although statutory exemptions are not among the\naffirmative defenses enumerated in Rule 8(c), see Fed. R.\nCiv. P. 8(c), we have explained that they must be raised in a\nparty’s “initial responsive pleading.” See Magana, 107 F.3d\nat 1445. Accordingly, the Salvation Army was required to\nplead the ROE as an affirmative defense. Cf. Oden v.\nOktibbeha County, 246 F.3d 458, 466–67 (5th Cir. 2001)\n(holding that Title VII’s personal-staff exception must be\npleaded as an affirmative defense). This makes sense given\nthat the defendant “bear[s] the burden of proving [it is\nexempt],” Kamehameha, 990 F.2d at 460, and the facts\nnecessary to invoke the exemption are “outside of the\nplaintiff’s prima facie case,” Wright & Miller, supra,\n§ 1271.\n\n The Salvation Army did not raise the ROE in its answer\nbut insists that it pleaded enough to put Garcia on notice.\nFirst, it pleaded that Garcia failed to state a claim upon which\nrelief can be granted, a defense which may be raised up and\nuntil the close of trial. Fed. R. Civ. P. 12(h)(2). But simply\nstating that the plaintiff failed to state a claim is insufficient\nto provide notice of a specific affirmative defense. See\nSimmons v. Navajo County, 609 F.3d 1011, 1023 (9th Cir.\n2010) (“The key to determining the sufficiency of pleading\nan affirmative defense is whether it gives plaintiff fair notice\nof the defense.” (quoting Wyshak v. City Nat’l Bank, 607\nF.2d 824, 827 (9th Cir. 1979))), abrogated in part by Castro\n\f GARCIA V. SALVATION ARMY 19\n\nv. County of Los Angeles, 833 F.3d 1060 (9th Cir. 2016) (en\nbanc). And independent of Rule 12(b)(6), Rule 8(c) requires\nthat “[i]n responding to a pleading, a party must\naffirmatively state any avoidance or affirmative defense.”\nFed. R. Civ. P. 8(c).\n\n The Salvation Army also pleaded “that any adverse\nemployment actions taken against plaintiff were taken for\nlawful, legitimate, non-retaliatory, and non-discriminatory\nreasons.” On its face, this defense does not refer to the ROE\nbut rather the standard for rebutting a prima facie claim of\ndiscrimination on the merits. See Coghlan v. Am. Seafoods\nCo. LLC, 413 F.3d 1090, 1094 (9th Cir. 2005). Thus, the\nSalvation Army failed to timely raise the ROE.\n\n Even so, “[i]n the absence of a showing of prejudice . . .\nan affirmative defense may be raised for the first time at\nsummary judgment.” Camarillo v. McCarthy, 998 F.2d 638,\n639 (9th Cir. 1993). There is no prejudice to a plaintiff where\nan “affirmative defense would have been dispositive” if\nasserted “when the action was filed.” Owens v. Kaiser\nFound. Health Plan, Inc., 244 F.3d 708, 713 (9th Cir. 2001).\nRather, a party must point to a “tangible way in which it was\nprejudiced by the delay.” See Ledo Fin. Corp. v. Summers,\n122 F.3d 825, 827 (9th Cir. 1997).\n\n The only prejudice Garcia asserts is that she was denied\ndiscovery to test the Salvation Army’s defense. But as a\nformer member of the Salvation Army’s congregation,\nGarcia was intimately familiar with its religious focus and\nmission. See, e.g., ER 133–34 (deposition excerpts\ndescribing Garcia’s participation in religious services); ER\n633 (Garcia’s opposition to motion for summary judgment\n\f20 GARCIA V. SALVATION ARMY\n\nindicating “[s]he believed she would be working in an\nenvironment that was respectful, [c]haritable and\nencouraging and above all Christ[-]like”). And the Salvation\nArmy’s status as a nonprofit corporation is public, along\nwith its yearly financial reports. Absent prejudice, the\nSalvation Army permissibly invoked the ROE at summary\njudgment, and it applies to foreclose Garcia’s Title VII\nclaims.\n\nADA Claim\n\n The district court properly granted the Salvation Army\nsummary judgment on Garcia’s ADA claim. We begin by\nbriefly recounting the underlying circumstances. Following\na period of leave under the Family and Medical Leave Act\nof 1993 (“FMLA”) from October to December 2013, Garcia\nsought additional personal leave, explaining that she “ha[d]\nbeen advised not to return to a stressful working\nenvironment for health reasons.” The Salvation Army\nresponded by asking her to provide medical documentation\n“outlining working restrictions and estimated date of return\nto full duty.” Garcia complied, and her request was granted.\nThe Salvation Army repeatedly extended Garcia’s leave\nthrough May 5, 2014.\n\n On May 5, 2014, Garcia informed the Salvation Army\nthat her doctor had cleared her to return to work on May 26,\n“without restrictions.” But Garcia said she would not return\nto work because she was “not ready to go back into the exact\nsame working environment which [her] doctors ha[d]\nadvised against” and “[t]here seem[ed] to be a mental\nblock/barrier” regarding the customer complaint filed\nagainst her in July 2013. She therefore requested the\n\f GARCIA V. SALVATION ARMY 21\n\n“accommodation” that “a copy or summary of [the client]\ncomplaint . . . be made available to [her] before [she]\nreturn[ed] to work.” The Salvation Army declined and\ndemanded medical evidence supporting her disability and\nproposed accommodation.\n\n This time, Garcia did not submit the requested medical\ninformation. As a result, the Salvation Army made clear that\nGarcia’s continued absence was not excused, “jeopardizing\n[her] continued employment.” Even so, on June 17, 2014,\nthe Salvation Army provided a summary of the client\ncomplaint but explained that Garcia would “need a doctor’s\nnote to allow us to assess your accommodation request and\nyour continued absence.” Garcia disputed the summary of\nthe complaint and reiterated her request for a copy. She did\nnot report to work. On July 10, 2014, the Salvation Army\nterminated Garcia’s employment due to unexcused absence.\n\n “The ADA prohibits discrimination ‘against a qualified\nindividual on the basis of disability in regard to . . . job\ntraining[] and other terms, conditions, and privileges of\nemployment.’” EEOC v. UPS Supply Chain Sols., 620 F.3d\n1103, 1110 (9th Cir. 2010) (quoting 42 U.S.C. § 12112(a)).\n“[D]iscrimination includes an employer’s not making\nreasonable accommodations to the known physical or mental\nlimitations of an otherwise qualified . . . employee, unless\n[the employer] can demonstrate that the accommodation\nwould impose an undue hardship on the operation of [its]\nbusiness.” U.S. Airways, Inc. v. Barnett, 535 U.S. 391, 396\n(2002) (emphasis and quotation marks omitted). An\n“interactive process” is required upon a request for an\naccommodation. UPS Supply Chain, 620 F.3d at 1110\n\f22 GARCIA V. SALVATION ARMY\n\n(quoting Zivkovic v. S. Cal. Edison Co., 302 F.3d 1080, 1089\n(9th Cir. 2002)).\n\n “To be entitled to the interactive process that leads to a\nreasonable accommodation, an employee must have a\n‘disability’ within the meaning of the ADA.” Becerril v.\nPima Cty. Assessor’s Office, 587 F.3d 1162, 1164 (9th Cir.\n2009) (per curiam). A doctor’s release to work without\nrestrictions supports a finding that a person no longer suffers\nfrom a “disability.” See, e.g., Rivera v. FedEx Corp., 2013\nWL 6672401, at *4 (N.D. Cal. Dec. 18, 2013) (plaintiff\nfailed to demonstrate disability where cleared by doctor\nwithout restrictions); cf. Stevenson v. Abbott Labs., 639 F.\nApp’x 473, 474 (9th Cir. 2016) (“[A]fter Plaintiff was\nreleased to work, she was not disabled.”). Moreover, an\nemployer “do[es] not have a duty under the ADA . . . to\nengage in further interactive processes . . . in the absence of”\nrequested medical evidence. See Allen v. Pac. Bell, 348 F.3d\n1113, 1115 (9th Cir. 2003) (per curiam).\n\n First, Garcia claims that the Salvation Army failed to\ninteract with her in good faith to arrive at a reasonable\naccommodation in the months leading up to May 26, 2014,\nthe day Garcia was slated to return to work. The record does\nnot support this assertion. To the contrary, each step of the\nway, the Salvation Army extended Garcia’s leave. This was\nthe only accommodation requested by Garcia and\ndocumented by her physician. At no time before her doctor’s\nunconditional release did Garcia indicate that she desired to\nreturn to work on a modified basis. Accordingly, no\nreasonable jury could find that the Salvation Army failed to\nengage in an interactive process with Garcia from the time\nof her leave up and until the time she was cleared for work.\n\f GARCIA V. SALVATION ARMY 23\n\n Garcia next alleges that the Salvation Army refused to\nnegotiate in good faith after Garcia insisted on receiving a\ncopy of the secret customer complaint rather than a\nsummary. This request, however, only came after Garcia\nwas cleared for work “without restrictions.” Garcia cannot\nshow that she suffered from a disability at that time, and she\nfailed to provide supporting medical documentation despite\nmultiple requests. The Salvation Army was not required to\ncontinue an interactive process in the absence of medical\nevidence. See Allen, 348 F.3d at 1115.\n\n Moreover, even after the Salvation Army provided a\nsummary of the client complaint as requested, Garcia\nprotested and demanded to see the original complaint. This\nonly underscores that Garcia’s requested “accommodation”\nis not cognizable under the ADA. A “reasonable\naccommodation” includes “[m]odifications or adjustments\nto the work environment, or to the manner or circumstances\nunder which the position held or desired is customarily\nperformed, that enable an individual with a disability who is\nqualified to perform the essential functions of that position.”\n29 C.F.R. § 1630.2(o)(1)(ii). Obtaining a copy of a year-old\nclient complaint is unrelated to the “essential functions” of\nGarcia’s former position. See Gonzagowski v. Widnall, 115\nF.3d 744, 747–48 (10th Cir. 1997) (“While specific stressors\nin a work environment may in some cases be legitimate\ntargets of accommodation, it is unreasonable to require an\nemployer to create a work environment free of stress and\ncriticism.”).\n\n Finally, Garcia argues that the Salvation Army failed to\ninteract in good faith because it had decided to eliminate her\nposition rather than negotiate with her. This allegation,\n\f24 GARCIA V. SALVATION ARMY\n\nhowever, is derived from a draft letter to Garcia dated May\n27, 2014, which was never used or sent. To the contrary, the\nSalvation Army continued Garcia’s leave until July 10,\n2014, when she was finally terminated. Regardless, as\ndiscussed, the Salvation Army was under no obligation to\nengage in an interactive process in the absence of a\ndisability. For these reasons, Garcia fails to make out a claim\nunder the ADA.\n\n CONCLUSION\n\n We hold that the ROE is nonjurisdictional and subject to\nprocedural forfeiture. Absent prejudice resulting from the\nSalvation Army’s failure to timely raise the defense,\nhowever, the ROE forecloses Garcia’s Title VII claims for\nretaliation and hostile work environment. Garcia’s ADA\nclaim fails on the merits. The judgment of the district court\nis therefore AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Ann Garcia v. Salvation Army\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-18\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nANN GARCIA, No. 16-16827\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:14-cv-02225-DGC\\n\\nSALVATION ARMY, OPINION\\n Defendant-Appellee.\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n David G. Campbell, District Judge, Presiding\\n\\n Argued and Submitted November 13, 2018\\n San Francisco, California\\n\\n Filed March 18, 2019\\n\\n Before: Mary M. Schroeder and Paul J. Watford, Circuit\\n Judges, and Edward R. Korman, * District Judge.\\n\\n Opinion by Judge Korman\\n\\n\\n\\n\\n *\\n The Honorable Edward R. Korman, United States District Judge\\nfor the Eastern District of New York, sitting by designation.\\n\\f2 GARCIA V. SALVATION ARMY\\n\\n SUMMARY **\\n\\n\\n Employment Discrimination\\n\\n The panel affirmed the district court’s summary\\njudgment in favor of the Salvation Army, the defendant in\\nan employment discrimination action under Title VII and the\\nAmericans with Disabilities Act.\\n\\n The panel held that Title VII’s religious organization\\nexemption is not jurisdictional and is subject to procedural\\nforfeiture. Absent prejudice resulting from the Salvation\\nArmy’s failure to timely raise the defense, however, the\\nreligious organization exemption foreclosed plaintiff’s Title\\nVII claims because the Salvation Army’s purpose and\\ncharacter were primarily religious. The panel held that the\\nexemption does not apply only to hiring and firing decisions,\\nbut rather extends to both retaliation and hostile work\\nenvironment claims.\\n\\n Affirming the district court’s grant of summary\\njudgment on plaintiff’s ADA claim, the panel held that there\\nwas no triable issue whether the Salvation Army failed to\\nengage in an interactive process in good faith with the\\nplaintiff up to the time she was cleared for work after a\\nperiod of leave. After the clearance for work, the plaintiff\\ncould not show that she was disabled.\\n\\n\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f GARCIA V. SALVATION ARMY 3\\n\\n COUNSEL\\n\\nMenno Goedman (argued), Boies Schiller Flexner LLP,\\nWashington, D.C.; Kathleen Hartnett, Boies Schiller Flexner\\nLLP, Palo Alto, California; for Plaintiff-Appellant.\\n\\nR. Shawn Oller (argued) and Ryan G. Lockner, Littler\\nMendelson, P.C., Phoenix, Arizona; for Defendant-\\nAppellee.\\n\\n\\n OPINION\\n\\nKORMAN, District Judge:\\n\\n The Salvation Army is an evangelical ministry founded\\nin 1865 by William Booth, a former Methodist minister. 1\\nThe Salvation Army’s religious tenets differed from\\ntraditional Methodism in rejecting the importance of\\nsacraments and emphasizing strong central governance. 2 To\\nthat end, Booth—“General” of the Salvation Army—\\n\\n\\n\\n\\n 1\\n Unless otherwise indicated, the following historical context is\\ngleaned from the record. Otherwise—and purely for background\\npurposes—we take judicial notice of certain historical facts that “can be\\naccurately and readily determined from sources whose accuracy cannot\\nreasonably be questioned.” Fed. R. Evid. 201(b); see Singh v. Ashcroft,\\n393 F.3d 903, 905 (9th Cir. 2004).\\n\\n 2\\n Protestantism, ENCYCLOPÆDIA BRITANNICA,\\nhttps://www.britannica.com/topic/Protestantism (published Dec. 13,\\n2018).\\n\\f4 GARCIA V. SALVATION ARMY\\n\\nadopted the military-style hierarchy of the British Army3\\nunder which ranked officers were the equivalent of\\nministers. 4 In keeping with Protestantism’s nineteenth\\ncentury “camp revival,” Booth took his ministry to the\\nstreets 5 and began establishing mission centers catering to\\nLondon’s poor. 6\\n\\n What started as a single ministry in the East End of\\nLondon spread to the shores of the United States in 1880 7\\nand now operates in more than 80 countries through 16,000\\nevangelical centers and 3,000 social welfare institutions\\nworldwide. 8 The Salvation Army describes itself as “an\\nevangelical part of the universal Christian church,” whose\\nprofessed mission is “to preach the gospel of Jesus Christ\\nand to meet human needs in His name without\\ndiscrimination.” Here in the United States, the Salvation\\n\\n\\n\\n 3\\n William Booth, ENCYCLOPÆDIA BRITANNICA,\\nhttps://www.britannica.com/biography/William-Booth (published Aug.\\n16, 2018).\\n 4\\n Salvation Army, ENCYCLOPÆDIA BRITANNICA,\\nhttps://www.britannica.com/topic/Salvation-Army (published May 11,\\n2018).\\n\\n 5\\n Protestantism, supra note 2.\\n\\n 6\\n Salvation Army, supra note 4.\\n\\n 7\\n William Booth, supra note 3.\\n\\n 8\\n Salvation Army, supra note 4.\\n\\f GARCIA V. SALVATION ARMY 5\\n\\nArmy operates through 501(c)(3) nonprofit corporations. 9 In\\n2012 and 2013, direct public donations made up the lion’s\\nshare of the Salvation Army’s total revenue; sales to the\\npublic comprised fifteen percent.\\n\\n Ann Garcia’s relationship with the Salvation Army dates\\nto 1999, when she began attending religious services at the\\nEstrella Mountain Corps in Avondale, Arizona. In 2002, the\\nCorps hired Garcia to work as an assistant to the pastor, a\\nposition she held until July 2010, when Arlene and Dionisio\\nTorres became the new pastors. No longer in need of an\\nassistant, Arlene Torres reassigned Garcia to the position of\\nsocial services coordinator in January 2011. In that role,\\nGarcia aided clients under the supervision of Arlene Torres.\\nIn late 2011, Garcia and her husband “left the Church” and\\nstopped attending the Salvation Army’s religious services,\\nbut Garcia continued her work as social services coordinator.\\nAfterward, her relationship with Torres began to deteriorate.\\n\\n Tensions reached new heights in July 2013, when a client\\nfiled a lengthy complaint letter against Garcia, claiming that\\nshe “refused to provide help to [the client’s] family.” After\\nTorres informed Garcia that a complaint had been lodged,\\nGarcia demanded to see it. Torres refused, claiming that the\\ncomplaint was confidential. Three days later, Garcia filed an\\ninternal grievance of her own against Torres, claiming that\\nshe “fe[lt] discriminated against and excluded and isolated”\\n\\n 9\\n We take judicial notice of the Salvation Army’s nonprofit status,\\nas reflected in the publicly available IRS determination letters at Docket\\nEntry No. 45. See Fed. R. Evid. 201(b); Anderson v. Holder, 673 F.3d\\n1089, 1094 n.1 (9th Cir. 2012) (“We may take judicial notice of records\\nand reports of administrative bodies.” (internal quotation marks\\nomitted)).\\n\\f6 GARCIA V. SALVATION ARMY\\n\\nat work ever since leaving the church. The specter of the\\nundisclosed client grievance continued to disturb Garcia.\\nShe would go on to submit complaints to the EEOC and\\nArizona state authorities for religious discrimination and\\nretaliation.\\n\\n Following a lengthy period of medical leave due to\\nfibromyalgia, the Salvation Army fired Garcia after she\\nfailed to report to work despite being cleared by her doctor.\\nGarcia then filed a second complaint with the EEOC and\\nstate authorities alleging that, by declining to disclose the\\nclient complaint, the Salvation Army failed to accommodate\\nher disability.\\n\\n Garcia’s EEOC charges were dismissed, and right-to-sue\\nletters issued. Garcia subsequently brought two lawsuits\\nagainst the Salvation Army: one under Title VII of the Civil\\nRights Act of 1964 (Title VII), 42 U.S.C. § 2000e, et seq.,\\nand another under the Americans with Disabilities Act\\n(ADA), 42 U.S.C. § 12112, et seq., which were\\nconsolidated. In sum, Garcia alleged that the Salvation Army\\nsubjected her to a hostile work environment because she\\nstopped attending religious services and retaliated against\\nher for filing an internal grievance complaining of religion-\\nbased mistreatment. The resulting stress precipitated health\\nproblems that the Salvation Army failed to accommodate.\\n\\n The district judge (Campbell, J.) granted summary\\njudgment to the Salvation Army, holding that Title VII’s\\nreligious organization exemption (ROE) protects the\\nSalvation Army from suit, even if it failed to timely assert\\nthe defense. Garcia v. Salvation Army, 2016 WL 4732845,\\nat *4 (D. Ariz. Sept. 12, 2016). He reasoned that the ROE is\\n\\f GARCIA V. SALVATION ARMY 7\\n\\njurisdictional—a matter of courts’ Article III power to hear\\ncases and controversies—and cannot be forfeited. Id. The\\ndistrict judge also dismissed Garcia’s ADA claims on the\\nmerits. Id. at *5–6.\\n\\n Garcia appeals, raising two legal questions regarding the\\napplication and scope of the ROE. First, whether the ROE is\\njurisdictional, depriving federal courts of subject matter\\njurisdiction when invoked. And second, whether the ROE\\nextends beyond hiring and firing decisions to hostile work\\nenvironment and retaliation claims. She also challenges the\\ndistrict judge’s dismissal of her ADA claims (to which the\\nROE does not apply). We first address the application of the\\nROE before turning to the merits of the ADA claim.\\n\\n DISCUSSION\\n\\nTitle VII Claims\\n\\n A. The ROE Applies to the Salvation Army\\n\\n The ROE provides that Title VII’s protections against\\ndiscrimination\\n\\n shall not apply to an employer with respect\\n to . . . a religious corporation, association,\\n educational institution, or society with\\n respect to the employment of individuals of a\\n particular religion to perform work connected\\n with the carrying on by such corporation,\\n association, educational institution, or\\n society of its activities.\\n\\f8 GARCIA V. SALVATION ARMY\\n\\n42 U.S.C. § 2000e-1(a). The entity seeking the benefit of the\\nstatute bears the burden of proving it is exempt. EEOC v.\\nKamehameha Schs./Bishop Estate, 990 F.2d 458, 460 (9th\\nCir. 1993).\\n\\n In applying the ROE, we determine whether an\\ninstitution’s “purpose and character are primarily religious”\\nby weighing “[a]ll significant religious and secular\\ncharacteristics.” EEOC v. Townley Eng’g & Mfg. Co., 859\\nF.2d 610, 618 (9th Cir. 1988). It does not suffice that an\\ninstitution be “merely ‘affiliated’ with a religious\\norganization.” Id. at 617. Although we construe the ROE\\nnarrowly, often the organization seeking the exemption is\\n“clearly” religious. Id. at 618.\\n\\n This is such a case. See, e.g., Rev. Rul. 59-129, 1959-1\\nC.B. 58 (noting that the Salvation Army is a “church” under\\nthe Internal Revenue Code); Schleicher v. Salvation Army,\\n518 F.3d 472, 478 (7th Cir. 2008) (“The Salvation Army,\\nwhich has existed in the United States since 1880, is\\nacknowledged to be a completely legitimate church . . . .”);\\nMcClure v. Salvation Army, 460 F.2d 553, 554 (5th Cir.\\n1972) (“The Salvation Army is a church . . . .”). The\\nSalvation Army holds regular religious services. It offers\\nsocial services to customers regardless of their religion “to\\nreach new populations and spread the gospel.” Indeed, the\\nSalvation Army’s mission statement describes it as\\n\\n an evangelical part of the universal Christian\\n church. Its message is based on the Bible. Its\\n ministry is motivated by the love of God. Its\\n mission is to preach the gospel of Jesus Christ\\n\\f GARCIA V. SALVATION ARMY 9\\n\\n and to meet human needs in His name\\n without discrimination. 10\\n\\nThe record establishes that, as a church, the Salvation Army\\nis “clearly” religious. See Townley, 859 F.2d at 618 (“[T]he\\ncentral function of [the ROE] has been to exempt churches,\\nsynagogues, and the like, and organizations closely affiliated\\nwith those entities.”). Accord Lown v. Salvation Army, Inc.,\\n393 F. Supp. 2d 223, 246–54 (S.D.N.Y. 2005) (applying\\nROE to the Salvation Army); Clark v. Salvation Army, LLC,\\n2008 WL 11375384, at *1–2 (N.D. Ala. June 16, 2008)\\n(same).\\n\\n Garcia argues that the Salvation Army does not qualify\\nfor the ROE because it does not satisfy the fourth factor of a\\nfour-part test discussed in Spencer v. World Vision, Inc., 633\\nF.3d 723, 724 (9th Cir. 2011) (per curiam), which asks\\nwhether an entity “engage[s] primarily or substantially in the\\nexchange of goods or services for money beyond nominal\\namounts.” Garcia contends this factor is not satisfied\\nbecause the Salvation Army generates a large-dollar amount\\nof sales revenue, even though that amount constitutes a small\\nportion (fifteen percent) of its total income. Moreover, in\\n2012, 82 cents of every dollar spent by the Salvation Army\\nwent toward its program services. These considerations\\naside, the concurring opinions of Judges O’Scannlain and\\nKleinfeld, who wrote separately on how to evaluate the\\nfourth Spencer factor, support the Salvation Army’s\\n\\n 10\\n The Salvation Army’s mission statement is consistent with the\\nSupreme Court’s various articulations of what constitutes a “religion.”\\nSee generally Note, Ari J. Diaconis, The Religion of Alcoholics\\nAnonymous (AA): Applying the Clergy Privilege to Certain AA\\nCommunications, 99 Cornell L. Rev. 1185, 1213–17 (2014).\\n\\f10 GARCIA V. SALVATION ARMY\\n\\nentitlement to ROE protection. Under Judge O’Scannlain’s\\napproach, the ROE applies if the first three factors identified\\nin the per curiam opinion are satisfied and the organization\\nis a nonprofit. Spencer, 633 F.3d at 734 (O’Scannlain, J.,\\nconcurring). The Salvation Army easily satisfies that test.\\nWhile Judge Kleinfeld would alternatively ask how the\\norganization charges for its services, he cites the Salvation\\nArmy as a group that satisfies this criterion because it “gives\\nits homeless shelter and soup kitchen services away, or\\ncharges nominal fees.” Id. at 747 (Kleinfeld, J., concurring).\\n\\n B. The ROE Reaches Claims for Retaliation and Hostile\\n Work Environment\\n\\n Even assuming the ROE applies to the Salvation Army,\\nGarcia argues that it does not reach her claims for retaliation\\nand hostile work environment. In her view, the ROE applies\\nonly to hiring and firing decisions. Although we have not\\naddressed this question, other courts have held that the ROE\\nextends to both retaliation and hostile work environment\\nclaims. See, e.g., Kennedy v. St. Joseph’s Ministries, Inc.,\\n657 F.3d 189, 192–94 (4th Cir. 2011); Saeemodarae v.\\nMercy Health Servs., 456 F. Supp. 2d 1021, 1040–41 (N.D.\\nIowa 2006); Lown, 393 F. Supp. 2d at 254; Hopkins v.\\nWomen’s Div., Gen. Bd. of Glob. Ministries, 238 F. Supp. 2d\\n174, 180 (D.D.C. 2002); Aguillard v. La. Coll., 341 F. Supp.\\n3d 642, 647–48, 652 (W.D. La. 2018); Clark, 2008 WL\\n11375384, at *1. We agree.\\n\\n First, the ROE’s text reaches beyond hiring and firing.\\nCongress “painted with a broader brush, exempting religious\\norganizations from the entire subchapter of Title VII with\\nrespect to the employment of persons of a particular\\n\\f GARCIA V. SALVATION ARMY 11\\n\\nreligion.” Kennedy, 657 F.3d at 194 (internal quotation\\nmarks omitted) (emphasis added). The “entire subchapter”\\nof Title VII includes protections against retaliation and\\ndiscriminatory harassment amounting to a hostile work\\nenvironment. 11 See 42 U.S.C. §§ 2000e-2(a)(1), 2000e-3(a).\\nAnd “employment” encompasses “the breadth of the\\nrelationship between the employer and employee,” not just\\nhiring and firing. Kennedy, 657 F.3d at 193 (deriving term’s\\ndefinition from settled common-law meaning); see\\nNationwide Mut. Ins. Co. v. Darden, 503 U.S. 318, 323–24\\n(1992) (setting out common-law factors for evaluating an\\nemployment relationship, including all aspects of\\nemployer’s control over employee’s actions); cf. 20 C.F.R.\\n§ 404.1007(a) (“In general, you are a common-law\\nemployee if the person you work for may tell you what to do\\nand how, when, and where to do it.”).\\n\\n Limiting “employment” to hiring and firing decisions is\\nalso inconsistent with the term’s use throughout Title VII.\\nFor example, Section 2000e-2(a)(1) defines “unlawful\\n\\n 11\\n Garcia relies on EEOC v. Pacific Press Publishing Association,\\n676 F.2d 1272 (9th Cir. 1982), abrogated on other grounds as\\nrecognized by Am. Friends Serv. Comm. Corp. v. Thornburg, 951 F.2d\\n957, 960 (9th Cir. 1991), for the proposition that Congress did not intend\\nthe ROE to cover retaliation claims. Pacific Press, however, dealt with\\nsex-based Title VII claims against a religious publisher. Id. at 1274. The\\nplaintiff there claimed that her employer paid her less than her male\\ncoworkers and retaliated against her when she filed charges. Id. The\\nemployer sought to avoid liability on the grounds that the employee’s act\\nof filing a lawsuit violated its religious tenets. See id. at 1275. There, we\\nheld that religious employers facing sex-, race-, and national origin-\\nbased claims are not immune from liability for “retaliatory actions\\nagainst employees who exercise their rights under the statute.” See id. at\\n1276. We did not hold that religion-based retaliation claims are beyond\\nthe reach of the ROE. Id.\\n\\f12 GARCIA V. SALVATION ARMY\\n\\nemployment practice” as “to fail or refuse to hire or\\ndischarge any individual, or otherwise to discriminate\\nagainst any individual with respect to his compensation,\\nterms, conditions, or privileges of employment.” There,\\n“‘employment’ . . . encompasses more than hiring or firing;\\nif the term were so limited, the second clause would be\\nsuperfluous.” Kennedy, 657 F.3d at 193. We agree that\\n“‘[e]mployment,’ as used throughout Title VII, simply\\ncovers a much broader understanding than mere hiring and\\nfiring.” Id. Indeed, Garcia conceded at oral argument that a\\nprotected organization may lawfully demote an employee\\nbased on religious preference. Accepting as much upends\\nher theory that the ROE covers only hiring-and-firing\\ndecisions. 12\\n\\n In a final effort to avoid the ROE’s reach, Garcia\\nmaintains that she has not pleaded a cause of action labeled\\n“religious discrimination.” Even so, the complaint centers\\naround religious discrimination. Specifically, the first\\nsentence of Count 1 (retaliation) alleges that the Salvation\\nArmy subjected Garcia to “discrimination based on\\nreligion.” The complaint then goes on to specify alleged\\n\\n 12\\n Garcia also notes that the 2000 edition of the EEOC Compliance\\nManual suggests that the ROE “only applies to hiring and discharge.”\\nU.S. Equal Emp’t Opportunity Comm’n, Compliance Manual § 2-\\nIII(B)(4)(b)(i) (issued May 12, 2000) (hereinafter “EEOC Compliance\\nManual”)). However, the foregoing analysis demonstrates that there is\\nno statutory ambiguity for agency guidance to resolve. See Chevron,\\nU.S.A., Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 842–43 (1984)\\n(“If the intent of Congress is clear, that is the end of the matter . . . .”). In\\nany event, “this interpretation, which is viewed under Skidmore\\ndeference, contains no attendant rationale, lacks the power to persuade,\\nand does not warrant deference.” Kennedy, 657 F.3d at 194 n.9.\\n\\f GARCIA V. SALVATION ARMY 13\\n\\nincidents of mistreatment and concludes that they all\\noccurred “after [Garcia] engaged in protected activity by\\nfiling an internal grievance from what [she] believe[s] was\\ndiscrimination based on religion.” 13 Count 2 (hostile work\\nenvironment) simply repleads all the allegations in Count 1.\\n\\n It is true, as Garcia notes, that Title VII treats workplace\\ndiscrimination and retaliation claims in separate provisions.\\nSee 42 U.S.C. §§ 2000e-2(a), 2000e-3(a). Yet both types of\\nclaims require a nexus to discrimination. See id. Because the\\nROE permits religious organizations to discriminate based\\non religion, retaliation claims based on religious\\ndiscrimination fail against protected organizations because\\nthe practice “opposed” is not “unlawful.” See id. § 2000e-\\n3(a). Likewise, hostile work environment claims must\\ninvolve status-based harassment, id. § 2000e-2(a)(1);\\nMeritor Sav. Bank, FSB v. Vinson, 477 U.S. 57, 63–64\\n(1986), and fail against protected organizations where the\\nalleged harassment is based on religion.\\n\\n C. The ROE Bars Garcia’s Claims\\n\\n The Salvation Army failed to raise the ROE as an\\naffirmative defense in its responsive pleading, which would\\nnormally result in forfeiture. 14 See Fed. R. Civ. P. 12(h)(1)–\\n\\n 13\\n Indeed, Garcia’s internal grievance took issue with her treatment\\n“ever since [she and her husband] left the church.” And Garcia’s EEOC\\ncharge—for which she checked the form’s “religious discrimination”\\nand “retaliation” boxes—states, “Ever[] since I left the Church my\\nimmediate supervisor . . . has been subjecting me to disparate treatment.”\\n 14\\n “[F]orfeiture is the failure to make the timely assertion of a right[;]\\nwaiver is the intentional relinquishment or abandonment of a known\\n\\f14 GARCIA V. SALVATION ARMY\\n\\n(2). The district judge, however, held that the ROE is\\njurisdictional, and as such, the defense cannot be\\nrelinquished. See Fed. R. Civ. P. 12(h)(3) (“If the court\\ndetermines at any time that it lacks subject-matter\\njurisdiction, the court must dismiss the action.” (emphasis\\nadded)).\\n\\n 1. The ROE is Nonjurisdictional and Subject to\\n Forfeiture\\n\\n In Arbaugh v. Y & H Corp., 546 U.S. 500, 516 (2006),\\nthe Supreme Court established the “bright line” rule that\\n“when Congress does not rank a statutory limitation on\\ncoverage as jurisdictional, courts should treat the restriction\\nas nonjurisdictional in character.” Three factors guide our\\nanalysis. See Leeson v. Transamerica Disability Income\\nPlan, 671 F.3d 969, 976–77 (9th Cir. 2012). First is whether\\nthe provision is “clearly labeled jurisdictional.” Id. at 976\\n(quotation marks omitted). Second is whether the provision\\nis “located in a jurisdiction-granting provision.” Id. at 976–\\n77. Third is whether some “other reasons necessitate[] that\\nthe provision be construed as jurisdictional.” Id. at 977.\\n\\n The ROE is not labeled jurisdictional. See 42 U.S.C.\\n§ 2000e-1(a). It does not use the term “jurisdiction” nor\\nspeak in jurisdictional terms about the power of United\\nStates courts to hear cases under Title VII. Compare id. with\\n28 U.S.C. §§ 1345, 1348, 1350. Rather, it appears in a\\nseparate provision from that establishing federal courts’\\njurisdiction over Title VII claims, which militates against\\n\\nright.” Hamer v. Neighborhood Hous. Servs. of Chi., 138 S. Ct. 13, 17\\nn.1 (2017) (quoting United States v. Olano, 507 U.S. 725, 733 (1993)).\\nThus, the case before us concerns a procedural forfeiture.\\n\\f GARCIA V. SALVATION ARMY 15\\n\\nclassifying it as jurisdictional. See 42 U.S.C. § 2000e-5(f)(3)\\n(conferring jurisdiction over Title VII claims to federal\\ndistrict courts). Put otherwise, the ROE limits entitlement to\\nrelief in a narrow class of cases, not “the authority of federal\\ncourts to adjudicate claims under [Title VII].” Leeson, 671\\nF.3d at 978. The district judge erred by ranking the ROE\\njurisdictional where Congress did not. See Arbaugh, 546\\nU.S. at 516.\\n\\n The Salvation Army maintains that a claim cannot\\npossibly arise under Title VII where Congress has mandated\\nTitle VII “shall not apply.” But that is precisely the\\nconsequence of Arbaugh. Title VII imposes no liability on\\nemployers with fewer than 15 employees. See 42 U.S.C.\\n§ 2000e(b). Yet the defendant in Arbaugh—an employer\\nwith fewer than 15 employees—forfeited this argument by\\nfailing to raise it until after trial. Arbaugh, 546 U.S. at 503–\\n04, 510, 516. The employer was held liable, even though\\nCongress did not intend Title VII to reach it. Id. at 516. Such\\nis the consequence of failing to raise a nonjurisdictional\\ndefense. See id. at 510–11.\\n\\n Our decision in United States v. Hui Hsiung, 778 F.3d\\n738 (9th Cir. 2015), provides a useful analog. There, we held\\nthat the Foreign Trade Antitrust Improvements Act\\n(FTAIA), 15 U.S.C. § 6a, which limits conduct covered by\\nthe Sherman Act, is nonjurisdictional. Hui Hsiung, 778 F.3d\\nat 752–53. Section 4 of the Sherman Act confers upon\\nfederal district courts jurisdiction “to prevent and restrain\\nviolations of sections 1 to 7 of this title.” 15 U.S.C. § 4. The\\nFTAIA, like the ROE, provides that “[s]ections 1 to 7 of this\\ntitle shall not apply,” unless certain conditions are satisfied.\\nId. § 6a (emphasis added). Nevertheless, we held that the\\n\\f16 GARCIA V. SALVATION ARMY\\n\\nimpact of the FTAIA on Sherman Act claims “is a merits\\nquestion, not a jurisdictional one.” Hui Hsiung, 778 F.3d at\\n752. As the Supreme Court has explained, “to ask what\\nconduct [a statute] reaches is to ask what conduct [the\\nstatute] prohibits, which is a merits question.” Morrison v.\\nNat’l Austl. Bank Ltd., 561 U.S. 247, 254 (2010). The ROE,\\nlike the FTAIA, “removes conduct from [Title VII’s] reach,”\\nbearing on the question of “what conduct [Title VII]\\nprohibits,” not the power of courts to hear a case. Hui\\nHsiung, 778 F.3d at 752 (internal quotation marks and\\nalterations omitted).\\n\\n Indeed, the Seventh Circuit has held that Title VII’s\\ndomestic-work requirement, embodied in Section 2000e-\\n1(a) alongside the ROE, is nonjurisdictional, as it “appears\\noutside of the statute’s jurisdictional provision” and “[t]here\\nis no other reason to believe that Congress intended to ‘rank’\\nthe restrictions as jurisdictional.” Rabe v. United Air Lines,\\nInc., 636 F.3d 866, 869 (7th Cir. 2011); see also Maguire v.\\nMarquette Univ., 814 F.2d 1213, 1216 (7th Cir. 1987)\\n(application of 42 U.S.C. § 2000e-2(e)(1), permitting\\neducational institutions “to hire and employ employees of a\\nparticular religion” under certain circumstances, is not a\\nquestion of subject matter jurisdiction). And the only court\\nto expressly address whether the ROE itself is jurisdictional\\nfound that it is not. See Smith v. Angel Food Ministries, Inc.,\\n2008 WL 5115037, at *4 (M.D. Ga. Dec. 4, 2008), on\\nreconsideration 611 F. Supp. 2d 1346 (M.D. Ga. 2009).\\n\\n We recognize that a number of courts have suggested\\nthat the ROE can never be waived or forfeited. See, e.g., Hall\\nv. Baptist Mem’l Health Care Corp., 215 F.3d 618, 625 (6th\\nCir. 2000); Ark Encounter, LLC v. Parkinson, 152 F. Supp.\\n\\f GARCIA V. SALVATION ARMY 17\\n\\n3d 880, 915 n.25 (E.D. Ky. 2016); Saeemodarae, 456 F.\\nSupp. 2d at 1038–39; Siegel v. Truett-McConnell Coll., Inc.,\\n13 F. Supp. 2d 1335, 1345 (N.D. Ga. 1994). These cases,\\nlike the district court’s opinion below, can be traced to the\\nThird Circuit’s decision in Little v. Wuerl, 929 F.2d 944 (3d\\nCir. 1991), where a Catholic school invoked the ROE after\\ndeclining to rehire a Protestant teacher. The teacher argued\\nthat the school “waived” the ROE, because it knew she was\\nProtestant when it hired her. Id. at 951. Little held that the\\nROE was not “a privilege or interest” that could be waived\\nbut rather “a decision by Congress that the government\\ninterest in eliminating religious discrimination by religious\\norganizations is outweighed by the rights of those\\norganizations to be free from government intervention,”\\nmeaning that “no act by Little or the Parish could expand the\\nstatute’s scope.” Id. Other courts subsequently concluded\\nthat the ROE “cannot be waived by either party.” E.g., Hall,\\n215 F.3d at 625.\\n\\n Little is inapposite because it concerned the validity of\\nan implied waiver defense based on an organization’s pre-\\nsuit conduct—not a procedural forfeiture. See Little, 929\\nF.2d at 951. A statutory defense is not unrelinquishable per\\nse simply because it embodies Congress’s intent to limit a\\nclass of claims. To the contrary, even where Congress has\\nmandated that a statute “shall not apply” under certain\\ncircumstances, we have routinely applied forfeiture\\nprinciples. See, e.g., Magana v. Commonwealth of the\\nNorthern Mariana Islands, 107 F.3d 1436, 1445 (9th Cir.\\n1997) (failure to plead FLSA exemption can result in\\nforfeiture); Brennan v. Valley Towing Co., Inc., 515 F.2d\\n100, 104 (9th Cir. 1975) (same). See generally 5 Charles\\nAlan Wright & Arthur R. Miller, Federal Practice and\\n\\f18 GARCIA V. SALVATION ARMY\\n\\nProcedure § 1271 (3d ed. 2004) (collecting cases). We see\\nno reason to treat the ROE differently.\\n\\n 2. Absent Prejudice, the ROE May Be First Raised\\n at Summary Judgment\\n\\n Although statutory exemptions are not among the\\naffirmative defenses enumerated in Rule 8(c), see Fed. R.\\nCiv. P. 8(c), we have explained that they must be raised in a\\nparty’s “initial responsive pleading.” See Magana, 107 F.3d\\nat 1445. Accordingly, the Salvation Army was required to\\nplead the ROE as an affirmative defense. Cf. Oden v.\\nOktibbeha County, 246 F.3d 458, 466–67 (5th Cir. 2001)\\n(holding that Title VII’s personal-staff exception must be\\npleaded as an affirmative defense). This makes sense given\\nthat the defendant “bear[s] the burden of proving [it is\\nexempt],” Kamehameha, 990 F.2d at 460, and the facts\\nnecessary to invoke the exemption are “outside of the\\nplaintiff’s prima facie case,” Wright & Miller, supra,\\n§ 1271.\\n\\n The Salvation Army did not raise the ROE in its answer\\nbut insists that it pleaded enough to put Garcia on notice.\\nFirst, it pleaded that Garcia failed to state a claim upon which\\nrelief can be granted, a defense which may be raised up and\\nuntil the close of trial. Fed. R. Civ. P. 12(h)(2). But simply\\nstating that the plaintiff failed to state a claim is insufficient\\nto provide notice of a specific affirmative defense. See\\nSimmons v. Navajo County, 609 F.3d 1011, 1023 (9th Cir.\\n2010) (“The key to determining the sufficiency of pleading\\nan affirmative defense is whether it gives plaintiff fair notice\\nof the defense.” (quoting Wyshak v. City Nat’l Bank, 607\\nF.2d 824, 827 (9th Cir. 1979))), abrogated in part by Castro\\n\\f GARCIA V. SALVATION ARMY 19\\n\\nv. County of Los Angeles, 833 F.3d 1060 (9th Cir. 2016) (en\\nbanc). And independent of Rule 12(b)(6), Rule 8(c) requires\\nthat “[i]n responding to a pleading, a party must\\naffirmatively state any avoidance or affirmative defense.”\\nFed. R. Civ. P. 8(c).\\n\\n The Salvation Army also pleaded “that any adverse\\nemployment actions taken against plaintiff were taken for\\nlawful, legitimate, non-retaliatory, and non-discriminatory\\nreasons.” On its face, this defense does not refer to the ROE\\nbut rather the standard for rebutting a prima facie claim of\\ndiscrimination on the merits. See Coghlan v. Am. Seafoods\\nCo. LLC, 413 F.3d 1090, 1094 (9th Cir. 2005). Thus, the\\nSalvation Army failed to timely raise the ROE.\\n\\n Even so, “[i]n the absence of a showing of prejudice . . .\\nan affirmative defense may be raised for the first time at\\nsummary judgment.” Camarillo v. McCarthy, 998 F.2d 638,\\n639 (9th Cir. 1993). There is no prejudice to a plaintiff where\\nan “affirmative defense would have been dispositive” if\\nasserted “when the action was filed.” Owens v. Kaiser\\nFound. Health Plan, Inc., 244 F.3d 708, 713 (9th Cir. 2001).\\nRather, a party must point to a “tangible way in which it was\\nprejudiced by the delay.” See Ledo Fin. Corp. v. Summers,\\n122 F.3d 825, 827 (9th Cir. 1997).\\n\\n The only prejudice Garcia asserts is that she was denied\\ndiscovery to test the Salvation Army’s defense. But as a\\nformer member of the Salvation Army’s congregation,\\nGarcia was intimately familiar with its religious focus and\\nmission. See, e.g., ER 133–34 (deposition excerpts\\ndescribing Garcia’s participation in religious services); ER\\n633 (Garcia’s opposition to motion for summary judgment\\n\\f20 GARCIA V. SALVATION ARMY\\n\\nindicating “[s]he believed she would be working in an\\nenvironment that was respectful, [c]haritable and\\nencouraging and above all Christ[-]like”). And the Salvation\\nArmy’s status as a nonprofit corporation is public, along\\nwith its yearly financial reports. Absent prejudice, the\\nSalvation Army permissibly invoked the ROE at summary\\njudgment, and it applies to foreclose Garcia’s Title VII\\nclaims.\\n\\nADA Claim\\n\\n The district court properly granted the Salvation Army\\nsummary judgment on Garcia’s ADA claim. We begin by\\nbriefly recounting the underlying circumstances. Following\\na period of leave under the Family and Medical Leave Act\\nof 1993 (“FMLA”) from October to December 2013, Garcia\\nsought additional personal leave, explaining that she “ha[d]\\nbeen advised not to return to a stressful working\\nenvironment for health reasons.” The Salvation Army\\nresponded by asking her to provide medical documentation\\n“outlining working restrictions and estimated date of return\\nto full duty.” Garcia complied, and her request was granted.\\nThe Salvation Army repeatedly extended Garcia’s leave\\nthrough May 5, 2014.\\n\\n On May 5, 2014, Garcia informed the Salvation Army\\nthat her doctor had cleared her to return to work on May 26,\\n“without restrictions.” But Garcia said she would not return\\nto work because she was “not ready to go back into the exact\\nsame working environment which [her] doctors ha[d]\\nadvised against” and “[t]here seem[ed] to be a mental\\nblock/barrier” regarding the customer complaint filed\\nagainst her in July 2013. She therefore requested the\\n\\f GARCIA V. SALVATION ARMY 21\\n\\n“accommodation” that “a copy or summary of [the client]\\ncomplaint . . . be made available to [her] before [she]\\nreturn[ed] to work.” The Salvation Army declined and\\ndemanded medical evidence supporting her disability and\\nproposed accommodation.\\n\\n This time, Garcia did not submit the requested medical\\ninformation. As a result, the Salvation Army made clear that\\nGarcia’s continued absence was not excused, “jeopardizing\\n[her] continued employment.” Even so, on June 17, 2014,\\nthe Salvation Army provided a summary of the client\\ncomplaint but explained that Garcia would “need a doctor’s\\nnote to allow us to assess your accommodation request and\\nyour continued absence.” Garcia disputed the summary of\\nthe complaint and reiterated her request for a copy. She did\\nnot report to work. On July 10, 2014, the Salvation Army\\nterminated Garcia’s employment due to unexcused absence.\\n\\n “The ADA prohibits discrimination ‘against a qualified\\nindividual on the basis of disability in regard to . . . job\\ntraining[] and other terms, conditions, and privileges of\\nemployment.’” EEOC v. UPS Supply Chain Sols., 620 F.3d\\n1103, 1110 (9th Cir. 2010) (quoting 42 U.S.C. § 12112(a)).\\n“[D]iscrimination includes an employer’s not making\\nreasonable accommodations to the known physical or mental\\nlimitations of an otherwise qualified . . . employee, unless\\n[the employer] can demonstrate that the accommodation\\nwould impose an undue hardship on the operation of [its]\\nbusiness.” U.S. Airways, Inc. v. Barnett, 535 U.S. 391, 396\\n(2002) (emphasis and quotation marks omitted). An\\n“interactive process” is required upon a request for an\\naccommodation. UPS Supply Chain, 620 F.3d at 1110\\n\\f22 GARCIA V. SALVATION ARMY\\n\\n(quoting Zivkovic v. S. Cal. Edison Co., 302 F.3d 1080, 1089\\n(9th Cir. 2002)).\\n\\n “To be entitled to the interactive process that leads to a\\nreasonable accommodation, an employee must have a\\n‘disability’ within the meaning of the ADA.” Becerril v.\\nPima Cty. Assessor’s Office, 587 F.3d 1162, 1164 (9th Cir.\\n2009) (per curiam). A doctor’s release to work without\\nrestrictions supports a finding that a person no longer suffers\\nfrom a “disability.” See, e.g., Rivera v. FedEx Corp., 2013\\nWL 6672401, at *4 (N.D. Cal. Dec. 18, 2013) (plaintiff\\nfailed to demonstrate disability where cleared by doctor\\nwithout restrictions); cf. Stevenson v. Abbott Labs., 639 F.\\nApp’x 473, 474 (9th Cir. 2016) (“[A]fter Plaintiff was\\nreleased to work, she was not disabled.”). Moreover, an\\nemployer “do[es] not have a duty under the ADA . . . to\\nengage in further interactive processes . . . in the absence of”\\nrequested medical evidence. See Allen v. Pac. Bell, 348 F.3d\\n1113, 1115 (9th Cir. 2003) (per curiam).\\n\\n First, Garcia claims that the Salvation Army failed to\\ninteract with her in good faith to arrive at a reasonable\\naccommodation in the months leading up to May 26, 2014,\\nthe day Garcia was slated to return to work. The record does\\nnot support this assertion. To the contrary, each step of the\\nway, the Salvation Army extended Garcia’s leave. This was\\nthe only accommodation requested by Garcia and\\ndocumented by her physician. At no time before her doctor’s\\nunconditional release did Garcia indicate that she desired to\\nreturn to work on a modified basis. Accordingly, no\\nreasonable jury could find that the Salvation Army failed to\\nengage in an interactive process with Garcia from the time\\nof her leave up and until the time she was cleared for work.\\n\\f GARCIA V. SALVATION ARMY 23\\n\\n Garcia next alleges that the Salvation Army refused to\\nnegotiate in good faith after Garcia insisted on receiving a\\ncopy of the secret customer complaint rather than a\\nsummary. This request, however, only came after Garcia\\nwas cleared for work “without restrictions.” Garcia cannot\\nshow that she suffered from a disability at that time, and she\\nfailed to provide supporting medical documentation despite\\nmultiple requests. The Salvation Army was not required to\\ncontinue an interactive process in the absence of medical\\nevidence. See Allen, 348 F.3d at 1115.\\n\\n Moreover, even after the Salvation Army provided a\\nsummary of the client complaint as requested, Garcia\\nprotested and demanded to see the original complaint. This\\nonly underscores that Garcia’s requested “accommodation”\\nis not cognizable under the ADA. A “reasonable\\naccommodation” includes “[m]odifications or adjustments\\nto the work environment, or to the manner or circumstances\\nunder which the position held or desired is customarily\\nperformed, that enable an individual with a disability who is\\nqualified to perform the essential functions of that position.”\\n29 C.F.R. § 1630.2(o)(1)(ii). Obtaining a copy of a year-old\\nclient complaint is unrelated to the “essential functions” of\\nGarcia’s former position. See Gonzagowski v. Widnall, 115\\nF.3d 744, 747–48 (10th Cir. 1997) (“While specific stressors\\nin a work environment may in some cases be legitimate\\ntargets of accommodation, it is unreasonable to require an\\nemployer to create a work environment free of stress and\\ncriticism.”).\\n\\n Finally, Garcia argues that the Salvation Army failed to\\ninteract in good faith because it had decided to eliminate her\\nposition rather than negotiate with her. This allegation,\\n\\f24 GARCIA V. SALVATION ARMY\\n\\nhowever, is derived from a draft letter to Garcia dated May\\n27, 2014, which was never used or sent. To the contrary, the\\nSalvation Army continued Garcia’s leave until July 10,\\n2014, when she was finally terminated. Regardless, as\\ndiscussed, the Salvation Army was under no obligation to\\nengage in an interactive process in the absence of a\\ndisability. For these reasons, Garcia fails to make out a claim\\nunder the ADA.\\n\\n CONCLUSION\\n\\n We hold that the ROE is nonjurisdictional and subject to\\nprocedural forfeiture. Absent prejudice resulting from the\\nSalvation Army’s failure to timely raise the defense,\\nhowever, the ROE forecloses Garcia’s Title VII claims for\\nretaliation and hostile work environment. Garcia’s ADA\\nclaim fails on the merits. The judgment of the district court\\nis therefore AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1298,"cells":{"id":{"kind":"number","value":4603410,"string":"4,603,410"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Phonsavanh Phongmanivan v. Ron Haynes"},"decision_date":{"kind":"string","value":"2019-03-19"},"docket_number":{"kind":"string","value":"16-36018"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Sandra S. Ikuta and Morgan Christen, Circuit Judges, and Jennifer Choe-Groves,** Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"The Honorable Jennifer Choe-Groves, Judge for the United\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n PHONSAVANH No. 16-36018\\n PHONGMANIVAN,\\n Petitioner-Appellant, D.C. No.\\n 2:16-cv-00556-RAJ\\n v.\\n ORDER CERTIFYING\\n RON HAYNES, SCCC QUESTION TO THE\\n Superintendent,* WASHINGTON\\n Respondent-Appellee. SUPREME COURT\\n\\n\\n Filed March 19, 2019*\\n\\n Before: Sandra S. Ikuta and Morgan Christen, Circuit\\n Judges, and Jennifer Choe-Groves,** Judge.\\n\\n Order\\n\\n\\n\\n\\n *\\n Ron Haynes has replaced Margaret Gilbert as Superintendent of\\nStafford Creek Corrections Center.\\n **\\n The Honorable Jennifer Choe-Groves, Judge for the United\\nStates Court of International Trade, sitting by designation.\\n\\f2 PHONGMANIVAN V. HAYNES\\n\\n SUMMARY***\\n\\n\\n Habeas Corpus\\n\\n The panel certified to the Washington Supreme Court the\\nfollowing question:\\n\\n Is the denial of a personal restraint petition\\n final when the Washington Supreme Court\\n denies a motion to modify an order of its\\n Commissioner denying discretionary review\\n of the state appellate court’s denial, or is the\\n denial not final until the Clerk of the\\n Washington Court of Appeals issues a\\n certificate of finality as required by Rule\\n 16.15(e)(1)(c) of the Rules of Appellate\\n Procedure?\\n\\n\\n COUNSEL\\n\\nAnn K. Wagner, Assistant Federal Public Defender, Federal\\nPublic Defender Office, Seattle, Washington, for Petitioner-\\nAppellant.\\n\\nAlex Kostin, Assistant Attorney General, Corrections\\nDivision; Robert W. Ferguson, Attorney General; Office of\\nthe Washington Attorney General, Olympia Washington; for\\nRespondent-Appellee.\\n\\n\\n ***\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f PHONGMANIVAN V. HAYNES 3\\n\\n ORDER\\n\\n We respectfully ask the Washington Supreme Court to\\nanswer the certified question presented below, pursuant to\\nRevised Code of Washington (“RCW”) § 2.60.020, because\\nwe have concluded that a dispositive question of state law\\napplies to the claim, and therefore “it is necessary to ascertain\\nthe local law of [Washington] state in order to dispose of\\n[this] proceeding and the local law has not been clearly\\ndetermined.”\\n\\n This case involves a statutory tolling claim under\\n28 U.S.C. § 2244(d)(2). The Antiterrorism and Effective\\nDeath Penalty Act of 1996 (“AEDPA”) sets a one-year statute\\nof limitations for filing a federal petition for a writ of habeas\\ncorpus. 28 U.S.C. § 2244(d); see also Williams v. Filson, 908\\nF.3d 546, 557 (9th Cir. 2018). The limitations period shall\\nrun from the latest of “the date on which the judgment\\nbecame final by the conclusion of direct review or the\\nexpiration of the time for seeking such review.” 28 U.S.C.\\n§ 2244(d)(1)(A). The statute of limitations is tolled when “a\\nproperly filed application for State post-conviction or other\\ncollateral review with respect to the pertinent judgment or\\nclaim is pending.” Id. § 2244(d)(2). The definition of\\n“pending” for the purposes of statutory tolling is a question\\nof state law. See Carey v. Saffold, 536 U.S. 214, 219–20\\n(2012); Hemmerle v. Schriro, 495 F.3d 1069, 1077 (9th Cir.\\n2007).\\n\\n The issue here is when Appellant’s personal restraint\\npetition (“PRP”) proceeding was no longer “pending” under\\nWashington state law for the purposes of calculating the\\nstatute of limitations applicable to Appellant’s federal habeas\\npetition.\\n\\f4 PHONGMANIVAN V. HAYNES\\n\\n I\\n\\n We summarize the material facts. Phongmanivan was\\nfound guilty of two counts of Washington Assault in the First\\nDegree, RCW § 9A.36.011, with two firearm enhancements,\\nafter a jury trial in 2011. Phongmanivan was sentenced to\\n306 months.\\n\\n Phongmanivan appealed his conviction. The Washington\\nCourt of Appeals affirmed the conviction and denied the\\nmotion for reconsideration. Phongmanivan petitioned the\\nWashington Supreme Court for review, which the court\\ndenied on December 11, 2013. Phongmanivan did not file a\\npetition for writ of certiorari with the Supreme Court of the\\nUnited States within the ninety-day window.\\n\\n Phongmanivan filed a PRP in the Washington Court of\\nAppeals on February 4, 2015. The Washington Court of\\nAppeals issued an order dismissing the PRP on its merits on\\nMay 4, 2015. Phongmanivan then filed a petition for review\\nin the Washington Supreme Court, which the Commissioner\\nof the Washington Supreme Court treated as a motion for\\ndiscretionary review under Rule 16.14(c) of the Washington\\nRules of Appellate Procedure (“RAP”) and denied.\\nPhongmanivan requested that the Washington Supreme Court\\nvacate or modify the Commissioner’s ruling. The\\nWashington Supreme Court denied Phongmanivan’s motion\\nto modify in the PRP proceedings on February 10, 2016. The\\nWashington Court of Appeals filed a certificate of finality on\\nApril 1, 2016.\\n\\n Phongmanivan filed his federal habeas petition with the\\nUnited States District Court for the Western District of\\nWashington on April 9, 2016. The district court, adopting the\\n\\f PHONGMANIVAN V. HAYNES 5\\n\\nmagistrate judge’s report and recommendation, held that\\nPhongmanivan untimely filed his petition and dismissed his\\ncase. The court found that the one-year statute of limitations\\nfor Phongmanivan’s habeas petition began to run on March\\n12, 2014, after Phongmanivan’s ninety-day window to file a\\npetition for writ of certiorari with the Supreme Court of the\\nUnited States. The statute of limitations was tolled 329 days\\nlater, on February 4, 2015, when Phongmanivan filed his\\npersonal restraint petition. Relevant here, the court\\ndetermined that the statute of limitations resumed when the\\nWashington Supreme Court denied the motion to modify the\\nCommissioner’s ruling on February 10, 2016. By the time\\nPhongmanivan signed his federal habeas petition on April 9,\\n2016, the statute of limitations had run for a total of 388 days\\nand therefore was untimely. The lower court denied\\nPhongmanivan’s claim for equitable tolling and dismissed his\\nhabeas petition with prejudice.\\n\\n II\\n\\n Phongmanivan argues on appeal that his federal habeas\\npetition was timely because his PRP was pending under\\nWashington law within the meaning of 28 U.S.C.\\n§ 2244(d)(2) (and so the one-year limitations period did not\\nresume running) until the Clerk of the Washington Court of\\nAppeals issued the certificate of finality on April 1, 2016.\\nThere is some support for this argument in state law. RAP\\n12.7, entitled “Finality of Decision,” provides that the\\nWashington Court of Appeals retains “the power to change or\\nmodify its decision” until one of three events occurs. RAP\\n12.7(a). One of these events includes the issuance of a\\ncertificate of finality with respect to a PRP, pursuant to RAP\\n16.15(e). Phongmanivan supports his argument further by\\nciting State v. Kilgore, 216 P.3d 393 (Wash. 2009) (en banc),\\n\\f6 PHONGMANIVAN V. HAYNES\\n\\nin which the Washington Supreme Court looked to RAP 12.7\\nto help determine the finality of a conviction for purposes of\\nretroactive application of a new rule of law. In reaching its\\nconclusion, Kilgore distinguished a case involving finality of\\na conviction in the PRP context in part because that case “did\\nnot address finality for purposes of retroactivity.” See id. at\\n397 n.5.\\n\\n Respondent-Appellee contends, and the district court\\nconcluded, that Phongmanivan’s PRP proceeding was no\\nlonger pending for the purposes of 28 U.S.C. § 2244(d)(2)\\nwhen the Washington Supreme Court denied the motion to\\nmodify the Commissioner’s ruling because the issuance of a\\ncertificate of finality is a ministerial act. There is also\\nsupport for this conclusion in state law. RAP 16.15(e)\\ndefines a certificate of finality as “the written notification of\\nthe clerk of the appellate court to the trial court and the\\nparties that the proceedings in the appellate court have come\\nto an end.” The Washington Supreme Court has described\\nthe certificate of finality as a document that “has traditionally\\nbeen issued by the clerk’s office when review of an\\ninterlocutory decision is at an end,” which “functions\\nessentially as a ‘mandate’ for such decisions.” In re Personal\\nRestraint Petition of Lord, 870 P.2d 964, 966 n.1 (Wash.\\n1994) (per curiam). Moreover, RAP 16.15(e)(1)(c) requires\\nthe clerk of the Court of Appeals to file a certificate of\\nfinality involving a PRP petition immediately “upon denial of\\nthe motion for discretionary review” by the Washington\\nSupreme Court, rather than ordering the certificate of finality\\nto be issued thirty days after the Court of Appeals decision is\\nfiled (unless a motion for reconsideration or discretionary\\nreview is filed earlier) or thirty days after the Court of\\nAppeals denies a motion for reconsideration (unless a motion\\nfor discretionary review is filed earlier), as in RAP\\n\\f PHONGMANIVAN V. HAYNES 7\\n\\n16.15(e)(1)(a) and (b). This suggests that after the\\nWashington Supreme Court denies the motion for\\ndiscretionary review, the court has nothing left to do but\\n“performance of a ministerial function.” Hemmerle, 495 F.3d\\nat 1077.\\n\\n III\\n\\n In light of the foregoing discussion, and because the\\nanswer to this question is “necessary to ascertain the local law\\nof this state in order to dispose” of this appeal, RCW\\n§ 2.60.020, we respectfully certify to the Washington\\nSupreme Court the following question:\\n\\n Is the denial of a personal restraint petition\\n final when the Washington Supreme Court\\n denies a motion to modify an order of its\\n Commissioner denying discretionary review\\n of the state appellate court’s denial, or is the\\n denial not final until the Clerk of the\\n Washington Court of Appeals issues a\\n certificate of finality as required by Rule\\n 16.15(e)(1)(c) of the Rules of Appellate\\n Procedure?\\n\\n We do not intend our framing of this question to restrict\\nthe Washington State Supreme Court’s consideration of any\\nissues that it determines are relevant. If the Washington State\\nSupreme Court decides to consider the certified question, it\\nmay in its discretion reformulate the question. Broad v.\\nMannesmann Anlagenbau AG, 196 F.3d 1075, 1076 (9th Cir.\\n1999).\\n\\f8 PHONGMANIVAN V. HAYNES\\n\\n If the Washington Supreme Court accepts review of the\\ncertified question, we designate Appellant Phongmanivan as\\nthe party to file the first brief pursuant to RAP 16.16(e)(1).\\n\\n The clerk of our court is hereby ordered to transmit\\nforthwith to the Washington Supreme Court, under official\\nseal of the United States Court of Appeals for the Ninth\\nCircuit, a copy of this order and all relevant briefs and\\nexcerpts of record pursuant to RCW §§ 2.60.010, 2.60.030,\\nand RAP 16.16.\\n\\n Further proceedings in our court are stayed pending the\\nWashington Supreme Court’s decision on whether it will\\naccept review, and if so, receipt of the answer to the certified\\nquestion. This case is withdrawn from submission until\\nfurther order from this court. The Clerk is directed to\\nadministratively close this docket, pending further order.\\n\\n The panel will resume control and jurisdiction on the\\ncertified question upon receiving an answer to the certified\\nquestion or upon the Washington Supreme Court’s decision\\nto decline to answer the certified question. When the\\nWashington Supreme Court decides whether or not to accept\\nthe certified question, the Parties shall file a joint report\\ninforming this court of the decision. If the Washington\\nSupreme Court accepts the certified question, the Parties shall\\nfile a joint status report every six months after the date of the\\nacceptance, or more frequently if circumstances warrant.\\n\\n It is so ORDERED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4380663/\", \"author_raw\": \"The Honorable Jennifer Choe-Groves, Judge for the United\"}]}"},"judge1":{"kind":"string","value":"SANDRA S IKUTA"},"judge2":{"kind":"string","value":"MORGAN CHRISTEN"},"judge3":{"kind":"string","value":"JENNIFER CHOE-GROVES"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4603410/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Phonsavanh Phongmanivan v. Ron Haynes\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-19\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n PHONSAVANH No. 16-36018\n PHONGMANIVAN,\n Petitioner-Appellant, D.C. No.\n 2:16-cv-00556-RAJ\n v.\n ORDER CERTIFYING\n RON HAYNES, SCCC QUESTION TO THE\n Superintendent,* WASHINGTON\n Respondent-Appellee. SUPREME COURT\n\n\n Filed March 19, 2019*\n\n Before: Sandra S. Ikuta and Morgan Christen, Circuit\n Judges, and Jennifer Choe-Groves,** Judge.\n\n Order\n\n\n\n\n *\n Ron Haynes has replaced Margaret Gilbert as Superintendent of\nStafford Creek Corrections Center.\n **\n The Honorable Jennifer Choe-Groves, Judge for the United\nStates Court of International Trade, sitting by designation.\n\f2 PHONGMANIVAN V. HAYNES\n\n SUMMARY***\n\n\n Habeas Corpus\n\n The panel certified to the Washington Supreme Court the\nfollowing question:\n\n Is the denial of a personal restraint petition\n final when the Washington Supreme Court\n denies a motion to modify an order of its\n Commissioner denying discretionary review\n of the state appellate court’s denial, or is the\n denial not final until the Clerk of the\n Washington Court of Appeals issues a\n certificate of finality as required by Rule\n 16.15(e)(1)(c) of the Rules of Appellate\n Procedure?\n\n\n COUNSEL\n\nAnn K. Wagner, Assistant Federal Public Defender, Federal\nPublic Defender Office, Seattle, Washington, for Petitioner-\nAppellant.\n\nAlex Kostin, Assistant Attorney General, Corrections\nDivision; Robert W. Ferguson, Attorney General; Office of\nthe Washington Attorney General, Olympia Washington; for\nRespondent-Appellee.\n\n\n ***\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\f PHONGMANIVAN V. HAYNES 3\n\n ORDER\n\n We respectfully ask the Washington Supreme Court to\nanswer the certified question presented below, pursuant to\nRevised Code of Washington (“RCW”) § 2.60.020, because\nwe have concluded that a dispositive question of state law\napplies to the claim, and therefore “it is necessary to ascertain\nthe local law of [Washington] state in order to dispose of\n[this] proceeding and the local law has not been clearly\ndetermined.”\n\n This case involves a statutory tolling claim under\n28 U.S.C. § 2244(d)(2). The Antiterrorism and Effective\nDeath Penalty Act of 1996 (“AEDPA”) sets a one-year statute\nof limitations for filing a federal petition for a writ of habeas\ncorpus. 28 U.S.C. § 2244(d); see also Williams v. Filson, 908\nF.3d 546, 557 (9th Cir. 2018). The limitations period shall\nrun from the latest of “the date on which the judgment\nbecame final by the conclusion of direct review or the\nexpiration of the time for seeking such review.” 28 U.S.C.\n§ 2244(d)(1)(A). The statute of limitations is tolled when “a\nproperly filed application for State post-conviction or other\ncollateral review with respect to the pertinent judgment or\nclaim is pending.” Id. § 2244(d)(2). The definition of\n“pending” for the purposes of statutory tolling is a question\nof state law. See Carey v. Saffold, 536 U.S. 214, 219–20\n(2012); Hemmerle v. Schriro, 495 F.3d 1069, 1077 (9th Cir.\n2007).\n\n The issue here is when Appellant’s personal restraint\npetition (“PRP”) proceeding was no longer “pending” under\nWashington state law for the purposes of calculating the\nstatute of limitations applicable to Appellant’s federal habeas\npetition.\n\f4 PHONGMANIVAN V. HAYNES\n\n I\n\n We summarize the material facts. Phongmanivan was\nfound guilty of two counts of Washington Assault in the First\nDegree, RCW § 9A.36.011, with two firearm enhancements,\nafter a jury trial in 2011. Phongmanivan was sentenced to\n306 months.\n\n Phongmanivan appealed his conviction. The Washington\nCourt of Appeals affirmed the conviction and denied the\nmotion for reconsideration. Phongmanivan petitioned the\nWashington Supreme Court for review, which the court\ndenied on December 11, 2013. Phongmanivan did not file a\npetition for writ of certiorari with the Supreme Court of the\nUnited States within the ninety-day window.\n\n Phongmanivan filed a PRP in the Washington Court of\nAppeals on February 4, 2015. The Washington Court of\nAppeals issued an order dismissing the PRP on its merits on\nMay 4, 2015. Phongmanivan then filed a petition for review\nin the Washington Supreme Court, which the Commissioner\nof the Washington Supreme Court treated as a motion for\ndiscretionary review under Rule 16.14(c) of the Washington\nRules of Appellate Procedure (“RAP”) and denied.\nPhongmanivan requested that the Washington Supreme Court\nvacate or modify the Commissioner’s ruling. The\nWashington Supreme Court denied Phongmanivan’s motion\nto modify in the PRP proceedings on February 10, 2016. The\nWashington Court of Appeals filed a certificate of finality on\nApril 1, 2016.\n\n Phongmanivan filed his federal habeas petition with the\nUnited States District Court for the Western District of\nWashington on April 9, 2016. The district court, adopting the\n\f PHONGMANIVAN V. HAYNES 5\n\nmagistrate judge’s report and recommendation, held that\nPhongmanivan untimely filed his petition and dismissed his\ncase. The court found that the one-year statute of limitations\nfor Phongmanivan’s habeas petition began to run on March\n12, 2014, after Phongmanivan’s ninety-day window to file a\npetition for writ of certiorari with the Supreme Court of the\nUnited States. The statute of limitations was tolled 329 days\nlater, on February 4, 2015, when Phongmanivan filed his\npersonal restraint petition. Relevant here, the court\ndetermined that the statute of limitations resumed when the\nWashington Supreme Court denied the motion to modify the\nCommissioner’s ruling on February 10, 2016. By the time\nPhongmanivan signed his federal habeas petition on April 9,\n2016, the statute of limitations had run for a total of 388 days\nand therefore was untimely. The lower court denied\nPhongmanivan’s claim for equitable tolling and dismissed his\nhabeas petition with prejudice.\n\n II\n\n Phongmanivan argues on appeal that his federal habeas\npetition was timely because his PRP was pending under\nWashington law within the meaning of 28 U.S.C.\n§ 2244(d)(2) (and so the one-year limitations period did not\nresume running) until the Clerk of the Washington Court of\nAppeals issued the certificate of finality on April 1, 2016.\nThere is some support for this argument in state law. RAP\n12.7, entitled “Finality of Decision,” provides that the\nWashington Court of Appeals retains “the power to change or\nmodify its decision” until one of three events occurs. RAP\n12.7(a). One of these events includes the issuance of a\ncertificate of finality with respect to a PRP, pursuant to RAP\n16.15(e). Phongmanivan supports his argument further by\nciting State v. Kilgore, 216 P.3d 393 (Wash. 2009) (en banc),\n\f6 PHONGMANIVAN V. HAYNES\n\nin which the Washington Supreme Court looked to RAP 12.7\nto help determine the finality of a conviction for purposes of\nretroactive application of a new rule of law. In reaching its\nconclusion, Kilgore distinguished a case involving finality of\na conviction in the PRP context in part because that case “did\nnot address finality for purposes of retroactivity.” See id. at\n397 n.5.\n\n Respondent-Appellee contends, and the district court\nconcluded, that Phongmanivan’s PRP proceeding was no\nlonger pending for the purposes of 28 U.S.C. § 2244(d)(2)\nwhen the Washington Supreme Court denied the motion to\nmodify the Commissioner’s ruling because the issuance of a\ncertificate of finality is a ministerial act. There is also\nsupport for this conclusion in state law. RAP 16.15(e)\ndefines a certificate of finality as “the written notification of\nthe clerk of the appellate court to the trial court and the\nparties that the proceedings in the appellate court have come\nto an end.” The Washington Supreme Court has described\nthe certificate of finality as a document that “has traditionally\nbeen issued by the clerk’s office when review of an\ninterlocutory decision is at an end,” which “functions\nessentially as a ‘mandate’ for such decisions.” In re Personal\nRestraint Petition of Lord, 870 P.2d 964, 966 n.1 (Wash.\n1994) (per curiam). Moreover, RAP 16.15(e)(1)(c) requires\nthe clerk of the Court of Appeals to file a certificate of\nfinality involving a PRP petition immediately “upon denial of\nthe motion for discretionary review” by the Washington\nSupreme Court, rather than ordering the certificate of finality\nto be issued thirty days after the Court of Appeals decision is\nfiled (unless a motion for reconsideration or discretionary\nreview is filed earlier) or thirty days after the Court of\nAppeals denies a motion for reconsideration (unless a motion\nfor discretionary review is filed earlier), as in RAP\n\f PHONGMANIVAN V. HAYNES 7\n\n16.15(e)(1)(a) and (b). This suggests that after the\nWashington Supreme Court denies the motion for\ndiscretionary review, the court has nothing left to do but\n“performance of a ministerial function.” Hemmerle, 495 F.3d\nat 1077.\n\n III\n\n In light of the foregoing discussion, and because the\nanswer to this question is “necessary to ascertain the local law\nof this state in order to dispose” of this appeal, RCW\n§ 2.60.020, we respectfully certify to the Washington\nSupreme Court the following question:\n\n Is the denial of a personal restraint petition\n final when the Washington Supreme Court\n denies a motion to modify an order of its\n Commissioner denying discretionary review\n of the state appellate court’s denial, or is the\n denial not final until the Clerk of the\n Washington Court of Appeals issues a\n certificate of finality as required by Rule\n 16.15(e)(1)(c) of the Rules of Appellate\n Procedure?\n\n We do not intend our framing of this question to restrict\nthe Washington State Supreme Court’s consideration of any\nissues that it determines are relevant. If the Washington State\nSupreme Court decides to consider the certified question, it\nmay in its discretion reformulate the question. Broad v.\nMannesmann Anlagenbau AG, 196 F.3d 1075, 1076 (9th Cir.\n1999).\n\f8 PHONGMANIVAN V. HAYNES\n\n If the Washington Supreme Court accepts review of the\ncertified question, we designate Appellant Phongmanivan as\nthe party to file the first brief pursuant to RAP 16.16(e)(1).\n\n The clerk of our court is hereby ordered to transmit\nforthwith to the Washington Supreme Court, under official\nseal of the United States Court of Appeals for the Ninth\nCircuit, a copy of this order and all relevant briefs and\nexcerpts of record pursuant to RCW §§ 2.60.010, 2.60.030,\nand RAP 16.16.\n\n Further proceedings in our court are stayed pending the\nWashington Supreme Court’s decision on whether it will\naccept review, and if so, receipt of the answer to the certified\nquestion. This case is withdrawn from submission until\nfurther order from this court. The Clerk is directed to\nadministratively close this docket, pending further order.\n\n The panel will resume control and jurisdiction on the\ncertified question upon receiving an answer to the certified\nquestion or upon the Washington Supreme Court’s decision\nto decline to answer the certified question. When the\nWashington Supreme Court decides whether or not to accept\nthe certified question, the Parties shall file a joint report\ninforming this court of the decision. If the Washington\nSupreme Court accepts the certified question, the Parties shall\nfile a joint status report every six months after the date of the\nacceptance, or more frequently if circumstances warrant.\n\n It is so ORDERED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Phonsavanh Phongmanivan v. Ron Haynes\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-19\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n PHONSAVANH No. 16-36018\\n PHONGMANIVAN,\\n Petitioner-Appellant, D.C. No.\\n 2:16-cv-00556-RAJ\\n v.\\n ORDER CERTIFYING\\n RON HAYNES, SCCC QUESTION TO THE\\n Superintendent,* WASHINGTON\\n Respondent-Appellee. SUPREME COURT\\n\\n\\n Filed March 19, 2019*\\n\\n Before: Sandra S. Ikuta and Morgan Christen, Circuit\\n Judges, and Jennifer Choe-Groves,** Judge.\\n\\n Order\\n\\n\\n\\n\\n *\\n Ron Haynes has replaced Margaret Gilbert as Superintendent of\\nStafford Creek Corrections Center.\\n **\\n The Honorable Jennifer Choe-Groves, Judge for the United\\nStates Court of International Trade, sitting by designation.\\n\\f2 PHONGMANIVAN V. HAYNES\\n\\n SUMMARY***\\n\\n\\n Habeas Corpus\\n\\n The panel certified to the Washington Supreme Court the\\nfollowing question:\\n\\n Is the denial of a personal restraint petition\\n final when the Washington Supreme Court\\n denies a motion to modify an order of its\\n Commissioner denying discretionary review\\n of the state appellate court’s denial, or is the\\n denial not final until the Clerk of the\\n Washington Court of Appeals issues a\\n certificate of finality as required by Rule\\n 16.15(e)(1)(c) of the Rules of Appellate\\n Procedure?\\n\\n\\n COUNSEL\\n\\nAnn K. Wagner, Assistant Federal Public Defender, Federal\\nPublic Defender Office, Seattle, Washington, for Petitioner-\\nAppellant.\\n\\nAlex Kostin, Assistant Attorney General, Corrections\\nDivision; Robert W. Ferguson, Attorney General; Office of\\nthe Washington Attorney General, Olympia Washington; for\\nRespondent-Appellee.\\n\\n\\n ***\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f PHONGMANIVAN V. HAYNES 3\\n\\n ORDER\\n\\n We respectfully ask the Washington Supreme Court to\\nanswer the certified question presented below, pursuant to\\nRevised Code of Washington (“RCW”) § 2.60.020, because\\nwe have concluded that a dispositive question of state law\\napplies to the claim, and therefore “it is necessary to ascertain\\nthe local law of [Washington] state in order to dispose of\\n[this] proceeding and the local law has not been clearly\\ndetermined.”\\n\\n This case involves a statutory tolling claim under\\n28 U.S.C. § 2244(d)(2). The Antiterrorism and Effective\\nDeath Penalty Act of 1996 (“AEDPA”) sets a one-year statute\\nof limitations for filing a federal petition for a writ of habeas\\ncorpus. 28 U.S.C. § 2244(d); see also Williams v. Filson, 908\\nF.3d 546, 557 (9th Cir. 2018). The limitations period shall\\nrun from the latest of “the date on which the judgment\\nbecame final by the conclusion of direct review or the\\nexpiration of the time for seeking such review.” 28 U.S.C.\\n§ 2244(d)(1)(A). The statute of limitations is tolled when “a\\nproperly filed application for State post-conviction or other\\ncollateral review with respect to the pertinent judgment or\\nclaim is pending.” Id. § 2244(d)(2). The definition of\\n“pending” for the purposes of statutory tolling is a question\\nof state law. See Carey v. Saffold, 536 U.S. 214, 219–20\\n(2012); Hemmerle v. Schriro, 495 F.3d 1069, 1077 (9th Cir.\\n2007).\\n\\n The issue here is when Appellant’s personal restraint\\npetition (“PRP”) proceeding was no longer “pending” under\\nWashington state law for the purposes of calculating the\\nstatute of limitations applicable to Appellant’s federal habeas\\npetition.\\n\\f4 PHONGMANIVAN V. HAYNES\\n\\n I\\n\\n We summarize the material facts. Phongmanivan was\\nfound guilty of two counts of Washington Assault in the First\\nDegree, RCW § 9A.36.011, with two firearm enhancements,\\nafter a jury trial in 2011. Phongmanivan was sentenced to\\n306 months.\\n\\n Phongmanivan appealed his conviction. The Washington\\nCourt of Appeals affirmed the conviction and denied the\\nmotion for reconsideration. Phongmanivan petitioned the\\nWashington Supreme Court for review, which the court\\ndenied on December 11, 2013. Phongmanivan did not file a\\npetition for writ of certiorari with the Supreme Court of the\\nUnited States within the ninety-day window.\\n\\n Phongmanivan filed a PRP in the Washington Court of\\nAppeals on February 4, 2015. The Washington Court of\\nAppeals issued an order dismissing the PRP on its merits on\\nMay 4, 2015. Phongmanivan then filed a petition for review\\nin the Washington Supreme Court, which the Commissioner\\nof the Washington Supreme Court treated as a motion for\\ndiscretionary review under Rule 16.14(c) of the Washington\\nRules of Appellate Procedure (“RAP”) and denied.\\nPhongmanivan requested that the Washington Supreme Court\\nvacate or modify the Commissioner’s ruling. The\\nWashington Supreme Court denied Phongmanivan’s motion\\nto modify in the PRP proceedings on February 10, 2016. The\\nWashington Court of Appeals filed a certificate of finality on\\nApril 1, 2016.\\n\\n Phongmanivan filed his federal habeas petition with the\\nUnited States District Court for the Western District of\\nWashington on April 9, 2016. The district court, adopting the\\n\\f PHONGMANIVAN V. HAYNES 5\\n\\nmagistrate judge’s report and recommendation, held that\\nPhongmanivan untimely filed his petition and dismissed his\\ncase. The court found that the one-year statute of limitations\\nfor Phongmanivan’s habeas petition began to run on March\\n12, 2014, after Phongmanivan’s ninety-day window to file a\\npetition for writ of certiorari with the Supreme Court of the\\nUnited States. The statute of limitations was tolled 329 days\\nlater, on February 4, 2015, when Phongmanivan filed his\\npersonal restraint petition. Relevant here, the court\\ndetermined that the statute of limitations resumed when the\\nWashington Supreme Court denied the motion to modify the\\nCommissioner’s ruling on February 10, 2016. By the time\\nPhongmanivan signed his federal habeas petition on April 9,\\n2016, the statute of limitations had run for a total of 388 days\\nand therefore was untimely. The lower court denied\\nPhongmanivan’s claim for equitable tolling and dismissed his\\nhabeas petition with prejudice.\\n\\n II\\n\\n Phongmanivan argues on appeal that his federal habeas\\npetition was timely because his PRP was pending under\\nWashington law within the meaning of 28 U.S.C.\\n§ 2244(d)(2) (and so the one-year limitations period did not\\nresume running) until the Clerk of the Washington Court of\\nAppeals issued the certificate of finality on April 1, 2016.\\nThere is some support for this argument in state law. RAP\\n12.7, entitled “Finality of Decision,” provides that the\\nWashington Court of Appeals retains “the power to change or\\nmodify its decision” until one of three events occurs. RAP\\n12.7(a). One of these events includes the issuance of a\\ncertificate of finality with respect to a PRP, pursuant to RAP\\n16.15(e). Phongmanivan supports his argument further by\\nciting State v. Kilgore, 216 P.3d 393 (Wash. 2009) (en banc),\\n\\f6 PHONGMANIVAN V. HAYNES\\n\\nin which the Washington Supreme Court looked to RAP 12.7\\nto help determine the finality of a conviction for purposes of\\nretroactive application of a new rule of law. In reaching its\\nconclusion, Kilgore distinguished a case involving finality of\\na conviction in the PRP context in part because that case “did\\nnot address finality for purposes of retroactivity.” See id. at\\n397 n.5.\\n\\n Respondent-Appellee contends, and the district court\\nconcluded, that Phongmanivan’s PRP proceeding was no\\nlonger pending for the purposes of 28 U.S.C. § 2244(d)(2)\\nwhen the Washington Supreme Court denied the motion to\\nmodify the Commissioner’s ruling because the issuance of a\\ncertificate of finality is a ministerial act. There is also\\nsupport for this conclusion in state law. RAP 16.15(e)\\ndefines a certificate of finality as “the written notification of\\nthe clerk of the appellate court to the trial court and the\\nparties that the proceedings in the appellate court have come\\nto an end.” The Washington Supreme Court has described\\nthe certificate of finality as a document that “has traditionally\\nbeen issued by the clerk’s office when review of an\\ninterlocutory decision is at an end,” which “functions\\nessentially as a ‘mandate’ for such decisions.” In re Personal\\nRestraint Petition of Lord, 870 P.2d 964, 966 n.1 (Wash.\\n1994) (per curiam). Moreover, RAP 16.15(e)(1)(c) requires\\nthe clerk of the Court of Appeals to file a certificate of\\nfinality involving a PRP petition immediately “upon denial of\\nthe motion for discretionary review” by the Washington\\nSupreme Court, rather than ordering the certificate of finality\\nto be issued thirty days after the Court of Appeals decision is\\nfiled (unless a motion for reconsideration or discretionary\\nreview is filed earlier) or thirty days after the Court of\\nAppeals denies a motion for reconsideration (unless a motion\\nfor discretionary review is filed earlier), as in RAP\\n\\f PHONGMANIVAN V. HAYNES 7\\n\\n16.15(e)(1)(a) and (b). This suggests that after the\\nWashington Supreme Court denies the motion for\\ndiscretionary review, the court has nothing left to do but\\n“performance of a ministerial function.” Hemmerle, 495 F.3d\\nat 1077.\\n\\n III\\n\\n In light of the foregoing discussion, and because the\\nanswer to this question is “necessary to ascertain the local law\\nof this state in order to dispose” of this appeal, RCW\\n§ 2.60.020, we respectfully certify to the Washington\\nSupreme Court the following question:\\n\\n Is the denial of a personal restraint petition\\n final when the Washington Supreme Court\\n denies a motion to modify an order of its\\n Commissioner denying discretionary review\\n of the state appellate court’s denial, or is the\\n denial not final until the Clerk of the\\n Washington Court of Appeals issues a\\n certificate of finality as required by Rule\\n 16.15(e)(1)(c) of the Rules of Appellate\\n Procedure?\\n\\n We do not intend our framing of this question to restrict\\nthe Washington State Supreme Court’s consideration of any\\nissues that it determines are relevant. If the Washington State\\nSupreme Court decides to consider the certified question, it\\nmay in its discretion reformulate the question. Broad v.\\nMannesmann Anlagenbau AG, 196 F.3d 1075, 1076 (9th Cir.\\n1999).\\n\\f8 PHONGMANIVAN V. HAYNES\\n\\n If the Washington Supreme Court accepts review of the\\ncertified question, we designate Appellant Phongmanivan as\\nthe party to file the first brief pursuant to RAP 16.16(e)(1).\\n\\n The clerk of our court is hereby ordered to transmit\\nforthwith to the Washington Supreme Court, under official\\nseal of the United States Court of Appeals for the Ninth\\nCircuit, a copy of this order and all relevant briefs and\\nexcerpts of record pursuant to RCW §§ 2.60.010, 2.60.030,\\nand RAP 16.16.\\n\\n Further proceedings in our court are stayed pending the\\nWashington Supreme Court’s decision on whether it will\\naccept review, and if so, receipt of the answer to the certified\\nquestion. This case is withdrawn from submission until\\nfurther order from this court. The Clerk is directed to\\nadministratively close this docket, pending further order.\\n\\n The panel will resume control and jurisdiction on the\\ncertified question upon receiving an answer to the certified\\nquestion or upon the Washington Supreme Court’s decision\\nto decline to answer the certified question. When the\\nWashington Supreme Court decides whether or not to accept\\nthe certified question, the Parties shall file a joint report\\ninforming this court of the decision. If the Washington\\nSupreme Court accepts the certified question, the Parties shall\\nfile a joint status report every six months after the date of the\\nacceptance, or more frequently if circumstances warrant.\\n\\n It is so ORDERED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":1299,"cells":{"id":{"kind":"number","value":4603407,"string":"4,603,407"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"William Herrera v. City of Palmdale"},"decision_date":{"kind":"string","value":"2019-03-20"},"docket_number":{"kind":"string","value":"17-55761"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Diarmuid F. O’Scannlain and Sandra S. Ikuta, Circuit Judges, and George Carem Steeh, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"Opinion by Judge O'Scannlain\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nWILLIAM ROBERT HERRERA; MONA No. 17-55761\\nMOLINA HERRERA, Individually and\\non behalf of their minor child N.P.H.; D.C. No.\\nWILLIAM RYAN HERRERA; PALMDALE 2:16-cv-09453-\\nLODGING, LLC, a California Limited MWF-FFM\\nLiability Company,\\n Plaintiffs-Appellants,\\n OPINION\\n v.\\n\\nCITY OF PALMDALE; COUNTY OF LOS\\nANGELES; BUD DAVIS; GEORGE\\nSCHNEIDER; ROB BRUCE; NARDY\\nLOPEZ; MARK DYLER; ANNE\\nAMBROSE; NOEL JAMES DURAN;\\nJAMES PURTEE; SARA SHREVES;\\nRAPOSAS; BLAKELY; ANTHONY\\nBONELLI; MARK MILLER; MUNOZ;\\nLEON; JACOBS; DANA; MYLES;\\nSORROW; DIAZ; ARCIDIANCONO;\\nDOLLENS; BRANDON; GALLAGHER;\\nWALDEN; DOES, 1–10 inclusive,\\n Defendants-Appellees.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Michael W. Fitzgerald, District Judge, Presiding\\n\\f2 HERRERA V. CITY OF PALMDALE\\n\\n Argued and Submitted December 6, 2018\\n Pasadena, California\\n\\n Filed March 20, 2019\\n\\n Before: Diarmuid F. O’Scannlain and Sandra S. Ikuta,\\n Circuit Judges, and George Carem Steeh, * District Judge.\\n\\n Opinion by Judge O’Scannlain\\n\\n\\n SUMMARY **\\n\\n\\n Civil Rights\\n\\n The panel affirmed in part and reversed in part the\\ndistrict court’s order granting defendants’ motions for\\nabstention in an action brought pursuant to 42 U.S.C. § 1983\\nand the Fair Housing Act alleging that the City of Palmdale\\nand the County of Los Angeles committed numerous\\nviolations in connection with assessing code violations on\\nplaintiffs’ motel property and evicting plaintiffs and their\\ntenants from the motel.\\n\\n When plaintiffs brought their suit in federal court, the\\nCity almost simultaneously filed a complaint in state court\\nasserting that plaintiffs’ hotel was a public nuisance and\\nseeking the appointment of a receiver to take possession and\\n\\n *\\n The Honorable George Caram Steeh III, United States District\\nJudge for the Eastern District of Michigan, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f HERRERA V. CITY OF PALMDALE 3\\n\\ncontrol of the property. The district court found that\\nabstention was proper under Younger v. Harris, 401 U.S. 37\\n(1971), and dismissed the claims for declaratory and\\ninjunctive relief, and stayed the claims for damages pending\\nresolution of proceedings in the state action.\\n\\n The panel first held that it had jurisdiction over the\\nappeal from the district court’s order granting the motions\\nfor abstention even though the district court merely stayed\\nrather than dismissed the damages claims. The panel held\\nthat the state nuisance enforcement action brought by the\\nCity (1) was a civil enforcement proceeding within the scope\\nof the Younger doctrine; (2) implicated important state\\ninterests; and (3) provided an adequate opportunity to raise\\nthe federal constitutional claims. The panel further\\nconcluded that because plaintiffs’ request for declaratory\\nrelief in the district court would have the same practical\\nimpact as injunctive relief on the pending state proceeding\\nas a result of the preclusive effect of the federal court\\njudgment, Younger abstention was also appropriate as to\\nsuch relief.\\n\\n Addressing the claims for damages under § 1983 for\\nviolations of the First, Fifth, and Fourteenth Amendments\\nand the Contract Clause of the Constitution and the Fair\\nHousing Act, the panel held that success by the plaintiffs on\\nsuch claims would invalidate the code enforcement\\nproceeding, and Younger abstention was therefore\\nappropriate as to those claims.\\n\\n Finally, the panel held that the district court erred in\\nabstaining from the § 1983 damages claim alleging\\nviolations of the Fourth Amendment. The panel held that the\\nrelief sought based on alleged Fourth Amendment violations\\nsimply did not meet the Court’s requirement that the relief\\n\\f4 HERRERA V. CITY OF PALMDALE\\n\\nhave the practical effect of enjoining the state court\\nproceeding. The panel stated that the Fourth Amendment\\nclaim must be severed from the other claims and that the\\ndistrict court should consider it on the merits upon remand.\\n\\n\\n COUNSEL\\n\\nFrank Alan Weiser (argued), Law Offices of Frank A.\\nWeiser, Los Angeles, California, for Plaintiffs-Appellants.\\n\\nJohn M. Fujii (argued) and Matthew R. Silver, Silver &\\nWright LLP, Irvine, California, for Defendant-Appellee City\\nof Palmdale.\\n\\nAshlee Clark (argued), Gilbert M. Nishimura, and Andrew\\nCharles Pongracz, Seki, Nishimura & Watase, LLP, Los\\nAngeles, California, for Defendant-Appellee County of Los\\nAngeles.\\n\\f HERRERA V. CITY OF PALMDALE 5\\n\\n OPINION\\n\\nO’SCANNLAIN, Circuit Judge:\\n\\n We must decide whether a district court may abstain\\nfrom addressing claims that seek federal relief while a\\nsimultaneous action is ongoing in state court.\\n\\n I\\n\\n A\\n\\n William (“Bill”) Herrera and his wife, Mona Herrera,\\noperate a 48-unit motel in Palmdale, California. The motel\\nis owned by Palmdale Lodging, LLC, a privately owned\\ncompany formed by Bill and Mona. The City of Palmdale\\n(“City”) licensed Palmdale Lodging to operate the motel.\\nAfter purchasing the motel in June 2016, Bill, Mona, and\\nPalmdale Lodging spent approximately $250,000 to upgrade\\nand to renovate the motel, and contracted with Motel 6 to\\noperate the motel as a franchise. They also evicted those\\nwhom they considered to be “problem tenants who had prior\\nto their ownership and possession caused crime problems at\\nthe motel.”\\n\\n On November 17, 2016, the City obtained a civil\\ninspection warrant to investigate suspected violations of the\\nPalmdale Municipal Code, the California Building\\nStandards Code, the California Health and Safety Code, and\\nother laws. The warrant was executed on November 21,\\n2016. Bill and Mona allege that the inspection warrant was\\nexecuted without notice and included a search of their\\npersonal residence, located within the motel, without their\\npermission. They further claim that sheriffs from the County\\nof Los Angeles (“County”) held Bill and their two children\\n\\f6 HERRERA V. CITY OF PALMDALE\\n\\nat gunpoint for an hour and a half during the inspection of\\ntheir personal residence.\\n\\n On December 13, 2016, the City issued a Notice and\\nOrder to Repair or Abate (“Notice and Order”), which\\nidentified more than 400 code violations on the motel\\nproperty. The Notice and Order required the violations to be\\nrepaired or abated within thirty days, and ordered the Herrera\\nfamily and all motel tenants to vacate the property within\\ntwo days.\\n\\n On December 15, 2016, the City and County closed the\\nmotel and evicted the Herrera family and all motel tenants.\\nBill and Mona contend that they were not afforded an\\nopportunity to appeal the Notice and Order. Bill and Mona\\nfurther contend that the City and County prevented them\\nfrom doing “any work to upgrade or repair the motel for the\\nalleged code violations” despite their requests to do so.\\n\\n B\\n\\n On December 21, 2016, Bill and Mona, individually and\\non behalf of their minor daughter, their son William Ryan\\nHerrera, and Palmdale Lodging (collectively, the\\n“Herreras”) filed this federal civil rights action in the Central\\nDistrict of California against the City, County, and\\nindividual defendants, alleging violations under 42 U.S.C.\\n§ 1983 and the Fair Housing Act (“FHA”), 42 U.S.C.\\n§ 3601, et. seq. (“the federal action”). A First Amended\\nComplaint alleged numerous federal constitutional\\nviolations by the City, County, and City and County\\nofficials, and a disparate impact claim under the FHA. The\\nHerreras requested declaratory and injunctive relief,\\nmonetary damages, and attorney’s fees.\\n\\f HERRERA V. CITY OF PALMDALE 7\\n\\n Almost simultaneously, the City filed a Nuisance\\nComplaint in Los Angeles County Superior Court against\\nBill Herrera and Palmdale Lodging (“the state action”). The\\nCity sought, among other things, a declaration that the motel\\nis a public nuisance, the appointment of a receiver to take\\npossession and control of the property, and injunctive relief\\nprohibiting the state-action defendants “from maintaining or\\nallowing any public nuisances” and “requiring [them] to\\nabate all violations of law” on the property.\\n\\n In the federal action, the City filed a motion for\\nabstention, or, in the alternative, a motion to dismiss. The\\nCounty filed a nearly identical motion the next day. The\\nHerreras opposed both motions. On May 16, 2017, the\\ndistrict court granted both motions, concluding that\\nabstention was appropriate under Younger v. Harris, 401\\nU.S. 37 (1971). The district court therefore dismissed the\\nclaims for declaratory and injunctive relief in the federal\\naction, and stayed the claims for damages pending resolution\\nof proceedings in the state action.\\n\\n The Herreras timely appeal, challenging the district\\ncourt’s order granting the motions for abstention. 1\\n\\n II\\n\\n As a threshold matter, the City argues that our court lacks\\njurisdiction over the appeal from the district court order\\ngranting the motions for abstention because the district court\\n\\n 1\\n The City’s unopposed motion for judicial notice of the state court\\ncomplaint, state court order appointing a receiver, and state court docket,\\nand the County’s two unopposed motions for judicial notice of the state\\naction defendants’ answer to the state court complaint and state action\\ndefendants’ reservation of federal claims in the state action are\\nGRANTED.\\n\\f8 HERRERA V. CITY OF PALMDALE\\n\\nhas not entered final judgment. Specifically, the City argues\\nthat we are without jurisdiction over the appeal because the\\ndistrict court has merely stayed, rather than dismissed, the\\nHerreras’ damages claims, and thus it retains jurisdiction\\nover such claims.\\n\\n A\\n\\n Under 28 U.S.C. § 1291, the Court of Appeals has\\njurisdiction over appeals from “final decisions of the district\\ncourts of the United States.” Generally, “[a] district court\\norder abstaining under Younger and dismissing the case ends\\nthe litigation. It is a final appealable order.” Confederated\\nSalish v. Simonich, 29 F.3d 1398, 1401 (9th Cir. 1994).\\nIndeed, when a court abstains under Younger, claims for\\ninjunctive and declaratory relief are typically dismissed. See\\nGilbertson v. Albright, 381 F.3d 965, 975, 981 (9th Cir.\\n2004) (en banc). However, our court has also recognized\\nthat, when a district court abstains from considering a\\ndamages claim under Younger, it must stay—rather than\\ndismiss—the damages action until state proceedings\\nconclude. See id. at 984 (“[W]hen damages are at issue\\nrather than discretionary relief, deference—rather than\\ndismissal—is the proper restraint. To stay instead of to\\ndismiss the federal action preserves the state’s interests in its\\nown procedures, the federal plaintiff’s opportunity to seek\\ncompensation in the forum of his choice, and an appropriate\\nbalance of federal-state jurisdiction.”). Pursuant to this rule,\\nthe district court dismissed the claims for declaratory and\\ninjunctive relief, and stayed the claims for damages pending\\nresolution of the state court proceedings.\\n\\n B\\n\\n “[A] stay is not ordinarily a final decision for purposes\\nof § 1291.” Moses H. Cone Mem’l Hosp. v. Mercury Constr.\\n\\f HERRERA V. CITY OF PALMDALE 9\\n\\nCorp., 460 U.S. 1, 10 n.11 (1983). However, a stay order\\nmay be “final for purposes of appellate jurisdiction” where\\nthe order puts the litigant “effectively out of court.” Id. at 9;\\nsee also Idlewild Liquor Corp. v. Epstein, 370 U.S. 713, 715\\nn.2 (1962). This is precisely what the stay of the Herreras’\\ndamages claims has effected here. “[T]he object of the stay\\nis to require . . . an essential part of the federal suit to be\\nlitigated in a state forum.” Moses H. Cone, 460 U.S. at 10\\nn.11. Yet such stay is “lengthy and indefinite,” which\\ncreates “a danger of denying justice by delay.” Blue Cross\\n& Blue Shield of Ala. v. Unity Outpatient Surgery Ctr., Inc.,\\n490 F.3d 718, 724 (9th Cir. 2007); see also id. at 723\\n(“Moses H. Cone applies whenever there is a possibility that\\nproceedings in another court could moot a suit or an issue,\\neven if there is no guarantee that they will do so.”). 2\\n\\n\\n\\n 2\\n We recognize that the district court has permitted the Herreras to\\nreserve their stayed damages claims for litigation in federal court under\\nEngland v. Louisiana State Board of Medical Examiners, 375 U.S. 411,\\n421–22 (1964). England protects a “litigant who has properly invoked\\nthe jurisdiction of a Federal District Court to consider federal\\nconstitutional claims” from being compelled “to accept instead a state\\ncourt’s determination of those claims.” Id. at 415. Although a district\\ncourt order granting an England reservation of jurisdiction “is not a final\\njudgment immediately appealable under 28 U.S.C. § 1291” because\\n“[s]uch order does not end the litigation on the merits,” Confederated\\nSalish, 29 F.3d at 1406, the England reservation is not at issue on appeal\\nhere. The district court order granting Younger abstention is appealable\\nunder Moses H. Cone, and we see no reason why the Herreras’\\nreservation of claims should preclude appellate review of the otherwise\\nappealable order granting Younger abstention. Cf. Confederated Salish,\\n29 F.3d at 1406–07 (concluding that our court lacked jurisdiction over\\nthe appeal of the order granting England reservation, but nonetheless\\nexercising jurisdiction over the appeal of the order granting abstention\\nunder Railroad Commission of Texas v. Pullman, 312 U.S. 496 (1941)).\\nHere, the reservation simply provides insurance to the Herreras that they\\n\\f10 HERRERA V. CITY OF PALMDALE\\n\\n Under Moses H. Cone, even the stay—the resolution\\nrequired by our court when granting Younger abstention on\\ndamages claims—is effectively a final decision and thus the\\ndistrict court order is final for purposes of appellate review.\\nSee also Parris v. Taft, 630 F. App’x 895, 898 (11th Cir.\\n2015) (“Because the district court stayed the case until Mr.\\nParris’s related state criminal proceedings were resolved, the\\norder left him effectively out of court. Thus, the order is a\\nfinal decision, and we have appellate jurisdiction.”). We are\\nsatisfied that we have jurisdiction to review the district court\\norder on appeal.\\n\\n III\\n\\n Turning to the merits, the Herreras argue that the district\\ncourt erred by abstaining under Younger.\\n\\n Younger abstention is grounded in a “longstanding\\npublic policy against federal court interference with state\\ncourt proceedings.” Younger, 401 U.S. at 43. The Supreme\\nCourt has “identified two sources for this policy: the\\nconstraints of equity jurisdiction and the concern for comity\\nin our federal system.” Gilbertson, 381 F.3d at 970. Most\\nimportantly, Younger abstention permits federal courts to\\n“preserve respect for state functions such that the national\\ngovernment protects federal rights and interests in a way that\\nwill not ‘unduly interfere with the legitimate activities of the\\nStates.’” Id. (quoting Younger, 401 U.S. at 44).\\n\\n A federal court may abstain under Younger in three\\ncategories of cases: “(1) parallel, pending state criminal\\nproceedings, (2) state civil proceedings that are akin to\\n\\nmay litigate their stayed damages claims after state court proceedings\\nhave finished.\\n\\f HERRERA V. CITY OF PALMDALE 11\\n\\ncriminal prosecutions, and (3) state civil proceedings that\\nimplicate a State’s interest in enforcing the orders and\\njudgments of its courts.” ReadyLink Healthcare, Inc. v.\\nState Comp. Ins. Fund, 754 F.3d 754, 759 (9th Cir. 2014)\\n(internal quotation marks and citations omitted). First\\nidentified in New Orleans Public Service, Inc. v. Council of\\nNew Orleans (“NOPSI”), 491 U.S. 350 (1989), these three\\ncategories are known as the NOPSI categories. See Sprint\\nCommc’ns, Inc. v. Jacobs, 571 U.S. 69, 72–73 (2013).\\n\\n To warrant Younger abstention, a state civil action must\\nfall into one of the NOPSI categories, and must also satisfy\\na three-part inquiry: the state proceeding must be\\n(1) “ongoing,” (2) “implicate important state interests,” and\\n(3) provide “an adequate opportunity . . . to raise\\nconstitutional challenges.” Middlesex Cty. Ethics Comm. v.\\nGarden State Bar Ass’n, 457 U.S. 423, 432 (1982); see also\\nReadyLink, 754 F.3d at 759. If the state proceeding falls into\\none of the NOPSI categories and meets the three Middlesex\\nfactors, a federal court may abstain under Younger so long\\nas “the federal action would have the practical effect of\\nenjoining the state proceedings.” ReadyLink, 754 F.3d at\\n759.\\n\\n A\\n\\n The Herreras first argue that the state nuisance\\nproceeding does not fall into one of the NOPSI categories\\nbecause it is not “a civil enforcement proceeding[] that is\\nakin to a criminal prosecution.”\\n\\n The Supreme Court has held to the contrary, and\\nrecognized that a state nuisance proceeding may warrant\\nYounger abstention from federal claims. See Huffman v.\\nPursue, Ltd., 420 U.S. 592, 607 (1975). In Huffman, the\\nCourt considered the applicability of Younger abstention in\\n\\f12 HERRERA V. CITY OF PALMDALE\\n\\na federal § 1983 action where there was an ongoing state\\nnuisance enforcement proceeding against the manager of an\\nobscene film theater. Id. at 595–98. Explaining that “the\\nproceeding is both in aid of and closely related to criminal\\nstatutes which prohibit the dissemination of obscene\\nmaterial,” the Court determined that “[t]he propriety of\\nfederal-court interference with an Ohio nuisance proceeding\\nmust . . . be controlled by application of those same\\nconsiderations of comity and federalism” at issue when the\\nstate proceedings are criminal in nature. Id. at 604, 607.\\nThus, the Court held that the district court should have\\nabstained, unless an exception to Younger applied. Id. at\\n611.\\n\\n In NOPSI itself, the Supreme Court cited Huffman to\\nclarify that Younger abstention may apply “beyond state\\ncriminal prosecutions, to civil enforcement proceedings.”\\nNOPSI, 491 U.S. at 368; see also Woodfeathers, Inc. v.\\nWashington County, 180 F.3d 1017, 1021 (9th Cir. 1999)\\n(“Civil actions brought by a government entity to enforce\\nnuisance laws have been held to justify Younger\\nabstention.”). And the Court recently reaffirmed such\\napplication of Younger:\\n\\n Our decisions applying Younger to instances\\n of civil enforcement have generally\\n concerned state proceedings akin to a\\n criminal prosecution in important respects.\\n Such enforcement actions are\\n characteristically initiated to sanction the\\n federal plaintiff, i.e., the party challenging\\n the state action, for some wrongful act. In\\n cases of this genre, a state actor is routinely a\\n party to the state proceeding and often\\n initiates the action. Investigations are\\n\\f HERRERA V. CITY OF PALMDALE 13\\n\\n commonly involved, often culminating in the\\n filing of a formal complaint or charges.\\n\\nSprint, 571 U.S. at 79–80 (internal quotation marks and\\ncitations omitted).\\n\\n The nuisance action pending in state court against Bill\\nHerrera and Palmdale Lodging closely resembles the civil\\nenforcement actions described in Sprint. The City, a state\\nactor, obtained and executed an inspection warrant, and\\nidentified more than four hundred violations of State and\\nlocal laws on the motel property. Such investigation by the\\nCity is characteristic of the state actions that warrant\\nYounger abstention under Sprint. The City then issued a\\nNotice and Order to Repair or Abate the violations, and, as\\ndescribed in Sprint, “initiate[d]” an action for nuisance\\nabatement and receivership, alleging that the motel property\\n“contains numerous violations of State and local laws and\\nposes a severe life and health and safety hazard to any\\noccupants, nearby residents, and the public.” The state\\nnuisance complaint requested, among other forms of relief:\\nthe appointment of a receiver to take possession and control\\nof the property, an injunction preventing Bill Herrera and\\nPalmdale Lodging from collecting rent or income from the\\nproperty and from claiming any state tax deduction on the\\nproperty, and imposition of civil penalties against Bill\\nHerrera and Palmdale Lodging. In keeping with the\\nobjective of the enforcement actions described in Sprint,\\nsuch relief would “sanction” the Herreras for their alleged\\nfailure to comply with state and local laws. The\\ninvestigation, initiation, and requested sanctions of the\\nproceeding here are therefore consistent with the\\nenforcement actions described in Sprint and at issue in\\nHuffman.\\n\\f14 HERRERA V. CITY OF PALMDALE\\n\\n We are satisfied that the state nuisance enforcement\\naction brought by the City against Bill Herrera and Palmdale\\nLodging is a civil enforcement proceeding within the scope\\nof the Younger doctrine.\\n\\n B\\n\\n The Herreras next argue that, even if the state action falls\\nwithin one of the NOPSI categories, Younger abstention is\\nnonetheless inappropriate because two of the three\\nMiddlesex factors are not met.\\n\\n 1\\n\\n Initially, the Herreras argue that the state proceeding\\ndoes not implicate important state interests. The City\\ndisputes such argument, claiming that it has an important\\ninterest in eliminating public nuisances and enforcing local\\nand state codes to protect the public from dangerous\\nconditions. The state action sought to enforce health and\\nsafety provisions, and to abate public nuisances.\\n\\n We have previously held that such nuisance actions\\nimplicate important state interests and thus satisfy this\\nsecond Middlesex factor. See, e.g., Woodfeathers, 180 F.3d\\nat 1021 (holding that a state action enforcing ordinances\\ndeclaring waste to be a nuisance implicated important state\\ninterests); Potrero Hills Landfill, Inc. v. County of Solano,\\n657 F.3d 876, 884 (9th Cir. 2011) (listing “a wide range of\\ncivil contexts,” including nuisance abatement, in which\\nabstention was “necessary to protect the state’s unique\\ninterest in exercising its basic executive functions”). The\\nHerreras offer no meaningful distinction between these cases\\nand their own. Thus, the Herreras’ argument to the contrary\\nis foreclosed by our court’s precedent, and we conclude that\\n\\f HERRERA V. CITY OF PALMDALE 15\\n\\nthe state nuisance proceeding at issue here implicates\\nimportant state interests.\\n\\n 2\\n\\n Next, the Herreras argue that the state proceedings do not\\nprovide an adequate opportunity to raise federal\\nconstitutional claims, as required by Middlesex.\\nSpecifically, the Herreras argue that (1) their alleged civil\\nrights violations are “irrelevant” to the issue whether the\\nmotel is a public nuisance, and that (2) Mona Herrera and\\nher children are unable to raise their civil rights claims\\nbecause they are not parties to the state action. We address\\neach argument in turn.\\n\\n a\\n\\n A federal court’s exercise of Younger abstention does not\\nturn on whether the federal plaintiff actually avails himself\\nof the opportunity to present federal constitutional claims in\\nthe state proceeding, but rather whether such an opportunity\\nexists. See Juidice v. Vail, 430 U.S. 327, 337 (1977)\\n(explaining that plaintiffs “need be accorded only an\\nopportunity to fairly pursue their constitutional claims in the\\nongoing state proceedings, and their failure to avail\\nthemselves of such opportunities does not mean that the state\\nprocedures were inadequate” (citation omitted)). “[T]he\\nburden on this point rests on the federal plaintiff to show\\n‘that state procedural law barred presentation of [its]\\nclaims.’” Pennzoil Co. v. Texaco, Inc., 481 U.S. 1, 14 (1987)\\n(second alteration in original) (quoting Moore v. Sims, 442\\nU.S. 415, 432 (1979)); see also Lebbos v. Judges of Superior\\nCourt, 883 F.2d 810, 815 (9th Cir. 1989).\\n\\n The Herreras’ argument that the federal action is\\n“irrelevant” to the state action does not satisfy such burden,\\n\\f16 HERRERA V. CITY OF PALMDALE\\n\\nand the Herreras have pointed to no other reason why the\\nstate action defendants could not raise their federal\\nconstitutional claims in the enforcement proceeding. The\\nstate proceeding is a civil action in Los Angeles County\\nSuperior Court over which the state court has general\\njurisdiction. According to the California Code of Civil\\nProcedure, in such a proceeding, “[a] party against whom a\\ncause of action has been asserted in a complaint or cross-\\ncomplaint may file a cross-complaint setting forth . . . [a]ny\\ncause of action he has against any of the parties who filed\\nthe complaint or cross-complaint against him.” Cal. Code\\nCiv. Pro. § 428.10. The Herreras have not demonstrated that\\nstate procedural law otherwise barred such a cross-complaint\\nin the state action here.\\n\\n Thus, the Herreras’ first challenge to the adequacy of\\ntheir opportunity to raise federal constitutional claims in the\\nstate action must fail.\\n\\n b\\n\\n The Herreras also urge that the state proceedings do not\\nprovide an adequate opportunity to Mona and her children to\\nraise federal constitutional claims because they are not\\nparties to the state action.\\n\\n Younger abstention generally applies only where the\\nfederal plaintiffs are also defendants in the ongoing state\\nproceeding. See Benavidez v. Eu, 34 F.3d 825, 832 (9th Cir.\\n1994). However, both the Supreme Court and our court have\\nrecognized that “there plainly may be some circumstances in\\nwhich legally distinct parties are so closely related that they\\nshould all be subject to the Younger considerations which\\ngovern any one of them.” Doran v. Salem Inn, Inc., 422 U.S.\\n922, 928 (1975). In Hicks v. Miranda, 422 U.S. 332, 337–\\n38 (1975), the Court considered a federal challenge brought\\n\\f HERRERA V. CITY OF PALMDALE 17\\n\\nby a theater, its owner, and its employees to an ongoing state\\nenforcement proceeding of a California obscenity statute.\\nAlthough some of the federal plaintiffs were not parties to\\nthe state enforcement proceeding at the time the federal\\naction was filed, the Court nonetheless invoked Younger\\nabstention. Id. at 348. The Court explained that, because the\\nclaims of the federal plaintiffs would nonetheless interfere\\nwith the state case, the “same comity considerations\\nappl[ied]” to warrant abstention. Id. at 349 (quoting Allee v.\\nMedrano, 416 U.S. 802, 831 (1974) (Burger, C.J.,\\nconcurring)). Because the parties’ interests were\\n“intertwined” with those of the state court defendants,\\nYounger abstention was proper as to all federal plaintiffs. Id.\\nat 348.\\n\\n Likewise, our court has suggested that parties with “a\\nsufficiently close relationship or sufficiently intertwined\\ninterests” may be “treated similarly for purposes of Younger\\nabstention.” Canatella v. California, 404 F.3d 1106, 1116\\n(9th Cir. 2005). We do not stand alone. Several of our sister\\ncircuits have upheld a decision to abstain under Younger\\nwhere the parties to the federal and state actions were not\\nidentical, so long as their interests were sufficiently\\nintertwined. See, e.g., Spargo v. N.Y. State Comm’n on\\nJudicial Conduct, 351 F.3d 65, 81–82 (2d Cir. 2003)\\n(considering whether legal interests were “sufficiently\\nintertwined”); Cedar Rapids Cellular Tel., L.P. v. Miller,\\n280 F.3d 874, 881–82 (8th Cir. 2002) (considering whether\\ninterests are “closely related”); Collins v. Kendall County,\\n807 F.2d 95, 101 (7th Cir. 1986) (concluding that Younger\\nabstention considerations applied equally where plaintiffs\\nwere “all related to the same business entity” and “share[d]\\nthe same interest in contesting the state litigation”);\\nWomen’s Cmty. Health Ctr. of Beaumont, Inc. v. Tex. Health\\nFacilities Comm’n, 685 F.2d 974, 981–82 (5th Cir. 1982)\\n\\f18 HERRERA V. CITY OF PALMDALE\\n\\n(abstaining where interests were “so completely intertwined\\nthat the same Younger bar must apply to all”).\\n\\n The Herreras are all related to the same corporation,\\nPalmdale Lodging. In fact, Mona—though not a named\\ndefendant in the state action—is a co-founder of state-\\ndefendant Palmdale Lodging. And the children, with their\\nparents, reside at the motel which is the subject of the state\\naction. Furthermore, the federal claims arise from a single\\nproceeding to abate code violations at Palmdale Lodging’s\\nmotel and the corresponding investigation. Such\\nrelationship goes beyond identity of interests; rather, the\\nfamily members were allegedly deprived of their civil rights\\ncollectively during the investigation, and the ongoing\\nnuisance proceeding related to the motel which Bill and\\nMona operate and at which the family resides. The federal\\nclaims of Mona and her children present the same risk of\\ninterference in the state proceeding as do the federal claims\\nof Bill and Palmdale Lodging—indeed, all the federal\\nplaintiffs seek the same relief from the state court\\nproceedings. As in Hicks, the comity considerations raised\\nby the federal claims of those not party to the state action are\\nindistinguishable from those raised by the state defendants.\\nThe parties’ interests are therefore sufficiently intertwined.\\n\\n We are thus persuaded that the closely intertwined\\ninterests of Mona, the children, and the state defendants, Bill\\nand Palmdale Lodging, warrant subjecting them all to the\\nsame Younger abstention considerations. The Herreras’\\nsecond challenge to the adequacy of their opportunity to\\nraise constitutional claims in the state action also fails.\\n\\n C\\n\\n Finally, the Herreras argue that the exercise of Younger\\nabstention was improper because the relief sought in the\\n\\f HERRERA V. CITY OF PALMDALE 19\\n\\nfederal action would not “enjoin—or have the practical\\neffect of enjoining—ongoing state proceedings.”\\nReadyLink, 754 F.3d at 758. Specifically, the Herreras argue\\nthat the issues raised in the federal action are “wholly distinct\\nfrom the state court question.”\\n\\n 1\\n\\n Alleging violations of their civil rights under § 1983 and\\nthe FHA, the Herreras request injunctive, declaratory, and\\nmonetary relief. Certainly the Herreras’ request that the\\ncourt enjoin the City from closing the motel and evicting the\\nHerreras from their personal residence would enjoin directly\\nthe state action. Furthermore, the Supreme Court has “held\\nthat Younger applies to requests for declaratory relief\\nbecause ‘ordinarily a declaratory judgment will result in\\nprecisely the same interference with and disruption of state\\nproceedings that the longstanding policy limiting injunctions\\nwas designed to avoid.’” Gilbertson, 381 F.3d at 971\\n(quoting Samuels v. Mackell, 401 U.S. 66, 72 (1971)).\\nBecause the request for declaratory relief would have “the\\nsame practical impact as injunctive relief on a pending state\\nproceeding as a result of the preclusive effect of the federal\\ncourt judgment,” Gilbertson, 381 F.3d at 975, Younger\\nabstention is also appropriate as to such relief.\\n\\n 2\\n\\n The Herreras’ request for monetary relief is not so\\nstraightforward. Our court has recognized that “Younger\\nprinciples apply to actions at law . . . because a\\ndetermination that the federal plaintiff’s . . . rights have been\\nviolated would have the same practical effect as a\\ndeclaration or injunction on pending state proceedings.” Id.\\nat 968. “[T]o determine whether the federal plaintiff is\\nentitled to damages,” the district court often first decides\\n\\f20 HERRERA V. CITY OF PALMDALE\\n\\nwhether a violation of the plaintiff’s civil rights has\\noccurred, imposing the same intrusion as a declaratory\\njudgment by the federal court. Id. at 979–80.\\n\\n a\\n\\n We first consider the Herreras’ claims for damages under\\n§ 1983 for alleged violations of the First, Fifth, and\\nFourteenth Amendments and the Contract Clause of the\\nConstitution, and under the FHA—in other words, all of\\ntheir damages claims except those concerning the Fourth\\nAmendment. Relief on such claims requires the district\\ncourt to determine first whether violations of their civil\\nrights have occurred in the course of the state enforcement\\nproceeding, which would create a federal court judgment\\nwith preclusive effect over the ongoing state action. See\\nGilbertson, 381 F.3d at 978 (“Preclusion rules may be\\nrelevant to determining the practical effect of a federal\\ncourt’s relief.”). For example, a holding by the district court\\nthat the Herreras are entitled to damages under § 1983\\nbecause City officials violated their Fifth Amendment rights\\nwould include a determination that the state action\\nconstituted a taking of property without just compensation.\\nPlainly, such determination that the state proceeding is itself\\nunconstitutional would interfere with the ongoing state\\nenforcement action in the same way as would a declaratory\\njudgment by the federal court. Id. at 979–80. Thus, like the\\nclaims for declaratory relief, resolution of the claims for\\ndamages under § 1983 for violations of the First, Fifth, and\\nFourteenth Amendments and the Contract Clause of the\\nConstitution and the FHA would collectively “frustrate the\\nstate’s interest in administering its judicial system, cast a\\nnegative light on the state court’s ability to enforce\\nconstitutional principles, and put the federal court in the\\nposition of prematurely or unnecessarily deciding a question\\n\\f HERRERA V. CITY OF PALMDALE 21\\n\\nof federal constitutional law.” Id. at 980. Success by the\\nHerreras on such claims would invalidate the code\\nenforcement proceeding, and Younger abstention is\\ntherefore appropriate as to such claims.\\n\\n b\\n\\n Finally, we consider separately the Herreras’ claims for\\ndamages under § 1983 for alleged violations of the Fourth\\nAmendment. Cf. AmerisourceBergen Corp. v. Roden, 495\\nF.3d 1143, 1147 (9th Cir. 2007) (considering only whether\\nthe district court properly abstained under Younger from\\nCount I, where the parties conceded that the district court\\nproperly dismissed Count II); Joseph A. ex rel. Corrine\\nWolfe v. Ingram, 275 F.3d 1253, 1272 (10th Cir. 2002)\\n(considering federal claim seeking enforcement of numerous\\nprovisions of a consent decree and noting that “[a] provision-\\nby-provision Younger analysis appears prudent, however,\\nfor the fact that one provision may not be enforceable in light\\nof Younger does not necessarily warrant voiding the entire\\nconsent decree . . . or dismissing the entire action”).\\nAlthough the allegations of Fourth Amendment violations\\narise from investigations conducted in the course of the state\\nenforcement action, we fail to see how determinations on\\nsuch claims by the federal court would intrude in the ongoing\\nstate nuisance proceeding in the impermissible way that the\\nHerreras’ other damages claims would.\\n\\n The Fourth Amendment claims arise from the\\ndefendants’ search of the motel and subsequent entry onto\\nthe property to enforce the abatement proceedings, rather\\nthan from a challenge to the state proceeding as a whole or\\nthe state’s allegedly discriminatory motivation in initiating\\nsuch action. A ruling in favor of the Herreras on such claims\\nwould presumably not invalidate the basis for the code-\\nviolation enforcement proceedings, and the Fourth\\n\\f22 HERRERA V. CITY OF PALMDALE\\n\\nAmendment claims themselves are not at issue in such\\nproceedings. See AmerisourceBergen, 495 F.3d at 1151–52\\n(holding that Younger abstention from a breach of contract\\nclaim was improper where federal jurisdiction over the claim\\n“would not have enjoined or in any way impeded the\\nongoing litigation” in state court where there was no\\n“counterclaim in state court proceedings attempting to\\nenforce” such contract claim). We may not abstain based on\\na mere “potential for conflict.” Id. at 1151. Thus, unlike a\\ndetermination that the civil proceeding itself is\\nconstitutionally deficient, a determination that a Fourth\\nAmendment violation occurred and that the Herreras are\\nentitled to monetary damages would not “have the same\\npractical effect as a declaration or injunction on pending\\nstate proceedings.” Gilbertson, 381 F.3d at 968.\\n\\n We thus conclude that the district court erred in\\nabstaining from the § 1983 damages claim alleging\\nviolations of the Fourth Amendment. We recognize that\\nsuch decision raises the possibility of piecemeal litigation.\\nAlthough related to the same sequence of events, the relief\\nsought based on alleged Fourth Amendment violations\\nsimply does not meet our court’s requirement that the relief\\nhave the practical effect of enjoining the state court\\nproceeding. See ReadyLink, 754 F.3d at 758. Recalling our\\n“oblig[ation] to decide cases within the scope of federal\\njurisdiction,” Sprint, 571 U.S. at 72, this claim must be\\nsevered and the district court shall consider it on the merits\\nupon remand. 3\\n\\n\\n 3\\n The Herreras also argue that the district court erred in abstaining\\nbecause an exception to Younger abstention for bad faith or other\\nextraordinary circumstances applies. See Middlesex, 457 U.S. at 435;\\nSan Jose Silicon Valley Chamber of Commerce Political Action Comm.\\n\\f HERRERA V. CITY OF PALMDALE 23\\n\\n IV\\n\\n In sum, we are satisfied that the district court abstained\\nproperly in every aspect, except with respect to the allegedly\\nunreasonable search, which must be severed from the other\\nclaims. On remand, the district court should consider the\\nHerreras’ claims for damages under § 1983 for alleged\\nviolations of the Fourth Amendment. 4\\n\\n The order of the district court is AFFIRMED IN PART\\nAND REVERSED AND REMANDED IN PART. The\\nparties shall bear their own costs on appeal.\\n\\n\\n\\n\\nv. City of San Jose, 546 F.3d 1087, 1092 (9th Cir. 2008). However,\\nbecause the Herreras failed to raise this argument below, we consider\\nthis argument waived on appeal. See In re Mercury Interactive Corp.\\nSec. Litig., 618 F.3d 988, 992 (9th Cir. 2010).\\n 4\\n In light of our disposition in this case affirming on abstention\\ngrounds the district court’s decision as to nearly all the Herreras’ claims,\\nwe need not reach the City’s alternative argument that we should affirm\\nthe district court’s dismissal of the claims based on the Herreras’ alleged\\nfailure to state a claim of municipal liability under Monell v. Department\\nof Social Services of City of New York, 436 U.S. 658 (1978). On remand,\\nthe district court may consider such argument with respect to the\\nHerreras’ claim for damages under § 1983 for alleged violations of the\\nFourth Amendment.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4380660/\", \"author_raw\": \"Opinion by Judge O'Scannlain\"}]}"},"judge1":{"kind":"string","value":"DIARMUID F O'SCANNLAIN"},"judge2":{"kind":"string","value":"SANDRA S IKUTA"},"judge3":{"kind":"string","value":"GEORGE CAREM STEEH"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4603407/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":3,"string":"3"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: William Herrera v. City of Palmdale\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-03-20\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nWILLIAM ROBERT HERRERA; MONA No. 17-55761\nMOLINA HERRERA, Individually and\non behalf of their minor child N.P.H.; D.C. No.\nWILLIAM RYAN HERRERA; PALMDALE 2:16-cv-09453-\nLODGING, LLC, a California Limited MWF-FFM\nLiability Company,\n Plaintiffs-Appellants,\n OPINION\n v.\n\nCITY OF PALMDALE; COUNTY OF LOS\nANGELES; BUD DAVIS; GEORGE\nSCHNEIDER; ROB BRUCE; NARDY\nLOPEZ; MARK DYLER; ANNE\nAMBROSE; NOEL JAMES DURAN;\nJAMES PURTEE; SARA SHREVES;\nRAPOSAS; BLAKELY; ANTHONY\nBONELLI; MARK MILLER; MUNOZ;\nLEON; JACOBS; DANA; MYLES;\nSORROW; DIAZ; ARCIDIANCONO;\nDOLLENS; BRANDON; GALLAGHER;\nWALDEN; DOES, 1–10 inclusive,\n Defendants-Appellees.\n\n Appeal from the United States District Court\n for the Central District of California\n Michael W. Fitzgerald, District Judge, Presiding\n\f2 HERRERA V. CITY OF PALMDALE\n\n Argued and Submitted December 6, 2018\n Pasadena, California\n\n Filed March 20, 2019\n\n Before: Diarmuid F. O’Scannlain and Sandra S. Ikuta,\n Circuit Judges, and George Carem Steeh, * District Judge.\n\n Opinion by Judge O’Scannlain\n\n\n SUMMARY **\n\n\n Civil Rights\n\n The panel affirmed in part and reversed in part the\ndistrict court’s order granting defendants’ motions for\nabstention in an action brought pursuant to 42 U.S.C. § 1983\nand the Fair Housing Act alleging that the City of Palmdale\nand the County of Los Angeles committed numerous\nviolations in connection with assessing code violations on\nplaintiffs’ motel property and evicting plaintiffs and their\ntenants from the motel.\n\n When plaintiffs brought their suit in federal court, the\nCity almost simultaneously filed a complaint in state court\nasserting that plaintiffs’ hotel was a public nuisance and\nseeking the appointment of a receiver to take possession and\n\n *\n The Honorable George Caram Steeh III, United States District\nJudge for the Eastern District of Michigan, sitting by designation.\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f HERRERA V. CITY OF PALMDALE 3\n\ncontrol of the property. The district court found that\nabstention was proper under Younger v. Harris, 401 U.S. 37\n(1971), and dismissed the claims for declaratory and\ninjunctive relief, and stayed the claims for damages pending\nresolution of proceedings in the state action.\n\n The panel first held that it had jurisdiction over the\nappeal from the district court’s order granting the motions\nfor abstention even though the district court merely stayed\nrather than dismissed the damages claims. The panel held\nthat the state nuisance enforcement action brought by the\nCity (1) was a civil enforcement proceeding within the scope\nof the Younger doctrine; (2) implicated important state\ninterests; and (3) provided an adequate opportunity to raise\nthe federal constitutional claims. The panel further\nconcluded that because plaintiffs’ request for declaratory\nrelief in the district court would have the same practical\nimpact as injunctive relief on the pending state proceeding\nas a result of the preclusive effect of the federal court\njudgment, Younger abstention was also appropriate as to\nsuch relief.\n\n Addressing the claims for damages under § 1983 for\nviolations of the First, Fifth, and Fourteenth Amendments\nand the Contract Clause of the Constitution and the Fair\nHousing Act, the panel held that success by the plaintiffs on\nsuch claims would invalidate the code enforcement\nproceeding, and Younger abstention was therefore\nappropriate as to those claims.\n\n Finally, the panel held that the district court erred in\nabstaining from the § 1983 damages claim alleging\nviolations of the Fourth Amendment. The panel held that the\nrelief sought based on alleged Fourth Amendment violations\nsimply did not meet the Court’s requirement that the relief\n\f4 HERRERA V. CITY OF PALMDALE\n\nhave the practical effect of enjoining the state court\nproceeding. The panel stated that the Fourth Amendment\nclaim must be severed from the other claims and that the\ndistrict court should consider it on the merits upon remand.\n\n\n COUNSEL\n\nFrank Alan Weiser (argued), Law Offices of Frank A.\nWeiser, Los Angeles, California, for Plaintiffs-Appellants.\n\nJohn M. Fujii (argued) and Matthew R. Silver, Silver &\nWright LLP, Irvine, California, for Defendant-Appellee City\nof Palmdale.\n\nAshlee Clark (argued), Gilbert M. Nishimura, and Andrew\nCharles Pongracz, Seki, Nishimura & Watase, LLP, Los\nAngeles, California, for Defendant-Appellee County of Los\nAngeles.\n\f HERRERA V. CITY OF PALMDALE 5\n\n OPINION\n\nO’SCANNLAIN, Circuit Judge:\n\n We must decide whether a district court may abstain\nfrom addressing claims that seek federal relief while a\nsimultaneous action is ongoing in state court.\n\n I\n\n A\n\n William (“Bill”) Herrera and his wife, Mona Herrera,\noperate a 48-unit motel in Palmdale, California. The motel\nis owned by Palmdale Lodging, LLC, a privately owned\ncompany formed by Bill and Mona. The City of Palmdale\n(“City”) licensed Palmdale Lodging to operate the motel.\nAfter purchasing the motel in June 2016, Bill, Mona, and\nPalmdale Lodging spent approximately $250,000 to upgrade\nand to renovate the motel, and contracted with Motel 6 to\noperate the motel as a franchise. They also evicted those\nwhom they considered to be “problem tenants who had prior\nto their ownership and possession caused crime problems at\nthe motel.”\n\n On November 17, 2016, the City obtained a civil\ninspection warrant to investigate suspected violations of the\nPalmdale Municipal Code, the California Building\nStandards Code, the California Health and Safety Code, and\nother laws. The warrant was executed on November 21,\n2016. Bill and Mona allege that the inspection warrant was\nexecuted without notice and included a search of their\npersonal residence, located within the motel, without their\npermission. They further claim that sheriffs from the County\nof Los Angeles (“County”) held Bill and their two children\n\f6 HERRERA V. CITY OF PALMDALE\n\nat gunpoint for an hour and a half during the inspection of\ntheir personal residence.\n\n On December 13, 2016, the City issued a Notice and\nOrder to Repair or Abate (“Notice and Order”), which\nidentified more than 400 code violations on the motel\nproperty. The Notice and Order required the violations to be\nrepaired or abated within thirty days, and ordered the Herrera\nfamily and all motel tenants to vacate the property within\ntwo days.\n\n On December 15, 2016, the City and County closed the\nmotel and evicted the Herrera family and all motel tenants.\nBill and Mona contend that they were not afforded an\nopportunity to appeal the Notice and Order. Bill and Mona\nfurther contend that the City and County prevented them\nfrom doing “any work to upgrade or repair the motel for the\nalleged code violations” despite their requests to do so.\n\n B\n\n On December 21, 2016, Bill and Mona, individually and\non behalf of their minor daughter, their son William Ryan\nHerrera, and Palmdale Lodging (collectively, the\n“Herreras”) filed this federal civil rights action in the Central\nDistrict of California against the City, County, and\nindividual defendants, alleging violations under 42 U.S.C.\n§ 1983 and the Fair Housing Act (“FHA”), 42 U.S.C.\n§ 3601, et. seq. (“the federal action”). A First Amended\nComplaint alleged numerous federal constitutional\nviolations by the City, County, and City and County\nofficials, and a disparate impact claim under the FHA. The\nHerreras requested declaratory and injunctive relief,\nmonetary damages, and attorney’s fees.\n\f HERRERA V. CITY OF PALMDALE 7\n\n Almost simultaneously, the City filed a Nuisance\nComplaint in Los Angeles County Superior Court against\nBill Herrera and Palmdale Lodging (“the state action”). The\nCity sought, among other things, a declaration that the motel\nis a public nuisance, the appointment of a receiver to take\npossession and control of the property, and injunctive relief\nprohibiting the state-action defendants “from maintaining or\nallowing any public nuisances” and “requiring [them] to\nabate all violations of law” on the property.\n\n In the federal action, the City filed a motion for\nabstention, or, in the alternative, a motion to dismiss. The\nCounty filed a nearly identical motion the next day. The\nHerreras opposed both motions. On May 16, 2017, the\ndistrict court granted both motions, concluding that\nabstention was appropriate under Younger v. Harris, 401\nU.S. 37 (1971). The district court therefore dismissed the\nclaims for declaratory and injunctive relief in the federal\naction, and stayed the claims for damages pending resolution\nof proceedings in the state action.\n\n The Herreras timely appeal, challenging the district\ncourt’s order granting the motions for abstention. 1\n\n II\n\n As a threshold matter, the City argues that our court lacks\njurisdiction over the appeal from the district court order\ngranting the motions for abstention because the district court\n\n 1\n The City’s unopposed motion for judicial notice of the state court\ncomplaint, state court order appointing a receiver, and state court docket,\nand the County’s two unopposed motions for judicial notice of the state\naction defendants’ answer to the state court complaint and state action\ndefendants’ reservation of federal claims in the state action are\nGRANTED.\n\f8 HERRERA V. CITY OF PALMDALE\n\nhas not entered final judgment. Specifically, the City argues\nthat we are without jurisdiction over the appeal because the\ndistrict court has merely stayed, rather than dismissed, the\nHerreras’ damages claims, and thus it retains jurisdiction\nover such claims.\n\n A\n\n Under 28 U.S.C. § 1291, the Court of Appeals has\njurisdiction over appeals from “final decisions of the district\ncourts of the United States.” Generally, “[a] district court\norder abstaining under Younger and dismissing the case ends\nthe litigation. It is a final appealable order.” Confederated\nSalish v. Simonich, 29 F.3d 1398, 1401 (9th Cir. 1994).\nIndeed, when a court abstains under Younger, claims for\ninjunctive and declaratory relief are typically dismissed. See\nGilbertson v. Albright, 381 F.3d 965, 975, 981 (9th Cir.\n2004) (en banc). However, our court has also recognized\nthat, when a district court abstains from considering a\ndamages claim under Younger, it must stay—rather than\ndismiss—the damages action until state proceedings\nconclude. See id. at 984 (“[W]hen damages are at issue\nrather than discretionary relief, deference—rather than\ndismissal—is the proper restraint. To stay instead of to\ndismiss the federal action preserves the state’s interests in its\nown procedures, the federal plaintiff’s opportunity to seek\ncompensation in the forum of his choice, and an appropriate\nbalance of federal-state jurisdiction.”). Pursuant to this rule,\nthe district court dismissed the claims for declaratory and\ninjunctive relief, and stayed the claims for damages pending\nresolution of the state court proceedings.\n\n B\n\n “[A] stay is not ordinarily a final decision for purposes\nof § 1291.” Moses H. Cone Mem’l Hosp. v. Mercury Constr.\n\f HERRERA V. CITY OF PALMDALE 9\n\nCorp., 460 U.S. 1, 10 n.11 (1983). However, a stay order\nmay be “final for purposes of appellate jurisdiction” where\nthe order puts the litigant “effectively out of court.” Id. at 9;\nsee also Idlewild Liquor Corp. v. Epstein, 370 U.S. 713, 715\nn.2 (1962). This is precisely what the stay of the Herreras’\ndamages claims has effected here. “[T]he object of the stay\nis to require . . . an essential part of the federal suit to be\nlitigated in a state forum.” Moses H. Cone, 460 U.S. at 10\nn.11. Yet such stay is “lengthy and indefinite,” which\ncreates “a danger of denying justice by delay.” Blue Cross\n& Blue Shield of Ala. v. Unity Outpatient Surgery Ctr., Inc.,\n490 F.3d 718, 724 (9th Cir. 2007); see also id. at 723\n(“Moses H. Cone applies whenever there is a possibility that\nproceedings in another court could moot a suit or an issue,\neven if there is no guarantee that they will do so.”). 2\n\n\n\n 2\n We recognize that the district court has permitted the Herreras to\nreserve their stayed damages claims for litigation in federal court under\nEngland v. Louisiana State Board of Medical Examiners, 375 U.S. 411,\n421–22 (1964). England protects a “litigant who has properly invoked\nthe jurisdiction of a Federal District Court to consider federal\nconstitutional claims” from being compelled “to accept instead a state\ncourt’s determination of those claims.” Id. at 415. Although a district\ncourt order granting an England reservation of jurisdiction “is not a final\njudgment immediately appealable under 28 U.S.C. § 1291” because\n“[s]uch order does not end the litigation on the merits,” Confederated\nSalish, 29 F.3d at 1406, the England reservation is not at issue on appeal\nhere. The district court order granting Younger abstention is appealable\nunder Moses H. Cone, and we see no reason why the Herreras’\nreservation of claims should preclude appellate review of the otherwise\nappealable order granting Younger abstention. Cf. Confederated Salish,\n29 F.3d at 1406–07 (concluding that our court lacked jurisdiction over\nthe appeal of the order granting England reservation, but nonetheless\nexercising jurisdiction over the appeal of the order granting abstention\nunder Railroad Commission of Texas v. Pullman, 312 U.S. 496 (1941)).\nHere, the reservation simply provides insurance to the Herreras that they\n\f10 HERRERA V. CITY OF PALMDALE\n\n Under Moses H. Cone, even the stay—the resolution\nrequired by our court when granting Younger abstention on\ndamages claims—is effectively a final decision and thus the\ndistrict court order is final for purposes of appellate review.\nSee also Parris v. Taft, 630 F. App’x 895, 898 (11th Cir.\n2015) (“Because the district court stayed the case until Mr.\nParris’s related state criminal proceedings were resolved, the\norder left him effectively out of court. Thus, the order is a\nfinal decision, and we have appellate jurisdiction.”). We are\nsatisfied that we have jurisdiction to review the district court\norder on appeal.\n\n III\n\n Turning to the merits, the Herreras argue that the district\ncourt erred by abstaining under Younger.\n\n Younger abstention is grounded in a “longstanding\npublic policy against federal court interference with state\ncourt proceedings.” Younger, 401 U.S. at 43. The Supreme\nCourt has “identified two sources for this policy: the\nconstraints of equity jurisdiction and the concern for comity\nin our federal system.” Gilbertson, 381 F.3d at 970. Most\nimportantly, Younger abstention permits federal courts to\n“preserve respect for state functions such that the national\ngovernment protects federal rights and interests in a way that\nwill not ‘unduly interfere with the legitimate activities of the\nStates.’” Id. (quoting Younger, 401 U.S. at 44).\n\n A federal court may abstain under Younger in three\ncategories of cases: “(1) parallel, pending state criminal\nproceedings, (2) state civil proceedings that are akin to\n\nmay litigate their stayed damages claims after state court proceedings\nhave finished.\n\f HERRERA V. CITY OF PALMDALE 11\n\ncriminal prosecutions, and (3) state civil proceedings that\nimplicate a State’s interest in enforcing the orders and\njudgments of its courts.” ReadyLink Healthcare, Inc. v.\nState Comp. Ins. Fund, 754 F.3d 754, 759 (9th Cir. 2014)\n(internal quotation marks and citations omitted). First\nidentified in New Orleans Public Service, Inc. v. Council of\nNew Orleans (“NOPSI”), 491 U.S. 350 (1989), these three\ncategories are known as the NOPSI categories. See Sprint\nCommc’ns, Inc. v. Jacobs, 571 U.S. 69, 72–73 (2013).\n\n To warrant Younger abstention, a state civil action must\nfall into one of the NOPSI categories, and must also satisfy\na three-part inquiry: the state proceeding must be\n(1) “ongoing,” (2) “implicate important state interests,” and\n(3) provide “an adequate opportunity . . . to raise\nconstitutional challenges.” Middlesex Cty. Ethics Comm. v.\nGarden State Bar Ass’n, 457 U.S. 423, 432 (1982); see also\nReadyLink, 754 F.3d at 759. If the state proceeding falls into\none of the NOPSI categories and meets the three Middlesex\nfactors, a federal court may abstain under Younger so long\nas “the federal action would have the practical effect of\nenjoining the state proceedings.” ReadyLink, 754 F.3d at\n759.\n\n A\n\n The Herreras first argue that the state nuisance\nproceeding does not fall into one of the NOPSI categories\nbecause it is not “a civil enforcement proceeding[] that is\nakin to a criminal prosecution.”\n\n The Supreme Court has held to the contrary, and\nrecognized that a state nuisance proceeding may warrant\nYounger abstention from federal claims. See Huffman v.\nPursue, Ltd., 420 U.S. 592, 607 (1975). In Huffman, the\nCourt considered the applicability of Younger abstention in\n\f12 HERRERA V. CITY OF PALMDALE\n\na federal § 1983 action where there was an ongoing state\nnuisance enforcement proceeding against the manager of an\nobscene film theater. Id. at 595–98. Explaining that “the\nproceeding is both in aid of and closely related to criminal\nstatutes which prohibit the dissemination of obscene\nmaterial,” the Court determined that “[t]he propriety of\nfederal-court interference with an Ohio nuisance proceeding\nmust . . . be controlled by application of those same\nconsiderations of comity and federalism” at issue when the\nstate proceedings are criminal in nature. Id. at 604, 607.\nThus, the Court held that the district court should have\nabstained, unless an exception to Younger applied. Id. at\n611.\n\n In NOPSI itself, the Supreme Court cited Huffman to\nclarify that Younger abstention may apply “beyond state\ncriminal prosecutions, to civil enforcement proceedings.”\nNOPSI, 491 U.S. at 368; see also Woodfeathers, Inc. v.\nWashington County, 180 F.3d 1017, 1021 (9th Cir. 1999)\n(“Civil actions brought by a government entity to enforce\nnuisance laws have been held to justify Younger\nabstention.”). And the Court recently reaffirmed such\napplication of Younger:\n\n Our decisions applying Younger to instances\n of civil enforcement have generally\n concerned state proceedings akin to a\n criminal prosecution in important respects.\n Such enforcement actions are\n characteristically initiated to sanction the\n federal plaintiff, i.e., the party challenging\n the state action, for some wrongful act. In\n cases of this genre, a state actor is routinely a\n party to the state proceeding and often\n initiates the action. Investigations are\n\f HERRERA V. CITY OF PALMDALE 13\n\n commonly involved, often culminating in the\n filing of a formal complaint or charges.\n\nSprint, 571 U.S. at 79–80 (internal quotation marks and\ncitations omitted).\n\n The nuisance action pending in state court against Bill\nHerrera and Palmdale Lodging closely resembles the civil\nenforcement actions described in Sprint. The City, a state\nactor, obtained and executed an inspection warrant, and\nidentified more than four hundred violations of State and\nlocal laws on the motel property. Such investigation by the\nCity is characteristic of the state actions that warrant\nYounger abstention under Sprint. The City then issued a\nNotice and Order to Repair or Abate the violations, and, as\ndescribed in Sprint, “initiate[d]” an action for nuisance\nabatement and receivership, alleging that the motel property\n“contains numerous violations of State and local laws and\nposes a severe life and health and safety hazard to any\noccupants, nearby residents, and the public.” The state\nnuisance complaint requested, among other forms of relief:\nthe appointment of a receiver to take possession and control\nof the property, an injunction preventing Bill Herrera and\nPalmdale Lodging from collecting rent or income from the\nproperty and from claiming any state tax deduction on the\nproperty, and imposition of civil penalties against Bill\nHerrera and Palmdale Lodging. In keeping with the\nobjective of the enforcement actions described in Sprint,\nsuch relief would “sanction” the Herreras for their alleged\nfailure to comply with state and local laws. The\ninvestigation, initiation, and requested sanctions of the\nproceeding here are therefore consistent with the\nenforcement actions described in Sprint and at issue in\nHuffman.\n\f14 HERRERA V. CITY OF PALMDALE\n\n We are satisfied that the state nuisance enforcement\naction brought by the City against Bill Herrera and Palmdale\nLodging is a civil enforcement proceeding within the scope\nof the Younger doctrine.\n\n B\n\n The Herreras next argue that, even if the state action falls\nwithin one of the NOPSI categories, Younger abstention is\nnonetheless inappropriate because two of the three\nMiddlesex factors are not met.\n\n 1\n\n Initially, the Herreras argue that the state proceeding\ndoes not implicate important state interests. The City\ndisputes such argument, claiming that it has an important\ninterest in eliminating public nuisances and enforcing local\nand state codes to protect the public from dangerous\nconditions. The state action sought to enforce health and\nsafety provisions, and to abate public nuisances.\n\n We have previously held that such nuisance actions\nimplicate important state interests and thus satisfy this\nsecond Middlesex factor. See, e.g., Woodfeathers, 180 F.3d\nat 1021 (holding that a state action enforcing ordinances\ndeclaring waste to be a nuisance implicated important state\ninterests); Potrero Hills Landfill, Inc. v. County of Solano,\n657 F.3d 876, 884 (9th Cir. 2011) (listing “a wide range of\ncivil contexts,” including nuisance abatement, in which\nabstention was “necessary to protect the state’s unique\ninterest in exercising its basic executive functions”). The\nHerreras offer no meaningful distinction between these cases\nand their own. Thus, the Herreras’ argument to the contrary\nis foreclosed by our court’s precedent, and we conclude that\n\f HERRERA V. CITY OF PALMDALE 15\n\nthe state nuisance proceeding at issue here implicates\nimportant state interests.\n\n 2\n\n Next, the Herreras argue that the state proceedings do not\nprovide an adequate opportunity to raise federal\nconstitutional claims, as required by Middlesex.\nSpecifically, the Herreras argue that (1) their alleged civil\nrights violations are “irrelevant” to the issue whether the\nmotel is a public nuisance, and that (2) Mona Herrera and\nher children are unable to raise their civil rights claims\nbecause they are not parties to the state action. We address\neach argument in turn.\n\n a\n\n A federal court’s exercise of Younger abstention does not\nturn on whether the federal plaintiff actually avails himself\nof the opportunity to present federal constitutional claims in\nthe state proceeding, but rather whether such an opportunity\nexists. See Juidice v. Vail, 430 U.S. 327, 337 (1977)\n(explaining that plaintiffs “need be accorded only an\nopportunity to fairly pursue their constitutional claims in the\nongoing state proceedings, and their failure to avail\nthemselves of such opportunities does not mean that the state\nprocedures were inadequate” (citation omitted)). “[T]he\nburden on this point rests on the federal plaintiff to show\n‘that state procedural law barred presentation of [its]\nclaims.’” Pennzoil Co. v. Texaco, Inc., 481 U.S. 1, 14 (1987)\n(second alteration in original) (quoting Moore v. Sims, 442\nU.S. 415, 432 (1979)); see also Lebbos v. Judges of Superior\nCourt, 883 F.2d 810, 815 (9th Cir. 1989).\n\n The Herreras’ argument that the federal action is\n“irrelevant” to the state action does not satisfy such burden,\n\f16 HERRERA V. CITY OF PALMDALE\n\nand the Herreras have pointed to no other reason why the\nstate action defendants could not raise their federal\nconstitutional claims in the enforcement proceeding. The\nstate proceeding is a civil action in Los Angeles County\nSuperior Court over which the state court has general\njurisdiction. According to the California Code of Civil\nProcedure, in such a proceeding, “[a] party against whom a\ncause of action has been asserted in a complaint or cross-\ncomplaint may file a cross-complaint setting forth . . . [a]ny\ncause of action he has against any of the parties who filed\nthe complaint or cross-complaint against him.” Cal. Code\nCiv. Pro. § 428.10. The Herreras have not demonstrated that\nstate procedural law otherwise barred such a cross-complaint\nin the state action here.\n\n Thus, the Herreras’ first challenge to the adequacy of\ntheir opportunity to raise federal constitutional claims in the\nstate action must fail.\n\n b\n\n The Herreras also urge that the state proceedings do not\nprovide an adequate opportunity to Mona and her children to\nraise federal constitutional claims because they are not\nparties to the state action.\n\n Younger abstention generally applies only where the\nfederal plaintiffs are also defendants in the ongoing state\nproceeding. See Benavidez v. Eu, 34 F.3d 825, 832 (9th Cir.\n1994). However, both the Supreme Court and our court have\nrecognized that “there plainly may be some circumstances in\nwhich legally distinct parties are so closely related that they\nshould all be subject to the Younger considerations which\ngovern any one of them.” Doran v. Salem Inn, Inc., 422 U.S.\n922, 928 (1975). In Hicks v. Miranda, 422 U.S. 332, 337–\n38 (1975), the Court considered a federal challenge brought\n\f HERRERA V. CITY OF PALMDALE 17\n\nby a theater, its owner, and its employees to an ongoing state\nenforcement proceeding of a California obscenity statute.\nAlthough some of the federal plaintiffs were not parties to\nthe state enforcement proceeding at the time the federal\naction was filed, the Court nonetheless invoked Younger\nabstention. Id. at 348. The Court explained that, because the\nclaims of the federal plaintiffs would nonetheless interfere\nwith the state case, the “same comity considerations\nappl[ied]” to warrant abstention. Id. at 349 (quoting Allee v.\nMedrano, 416 U.S. 802, 831 (1974) (Burger, C.J.,\nconcurring)). Because the parties’ interests were\n“intertwined” with those of the state court defendants,\nYounger abstention was proper as to all federal plaintiffs. Id.\nat 348.\n\n Likewise, our court has suggested that parties with “a\nsufficiently close relationship or sufficiently intertwined\ninterests” may be “treated similarly for purposes of Younger\nabstention.” Canatella v. California, 404 F.3d 1106, 1116\n(9th Cir. 2005). We do not stand alone. Several of our sister\ncircuits have upheld a decision to abstain under Younger\nwhere the parties to the federal and state actions were not\nidentical, so long as their interests were sufficiently\nintertwined. See, e.g., Spargo v. N.Y. State Comm’n on\nJudicial Conduct, 351 F.3d 65, 81–82 (2d Cir. 2003)\n(considering whether legal interests were “sufficiently\nintertwined”); Cedar Rapids Cellular Tel., L.P. v. Miller,\n280 F.3d 874, 881–82 (8th Cir. 2002) (considering whether\ninterests are “closely related”); Collins v. Kendall County,\n807 F.2d 95, 101 (7th Cir. 1986) (concluding that Younger\nabstention considerations applied equally where plaintiffs\nwere “all related to the same business entity” and “share[d]\nthe same interest in contesting the state litigation”);\nWomen’s Cmty. Health Ctr. of Beaumont, Inc. v. Tex. Health\nFacilities Comm’n, 685 F.2d 974, 981–82 (5th Cir. 1982)\n\f18 HERRERA V. CITY OF PALMDALE\n\n(abstaining where interests were “so completely intertwined\nthat the same Younger bar must apply to all”).\n\n The Herreras are all related to the same corporation,\nPalmdale Lodging. In fact, Mona—though not a named\ndefendant in the state action—is a co-founder of state-\ndefendant Palmdale Lodging. And the children, with their\nparents, reside at the motel which is the subject of the state\naction. Furthermore, the federal claims arise from a single\nproceeding to abate code violations at Palmdale Lodging’s\nmotel and the corresponding investigation. Such\nrelationship goes beyond identity of interests; rather, the\nfamily members were allegedly deprived of their civil rights\ncollectively during the investigation, and the ongoing\nnuisance proceeding related to the motel which Bill and\nMona operate and at which the family resides. The federal\nclaims of Mona and her children present the same risk of\ninterference in the state proceeding as do the federal claims\nof Bill and Palmdale Lodging—indeed, all the federal\nplaintiffs seek the same relief from the state court\nproceedings. As in Hicks, the comity considerations raised\nby the federal claims of those not party to the state action are\nindistinguishable from those raised by the state defendants.\nThe parties’ interests are therefore sufficiently intertwined.\n\n We are thus persuaded that the closely intertwined\ninterests of Mona, the children, and the state defendants, Bill\nand Palmdale Lodging, warrant subjecting them all to the\nsame Younger abstention considerations. The Herreras’\nsecond challenge to the adequacy of their opportunity to\nraise constitutional claims in the state action also fails.\n\n C\n\n Finally, the Herreras argue that the exercise of Younger\nabstention was improper because the relief sought in the\n\f HERRERA V. CITY OF PALMDALE 19\n\nfederal action would not “enjoin—or have the practical\neffect of enjoining—ongoing state proceedings.”\nReadyLink, 754 F.3d at 758. Specifically, the Herreras argue\nthat the issues raised in the federal action are “wholly distinct\nfrom the state court question.”\n\n 1\n\n Alleging violations of their civil rights under § 1983 and\nthe FHA, the Herreras request injunctive, declaratory, and\nmonetary relief. Certainly the Herreras’ request that the\ncourt enjoin the City from closing the motel and evicting the\nHerreras from their personal residence would enjoin directly\nthe state action. Furthermore, the Supreme Court has “held\nthat Younger applies to requests for declaratory relief\nbecause ‘ordinarily a declaratory judgment will result in\nprecisely the same interference with and disruption of state\nproceedings that the longstanding policy limiting injunctions\nwas designed to avoid.’” Gilbertson, 381 F.3d at 971\n(quoting Samuels v. Mackell, 401 U.S. 66, 72 (1971)).\nBecause the request for declaratory relief would have “the\nsame practical impact as injunctive relief on a pending state\nproceeding as a result of the preclusive effect of the federal\ncourt judgment,” Gilbertson, 381 F.3d at 975, Younger\nabstention is also appropriate as to such relief.\n\n 2\n\n The Herreras’ request for monetary relief is not so\nstraightforward. Our court has recognized that “Younger\nprinciples apply to actions at law . . . because a\ndetermination that the federal plaintiff’s . . . rights have been\nviolated would have the same practical effect as a\ndeclaration or injunction on pending state proceedings.” Id.\nat 968. “[T]o determine whether the federal plaintiff is\nentitled to damages,” the district court often first decides\n\f20 HERRERA V. CITY OF PALMDALE\n\nwhether a violation of the plaintiff’s civil rights has\noccurred, imposing the same intrusion as a declaratory\njudgment by the federal court. Id. at 979–80.\n\n a\n\n We first consider the Herreras’ claims for damages under\n§ 1983 for alleged violations of the First, Fifth, and\nFourteenth Amendments and the Contract Clause of the\nConstitution, and under the FHA—in other words, all of\ntheir damages claims except those concerning the Fourth\nAmendment. Relief on such claims requires the district\ncourt to determine first whether violations of their civil\nrights have occurred in the course of the state enforcement\nproceeding, which would create a federal court judgment\nwith preclusive effect over the ongoing state action. See\nGilbertson, 381 F.3d at 978 (“Preclusion rules may be\nrelevant to determining the practical effect of a federal\ncourt’s relief.”). For example, a holding by the district court\nthat the Herreras are entitled to damages under § 1983\nbecause City officials violated their Fifth Amendment rights\nwould include a determination that the state action\nconstituted a taking of property without just compensation.\nPlainly, such determination that the state proceeding is itself\nunconstitutional would interfere with the ongoing state\nenforcement action in the same way as would a declaratory\njudgment by the federal court. Id. at 979–80. Thus, like the\nclaims for declaratory relief, resolution of the claims for\ndamages under § 1983 for violations of the First, Fifth, and\nFourteenth Amendments and the Contract Clause of the\nConstitution and the FHA would collectively “frustrate the\nstate’s interest in administering its judicial system, cast a\nnegative light on the state court’s ability to enforce\nconstitutional principles, and put the federal court in the\nposition of prematurely or unnecessarily deciding a question\n\f HERRERA V. CITY OF PALMDALE 21\n\nof federal constitutional law.” Id. at 980. Success by the\nHerreras on such claims would invalidate the code\nenforcement proceeding, and Younger abstention is\ntherefore appropriate as to such claims.\n\n b\n\n Finally, we consider separately the Herreras’ claims for\ndamages under § 1983 for alleged violations of the Fourth\nAmendment. Cf. AmerisourceBergen Corp. v. Roden, 495\nF.3d 1143, 1147 (9th Cir. 2007) (considering only whether\nthe district court properly abstained under Younger from\nCount I, where the parties conceded that the district court\nproperly dismissed Count II); Joseph A. ex rel. Corrine\nWolfe v. Ingram, 275 F.3d 1253, 1272 (10th Cir. 2002)\n(considering federal claim seeking enforcement of numerous\nprovisions of a consent decree and noting that “[a] provision-\nby-provision Younger analysis appears prudent, however,\nfor the fact that one provision may not be enforceable in light\nof Younger does not necessarily warrant voiding the entire\nconsent decree . . . or dismissing the entire action”).\nAlthough the allegations of Fourth Amendment violations\narise from investigations conducted in the course of the state\nenforcement action, we fail to see how determinations on\nsuch claims by the federal court would intrude in the ongoing\nstate nuisance proceeding in the impermissible way that the\nHerreras’ other damages claims would.\n\n The Fourth Amendment claims arise from the\ndefendants’ search of the motel and subsequent entry onto\nthe property to enforce the abatement proceedings, rather\nthan from a challenge to the state proceeding as a whole or\nthe state’s allegedly discriminatory motivation in initiating\nsuch action. A ruling in favor of the Herreras on such claims\nwould presumably not invalidate the basis for the code-\nviolation enforcement proceedings, and the Fourth\n\f22 HERRERA V. CITY OF PALMDALE\n\nAmendment claims themselves are not at issue in such\nproceedings. See AmerisourceBergen, 495 F.3d at 1151–52\n(holding that Younger abstention from a breach of contract\nclaim was improper where federal jurisdiction over the claim\n“would not have enjoined or in any way impeded the\nongoing litigation” in state court where there was no\n“counterclaim in state court proceedings attempting to\nenforce” such contract claim). We may not abstain based on\na mere “potential for conflict.” Id. at 1151. Thus, unlike a\ndetermination that the civil proceeding itself is\nconstitutionally deficient, a determination that a Fourth\nAmendment violation occurred and that the Herreras are\nentitled to monetary damages would not “have the same\npractical effect as a declaration or injunction on pending\nstate proceedings.” Gilbertson, 381 F.3d at 968.\n\n We thus conclude that the district court erred in\nabstaining from the § 1983 damages claim alleging\nviolations of the Fourth Amendment. We recognize that\nsuch decision raises the possibility of piecemeal litigation.\nAlthough related to the same sequence of events, the relief\nsought based on alleged Fourth Amendment violations\nsimply does not meet our court’s requirement that the relief\nhave the practical effect of enjoining the state court\nproceeding. See ReadyLink, 754 F.3d at 758. Recalling our\n“oblig[ation] to decide cases within the scope of federal\njurisdiction,” Sprint, 571 U.S. at 72, this claim must be\nsevered and the district court shall consider it on the merits\nupon remand. 3\n\n\n 3\n The Herreras also argue that the district court erred in abstaining\nbecause an exception to Younger abstention for bad faith or other\nextraordinary circumstances applies. See Middlesex, 457 U.S. at 435;\nSan Jose Silicon Valley Chamber of Commerce Political Action Comm.\n\f HERRERA V. CITY OF PALMDALE 23\n\n IV\n\n In sum, we are satisfied that the district court abstained\nproperly in every aspect, except with respect to the allegedly\nunreasonable search, which must be severed from the other\nclaims. On remand, the district court should consider the\nHerreras’ claims for damages under § 1983 for alleged\nviolations of the Fourth Amendment. 4\n\n The order of the district court is AFFIRMED IN PART\nAND REVERSED AND REMANDED IN PART. The\nparties shall bear their own costs on appeal.\n\n\n\n\nv. City of San Jose, 546 F.3d 1087, 1092 (9th Cir. 2008). However,\nbecause the Herreras failed to raise this argument below, we consider\nthis argument waived on appeal. See In re Mercury Interactive Corp.\nSec. Litig., 618 F.3d 988, 992 (9th Cir. 2010).\n 4\n In light of our disposition in this case affirming on abstention\ngrounds the district court’s decision as to nearly all the Herreras’ claims,\nwe need not reach the City’s alternative argument that we should affirm\nthe district court’s dismissal of the claims based on the Herreras’ alleged\nfailure to state a claim of municipal liability under Monell v. Department\nof Social Services of City of New York, 436 U.S. 658 (1978). On remand,\nthe district court may consider such argument with respect to the\nHerreras’ claim for damages under § 1983 for alleged violations of the\nFourth Amendment.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: William Herrera v. City of Palmdale\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-03-20\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nWILLIAM ROBERT HERRERA; MONA No. 17-55761\\nMOLINA HERRERA, Individually and\\non behalf of their minor child N.P.H.; D.C. No.\\nWILLIAM RYAN HERRERA; PALMDALE 2:16-cv-09453-\\nLODGING, LLC, a California Limited MWF-FFM\\nLiability Company,\\n Plaintiffs-Appellants,\\n OPINION\\n v.\\n\\nCITY OF PALMDALE; COUNTY OF LOS\\nANGELES; BUD DAVIS; GEORGE\\nSCHNEIDER; ROB BRUCE; NARDY\\nLOPEZ; MARK DYLER; ANNE\\nAMBROSE; NOEL JAMES DURAN;\\nJAMES PURTEE; SARA SHREVES;\\nRAPOSAS; BLAKELY; ANTHONY\\nBONELLI; MARK MILLER; MUNOZ;\\nLEON; JACOBS; DANA; MYLES;\\nSORROW; DIAZ; ARCIDIANCONO;\\nDOLLENS; BRANDON; GALLAGHER;\\nWALDEN; DOES, 1–10 inclusive,\\n Defendants-Appellees.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Michael W. Fitzgerald, District Judge, Presiding\\n\\f2 HERRERA V. CITY OF PALMDALE\\n\\n Argued and Submitted December 6, 2018\\n Pasadena, California\\n\\n Filed March 20, 2019\\n\\n Before: Diarmuid F. O’Scannlain and Sandra S. Ikuta,\\n Circuit Judges, and George Carem Steeh, * District Judge.\\n\\n Opinion by Judge O’Scannlain\\n\\n\\n SUMMARY **\\n\\n\\n Civil Rights\\n\\n The panel affirmed in part and reversed in part the\\ndistrict court’s order granting defendants’ motions for\\nabstention in an action brought pursuant to 42 U.S.C. § 1983\\nand the Fair Housing Act alleging that the City of Palmdale\\nand the County of Los Angeles committed numerous\\nviolations in connection with assessing code violations on\\nplaintiffs’ motel property and evicting plaintiffs and their\\ntenants from the motel.\\n\\n When plaintiffs brought their suit in federal court, the\\nCity almost simultaneously filed a complaint in state court\\nasserting that plaintiffs’ hotel was a public nuisance and\\nseeking the appointment of a receiver to take possession and\\n\\n *\\n The Honorable George Caram Steeh III, United States District\\nJudge for the Eastern District of Michigan, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f HERRERA V. CITY OF PALMDALE 3\\n\\ncontrol of the property. The district court found that\\nabstention was proper under Younger v. Harris, 401 U.S. 37\\n(1971), and dismissed the claims for declaratory and\\ninjunctive relief, and stayed the claims for damages pending\\nresolution of proceedings in the state action.\\n\\n The panel first held that it had jurisdiction over the\\nappeal from the district court’s order granting the motions\\nfor abstention even though the district court merely stayed\\nrather than dismissed the damages claims. The panel held\\nthat the state nuisance enforcement action brought by the\\nCity (1) was a civil enforcement proceeding within the scope\\nof the Younger doctrine; (2) implicated important state\\ninterests; and (3) provided an adequate opportunity to raise\\nthe federal constitutional claims. The panel further\\nconcluded that because plaintiffs’ request for declaratory\\nrelief in the district court would have the same practical\\nimpact as injunctive relief on the pending state proceeding\\nas a result of the preclusive effect of the federal court\\njudgment, Younger abstention was also appropriate as to\\nsuch relief.\\n\\n Addressing the claims for damages under § 1983 for\\nviolations of the First, Fifth, and Fourteenth Amendments\\nand the Contract Clause of the Constitution and the Fair\\nHousing Act, the panel held that success by the plaintiffs on\\nsuch claims would invalidate the code enforcement\\nproceeding, and Younger abstention was therefore\\nappropriate as to those claims.\\n\\n Finally, the panel held that the district court erred in\\nabstaining from the § 1983 damages claim alleging\\nviolations of the Fourth Amendment. The panel held that the\\nrelief sought based on alleged Fourth Amendment violations\\nsimply did not meet the Court’s requirement that the relief\\n\\f4 HERRERA V. CITY OF PALMDALE\\n\\nhave the practical effect of enjoining the state court\\nproceeding. The panel stated that the Fourth Amendment\\nclaim must be severed from the other claims and that the\\ndistrict court should consider it on the merits upon remand.\\n\\n\\n COUNSEL\\n\\nFrank Alan Weiser (argued), Law Offices of Frank A.\\nWeiser, Los Angeles, California, for Plaintiffs-Appellants.\\n\\nJohn M. Fujii (argued) and Matthew R. Silver, Silver &\\nWright LLP, Irvine, California, for Defendant-Appellee City\\nof Palmdale.\\n\\nAshlee Clark (argued), Gilbert M. Nishimura, and Andrew\\nCharles Pongracz, Seki, Nishimura & Watase, LLP, Los\\nAngeles, California, for Defendant-Appellee County of Los\\nAngeles.\\n\\f HERRERA V. CITY OF PALMDALE 5\\n\\n OPINION\\n\\nO’SCANNLAIN, Circuit Judge:\\n\\n We must decide whether a district court may abstain\\nfrom addressing claims that seek federal relief while a\\nsimultaneous action is ongoing in state court.\\n\\n I\\n\\n A\\n\\n William (“Bill”) Herrera and his wife, Mona Herrera,\\noperate a 48-unit motel in Palmdale, California. The motel\\nis owned by Palmdale Lodging, LLC, a privately owned\\ncompany formed by Bill and Mona. The City of Palmdale\\n(“City”) licensed Palmdale Lodging to operate the motel.\\nAfter purchasing the motel in June 2016, Bill, Mona, and\\nPalmdale Lodging spent approximately $250,000 to upgrade\\nand to renovate the motel, and contracted with Motel 6 to\\noperate the motel as a franchise. They also evicted those\\nwhom they considered to be “problem tenants who had prior\\nto their ownership and possession caused crime problems at\\nthe motel.”\\n\\n On November 17, 2016, the City obtained a civil\\ninspection warrant to investigate suspected violations of the\\nPalmdale Municipal Code, the California Building\\nStandards Code, the California Health and Safety Code, and\\nother laws. The warrant was executed on November 21,\\n2016. Bill and Mona allege that the inspection warrant was\\nexecuted without notice and included a search of their\\npersonal residence, located within the motel, without their\\npermission. They further claim that sheriffs from the County\\nof Los Angeles (“County”) held Bill and their two children\\n\\f6 HERRERA V. CITY OF PALMDALE\\n\\nat gunpoint for an hour and a half during the inspection of\\ntheir personal residence.\\n\\n On December 13, 2016, the City issued a Notice and\\nOrder to Repair or Abate (“Notice and Order”), which\\nidentified more than 400 code violations on the motel\\nproperty. The Notice and Order required the violations to be\\nrepaired or abated within thirty days, and ordered the Herrera\\nfamily and all motel tenants to vacate the property within\\ntwo days.\\n\\n On December 15, 2016, the City and County closed the\\nmotel and evicted the Herrera family and all motel tenants.\\nBill and Mona contend that they were not afforded an\\nopportunity to appeal the Notice and Order. Bill and Mona\\nfurther contend that the City and County prevented them\\nfrom doing “any work to upgrade or repair the motel for the\\nalleged code violations” despite their requests to do so.\\n\\n B\\n\\n On December 21, 2016, Bill and Mona, individually and\\non behalf of their minor daughter, their son William Ryan\\nHerrera, and Palmdale Lodging (collectively, the\\n“Herreras”) filed this federal civil rights action in the Central\\nDistrict of California against the City, County, and\\nindividual defendants, alleging violations under 42 U.S.C.\\n§ 1983 and the Fair Housing Act (“FHA”), 42 U.S.C.\\n§ 3601, et. seq. (“the federal action”). A First Amended\\nComplaint alleged numerous federal constitutional\\nviolations by the City, County, and City and County\\nofficials, and a disparate impact claim under the FHA. The\\nHerreras requested declaratory and injunctive relief,\\nmonetary damages, and attorney’s fees.\\n\\f HERRERA V. CITY OF PALMDALE 7\\n\\n Almost simultaneously, the City filed a Nuisance\\nComplaint in Los Angeles County Superior Court against\\nBill Herrera and Palmdale Lodging (“the state action”). The\\nCity sought, among other things, a declaration that the motel\\nis a public nuisance, the appointment of a receiver to take\\npossession and control of the property, and injunctive relief\\nprohibiting the state-action defendants “from maintaining or\\nallowing any public nuisances” and “requiring [them] to\\nabate all violations of law” on the property.\\n\\n In the federal action, the City filed a motion for\\nabstention, or, in the alternative, a motion to dismiss. The\\nCounty filed a nearly identical motion the next day. The\\nHerreras opposed both motions. On May 16, 2017, the\\ndistrict court granted both motions, concluding that\\nabstention was appropriate under Younger v. Harris, 401\\nU.S. 37 (1971). The district court therefore dismissed the\\nclaims for declaratory and injunctive relief in the federal\\naction, and stayed the claims for damages pending resolution\\nof proceedings in the state action.\\n\\n The Herreras timely appeal, challenging the district\\ncourt’s order granting the motions for abstention. 1\\n\\n II\\n\\n As a threshold matter, the City argues that our court lacks\\njurisdiction over the appeal from the district court order\\ngranting the motions for abstention because the district court\\n\\n 1\\n The City’s unopposed motion for judicial notice of the state court\\ncomplaint, state court order appointing a receiver, and state court docket,\\nand the County’s two unopposed motions for judicial notice of the state\\naction defendants’ answer to the state court complaint and state action\\ndefendants’ reservation of federal claims in the state action are\\nGRANTED.\\n\\f8 HERRERA V. CITY OF PALMDALE\\n\\nhas not entered final judgment. Specifically, the City argues\\nthat we are without jurisdiction over the appeal because the\\ndistrict court has merely stayed, rather than dismissed, the\\nHerreras’ damages claims, and thus it retains jurisdiction\\nover such claims.\\n\\n A\\n\\n Under 28 U.S.C. § 1291, the Court of Appeals has\\njurisdiction over appeals from “final decisions of the district\\ncourts of the United States.” Generally, “[a] district court\\norder abstaining under Younger and dismissing the case ends\\nthe litigation. It is a final appealable order.” Confederated\\nSalish v. Simonich, 29 F.3d 1398, 1401 (9th Cir. 1994).\\nIndeed, when a court abstains under Younger, claims for\\ninjunctive and declaratory relief are typically dismissed. See\\nGilbertson v. Albright, 381 F.3d 965, 975, 981 (9th Cir.\\n2004) (en banc). However, our court has also recognized\\nthat, when a district court abstains from considering a\\ndamages claim under Younger, it must stay—rather than\\ndismiss—the damages action until state proceedings\\nconclude. See id. at 984 (“[W]hen damages are at issue\\nrather than discretionary relief, deference—rather than\\ndismissal—is the proper restraint. To stay instead of to\\ndismiss the federal action preserves the state’s interests in its\\nown procedures, the federal plaintiff’s opportunity to seek\\ncompensation in the forum of his choice, and an appropriate\\nbalance of federal-state jurisdiction.”). Pursuant to this rule,\\nthe district court dismissed the claims for declaratory and\\ninjunctive relief, and stayed the claims for damages pending\\nresolution of the state court proceedings.\\n\\n B\\n\\n “[A] stay is not ordinarily a final decision for purposes\\nof § 1291.” Moses H. Cone Mem’l Hosp. v. Mercury Constr.\\n\\f HERRERA V. CITY OF PALMDALE 9\\n\\nCorp., 460 U.S. 1, 10 n.11 (1983). However, a stay order\\nmay be “final for purposes of appellate jurisdiction” where\\nthe order puts the litigant “effectively out of court.” Id. at 9;\\nsee also Idlewild Liquor Corp. v. Epstein, 370 U.S. 713, 715\\nn.2 (1962). This is precisely what the stay of the Herreras’\\ndamages claims has effected here. “[T]he object of the stay\\nis to require . . . an essential part of the federal suit to be\\nlitigated in a state forum.” Moses H. Cone, 460 U.S. at 10\\nn.11. Yet such stay is “lengthy and indefinite,” which\\ncreates “a danger of denying justice by delay.” Blue Cross\\n& Blue Shield of Ala. v. Unity Outpatient Surgery Ctr., Inc.,\\n490 F.3d 718, 724 (9th Cir. 2007); see also id. at 723\\n(“Moses H. Cone applies whenever there is a possibility that\\nproceedings in another court could moot a suit or an issue,\\neven if there is no guarantee that they will do so.”). 2\\n\\n\\n\\n 2\\n We recognize that the district court has permitted the Herreras to\\nreserve their stayed damages claims for litigation in federal court under\\nEngland v. Louisiana State Board of Medical Examiners, 375 U.S. 411,\\n421–22 (1964). England protects a “litigant who has properly invoked\\nthe jurisdiction of a Federal District Court to consider federal\\nconstitutional claims” from being compelled “to accept instead a state\\ncourt’s determination of those claims.” Id. at 415. Although a district\\ncourt order granting an England reservation of jurisdiction “is not a final\\njudgment immediately appealable under 28 U.S.C. § 1291” because\\n“[s]uch order does not end the litigation on the merits,” Confederated\\nSalish, 29 F.3d at 1406, the England reservation is not at issue on appeal\\nhere. The district court order granting Younger abstention is appealable\\nunder Moses H. Cone, and we see no reason why the Herreras’\\nreservation of claims should preclude appellate review of the otherwise\\nappealable order granting Younger abstention. Cf. Confederated Salish,\\n29 F.3d at 1406–07 (concluding that our court lacked jurisdiction over\\nthe appeal of the order granting England reservation, but nonetheless\\nexercising jurisdiction over the appeal of the order granting abstention\\nunder Railroad Commission of Texas v. Pullman, 312 U.S. 496 (1941)).\\nHere, the reservation simply provides insurance to the Herreras that they\\n\\f10 HERRERA V. CITY OF PALMDALE\\n\\n Under Moses H. Cone, even the stay—the resolution\\nrequired by our court when granting Younger abstention on\\ndamages claims—is effectively a final decision and thus the\\ndistrict court order is final for purposes of appellate review.\\nSee also Parris v. Taft, 630 F. App’x 895, 898 (11th Cir.\\n2015) (“Because the district court stayed the case until Mr.\\nParris’s related state criminal proceedings were resolved, the\\norder left him effectively out of court. Thus, the order is a\\nfinal decision, and we have appellate jurisdiction.”). We are\\nsatisfied that we have jurisdiction to review the district court\\norder on appeal.\\n\\n III\\n\\n Turning to the merits, the Herreras argue that the district\\ncourt erred by abstaining under Younger.\\n\\n Younger abstention is grounded in a “longstanding\\npublic policy against federal court interference with state\\ncourt proceedings.” Younger, 401 U.S. at 43. The Supreme\\nCourt has “identified two sources for this policy: the\\nconstraints of equity jurisdiction and the concern for comity\\nin our federal system.” Gilbertson, 381 F.3d at 970. Most\\nimportantly, Younger abstention permits federal courts to\\n“preserve respect for state functions such that the national\\ngovernment protects federal rights and interests in a way that\\nwill not ‘unduly interfere with the legitimate activities of the\\nStates.’” Id. (quoting Younger, 401 U.S. at 44).\\n\\n A federal court may abstain under Younger in three\\ncategories of cases: “(1) parallel, pending state criminal\\nproceedings, (2) state civil proceedings that are akin to\\n\\nmay litigate their stayed damages claims after state court proceedings\\nhave finished.\\n\\f HERRERA V. CITY OF PALMDALE 11\\n\\ncriminal prosecutions, and (3) state civil proceedings that\\nimplicate a State’s interest in enforcing the orders and\\njudgments of its courts.” ReadyLink Healthcare, Inc. v.\\nState Comp. Ins. Fund, 754 F.3d 754, 759 (9th Cir. 2014)\\n(internal quotation marks and citations omitted). First\\nidentified in New Orleans Public Service, Inc. v. Council of\\nNew Orleans (“NOPSI”), 491 U.S. 350 (1989), these three\\ncategories are known as the NOPSI categories. See Sprint\\nCommc’ns, Inc. v. Jacobs, 571 U.S. 69, 72–73 (2013).\\n\\n To warrant Younger abstention, a state civil action must\\nfall into one of the NOPSI categories, and must also satisfy\\na three-part inquiry: the state proceeding must be\\n(1) “ongoing,” (2) “implicate important state interests,” and\\n(3) provide “an adequate opportunity . . . to raise\\nconstitutional challenges.” Middlesex Cty. Ethics Comm. v.\\nGarden State Bar Ass’n, 457 U.S. 423, 432 (1982); see also\\nReadyLink, 754 F.3d at 759. If the state proceeding falls into\\none of the NOPSI categories and meets the three Middlesex\\nfactors, a federal court may abstain under Younger so long\\nas “the federal action would have the practical effect of\\nenjoining the state proceedings.” ReadyLink, 754 F.3d at\\n759.\\n\\n A\\n\\n The Herreras first argue that the state nuisance\\nproceeding does not fall into one of the NOPSI categories\\nbecause it is not “a civil enforcement proceeding[] that is\\nakin to a criminal prosecution.”\\n\\n The Supreme Court has held to the contrary, and\\nrecognized that a state nuisance proceeding may warrant\\nYounger abstention from federal claims. See Huffman v.\\nPursue, Ltd., 420 U.S. 592, 607 (1975). In Huffman, the\\nCourt considered the applicability of Younger abstention in\\n\\f12 HERRERA V. CITY OF PALMDALE\\n\\na federal § 1983 action where there was an ongoing state\\nnuisance enforcement proceeding against the manager of an\\nobscene film theater. Id. at 595–98. Explaining that “the\\nproceeding is both in aid of and closely related to criminal\\nstatutes which prohibit the dissemination of obscene\\nmaterial,” the Court determined that “[t]he propriety of\\nfederal-court interference with an Ohio nuisance proceeding\\nmust . . . be controlled by application of those same\\nconsiderations of comity and federalism” at issue when the\\nstate proceedings are criminal in nature. Id. at 604, 607.\\nThus, the Court held that the district court should have\\nabstained, unless an exception to Younger applied. Id. at\\n611.\\n\\n In NOPSI itself, the Supreme Court cited Huffman to\\nclarify that Younger abstention may apply “beyond state\\ncriminal prosecutions, to civil enforcement proceedings.”\\nNOPSI, 491 U.S. at 368; see also Woodfeathers, Inc. v.\\nWashington County, 180 F.3d 1017, 1021 (9th Cir. 1999)\\n(“Civil actions brought by a government entity to enforce\\nnuisance laws have been held to justify Younger\\nabstention.”). And the Court recently reaffirmed such\\napplication of Younger:\\n\\n Our decisions applying Younger to instances\\n of civil enforcement have generally\\n concerned state proceedings akin to a\\n criminal prosecution in important respects.\\n Such enforcement actions are\\n characteristically initiated to sanction the\\n federal plaintiff, i.e., the party challenging\\n the state action, for some wrongful act. In\\n cases of this genre, a state actor is routinely a\\n party to the state proceeding and often\\n initiates the action. Investigations are\\n\\f HERRERA V. CITY OF PALMDALE 13\\n\\n commonly involved, often culminating in the\\n filing of a formal complaint or charges.\\n\\nSprint, 571 U.S. at 79–80 (internal quotation marks and\\ncitations omitted).\\n\\n The nuisance action pending in state court against Bill\\nHerrera and Palmdale Lodging closely resembles the civil\\nenforcement actions described in Sprint. The City, a state\\nactor, obtained and executed an inspection warrant, and\\nidentified more than four hundred violations of State and\\nlocal laws on the motel property. Such investigation by the\\nCity is characteristic of the state actions that warrant\\nYounger abstention under Sprint. The City then issued a\\nNotice and Order to Repair or Abate the violations, and, as\\ndescribed in Sprint, “initiate[d]” an action for nuisance\\nabatement and receivership, alleging that the motel property\\n“contains numerous violations of State and local laws and\\nposes a severe life and health and safety hazard to any\\noccupants, nearby residents, and the public.” The state\\nnuisance complaint requested, among other forms of relief:\\nthe appointment of a receiver to take possession and control\\nof the property, an injunction preventing Bill Herrera and\\nPalmdale Lodging from collecting rent or income from the\\nproperty and from claiming any state tax deduction on the\\nproperty, and imposition of civil penalties against Bill\\nHerrera and Palmdale Lodging. In keeping with the\\nobjective of the enforcement actions described in Sprint,\\nsuch relief would “sanction” the Herreras for their alleged\\nfailure to comply with state and local laws. The\\ninvestigation, initiation, and requested sanctions of the\\nproceeding here are therefore consistent with the\\nenforcement actions described in Sprint and at issue in\\nHuffman.\\n\\f14 HERRERA V. CITY OF PALMDALE\\n\\n We are satisfied that the state nuisance enforcement\\naction brought by the City against Bill Herrera and Palmdale\\nLodging is a civil enforcement proceeding within the scope\\nof the Younger doctrine.\\n\\n B\\n\\n The Herreras next argue that, even if the state action falls\\nwithin one of the NOPSI categories, Younger abstention is\\nnonetheless inappropriate because two of the three\\nMiddlesex factors are not met.\\n\\n 1\\n\\n Initially, the Herreras argue that the state proceeding\\ndoes not implicate important state interests. The City\\ndisputes such argument, claiming that it has an important\\ninterest in eliminating public nuisances and enforcing local\\nand state codes to protect the public from dangerous\\nconditions. The state action sought to enforce health and\\nsafety provisions, and to abate public nuisances.\\n\\n We have previously held that such nuisance actions\\nimplicate important state interests and thus satisfy this\\nsecond Middlesex factor. See, e.g., Woodfeathers, 180 F.3d\\nat 1021 (holding that a state action enforcing ordinances\\ndeclaring waste to be a nuisance implicated important state\\ninterests); Potrero Hills Landfill, Inc. v. County of Solano,\\n657 F.3d 876, 884 (9th Cir. 2011) (listing “a wide range of\\ncivil contexts,” including nuisance abatement, in which\\nabstention was “necessary to protect the state’s unique\\ninterest in exercising its basic executive functions”). The\\nHerreras offer no meaningful distinction between these cases\\nand their own. Thus, the Herreras’ argument to the contrary\\nis foreclosed by our court’s precedent, and we conclude that\\n\\f HERRERA V. CITY OF PALMDALE 15\\n\\nthe state nuisance proceeding at issue here implicates\\nimportant state interests.\\n\\n 2\\n\\n Next, the Herreras argue that the state proceedings do not\\nprovide an adequate opportunity to raise federal\\nconstitutional claims, as required by Middlesex.\\nSpecifically, the Herreras argue that (1) their alleged civil\\nrights violations are “irrelevant” to the issue whether the\\nmotel is a public nuisance, and that (2) Mona Herrera and\\nher children are unable to raise their civil rights claims\\nbecause they are not parties to the state action. We address\\neach argument in turn.\\n\\n a\\n\\n A federal court’s exercise of Younger abstention does not\\nturn on whether the federal plaintiff actually avails himself\\nof the opportunity to present federal constitutional claims in\\nthe state proceeding, but rather whether such an opportunity\\nexists. See Juidice v. Vail, 430 U.S. 327, 337 (1977)\\n(explaining that plaintiffs “need be accorded only an\\nopportunity to fairly pursue their constitutional claims in the\\nongoing state proceedings, and their failure to avail\\nthemselves of such opportunities does not mean that the state\\nprocedures were inadequate” (citation omitted)). “[T]he\\nburden on this point rests on the federal plaintiff to show\\n‘that state procedural law barred presentation of [its]\\nclaims.’” Pennzoil Co. v. Texaco, Inc., 481 U.S. 1, 14 (1987)\\n(second alteration in original) (quoting Moore v. Sims, 442\\nU.S. 415, 432 (1979)); see also Lebbos v. Judges of Superior\\nCourt, 883 F.2d 810, 815 (9th Cir. 1989).\\n\\n The Herreras’ argument that the federal action is\\n“irrelevant” to the state action does not satisfy such burden,\\n\\f16 HERRERA V. CITY OF PALMDALE\\n\\nand the Herreras have pointed to no other reason why the\\nstate action defendants could not raise their federal\\nconstitutional claims in the enforcement proceeding. The\\nstate proceeding is a civil action in Los Angeles County\\nSuperior Court over which the state court has general\\njurisdiction. According to the California Code of Civil\\nProcedure, in such a proceeding, “[a] party against whom a\\ncause of action has been asserted in a complaint or cross-\\ncomplaint may file a cross-complaint setting forth . . . [a]ny\\ncause of action he has against any of the parties who filed\\nthe complaint or cross-complaint against him.” Cal. Code\\nCiv. Pro. § 428.10. The Herreras have not demonstrated that\\nstate procedural law otherwise barred such a cross-complaint\\nin the state action here.\\n\\n Thus, the Herreras’ first challenge to the adequacy of\\ntheir opportunity to raise federal constitutional claims in the\\nstate action must fail.\\n\\n b\\n\\n The Herreras also urge that the state proceedings do not\\nprovide an adequate opportunity to Mona and her children to\\nraise federal constitutional claims because they are not\\nparties to the state action.\\n\\n Younger abstention generally applies only where the\\nfederal plaintiffs are also defendants in the ongoing state\\nproceeding. See Benavidez v. Eu, 34 F.3d 825, 832 (9th Cir.\\n1994). However, both the Supreme Court and our court have\\nrecognized that “there plainly may be some circumstances in\\nwhich legally distinct parties are so closely related that they\\nshould all be subject to the Younger considerations which\\ngovern any one of them.” Doran v. Salem Inn, Inc., 422 U.S.\\n922, 928 (1975). In Hicks v. Miranda, 422 U.S. 332, 337–\\n38 (1975), the Court considered a federal challenge brought\\n\\f HERRERA V. CITY OF PALMDALE 17\\n\\nby a theater, its owner, and its employees to an ongoing state\\nenforcement proceeding of a California obscenity statute.\\nAlthough some of the federal plaintiffs were not parties to\\nthe state enforcement proceeding at the time the federal\\naction was filed, the Court nonetheless invoked Younger\\nabstention. Id. at 348. The Court explained that, because the\\nclaims of the federal plaintiffs would nonetheless interfere\\nwith the state case, the “same comity considerations\\nappl[ied]” to warrant abstention. Id. at 349 (quoting Allee v.\\nMedrano, 416 U.S. 802, 831 (1974) (Burger, C.J.,\\nconcurring)). Because the parties’ interests were\\n“intertwined” with those of the state court defendants,\\nYounger abstention was proper as to all federal plaintiffs. Id.\\nat 348.\\n\\n Likewise, our court has suggested that parties with “a\\nsufficiently close relationship or sufficiently intertwined\\ninterests” may be “treated similarly for purposes of Younger\\nabstention.” Canatella v. California, 404 F.3d 1106, 1116\\n(9th Cir. 2005). We do not stand alone. Several of our sister\\ncircuits have upheld a decision to abstain under Younger\\nwhere the parties to the federal and state actions were not\\nidentical, so long as their interests were sufficiently\\nintertwined. See, e.g., Spargo v. N.Y. State Comm’n on\\nJudicial Conduct, 351 F.3d 65, 81–82 (2d Cir. 2003)\\n(considering whether legal interests were “sufficiently\\nintertwined”); Cedar Rapids Cellular Tel., L.P. v. Miller,\\n280 F.3d 874, 881–82 (8th Cir. 2002) (considering whether\\ninterests are “closely related”); Collins v. Kendall County,\\n807 F.2d 95, 101 (7th Cir. 1986) (concluding that Younger\\nabstention considerations applied equally where plaintiffs\\nwere “all related to the same business entity” and “share[d]\\nthe same interest in contesting the state litigation”);\\nWomen’s Cmty. Health Ctr. of Beaumont, Inc. v. Tex. Health\\nFacilities Comm’n, 685 F.2d 974, 981–82 (5th Cir. 1982)\\n\\f18 HERRERA V. CITY OF PALMDALE\\n\\n(abstaining where interests were “so completely intertwined\\nthat the same Younger bar must apply to all”).\\n\\n The Herreras are all related to the same corporation,\\nPalmdale Lodging. In fact, Mona—though not a named\\ndefendant in the state action—is a co-founder of state-\\ndefendant Palmdale Lodging. And the children, with their\\nparents, reside at the motel which is the subject of the state\\naction. Furthermore, the federal claims arise from a single\\nproceeding to abate code violations at Palmdale Lodging’s\\nmotel and the corresponding investigation. Such\\nrelationship goes beyond identity of interests; rather, the\\nfamily members were allegedly deprived of their civil rights\\ncollectively during the investigation, and the ongoing\\nnuisance proceeding related to the motel which Bill and\\nMona operate and at which the family resides. The federal\\nclaims of Mona and her children present the same risk of\\ninterference in the state proceeding as do the federal claims\\nof Bill and Palmdale Lodging—indeed, all the federal\\nplaintiffs seek the same relief from the state court\\nproceedings. As in Hicks, the comity considerations raised\\nby the federal claims of those not party to the state action are\\nindistinguishable from those raised by the state defendants.\\nThe parties’ interests are therefore sufficiently intertwined.\\n\\n We are thus persuaded that the closely intertwined\\ninterests of Mona, the children, and the state defendants, Bill\\nand Palmdale Lodging, warrant subjecting them all to the\\nsame Younger abstention considerations. The Herreras’\\nsecond challenge to the adequacy of their opportunity to\\nraise constitutional claims in the state action also fails.\\n\\n C\\n\\n Finally, the Herreras argue that the exercise of Younger\\nabstention was improper because the relief sought in the\\n\\f HERRERA V. CITY OF PALMDALE 19\\n\\nfederal action would not “enjoin—or have the practical\\neffect of enjoining—ongoing state proceedings.”\\nReadyLink, 754 F.3d at 758. Specifically, the Herreras argue\\nthat the issues raised in the federal action are “wholly distinct\\nfrom the state court question.”\\n\\n 1\\n\\n Alleging violations of their civil rights under § 1983 and\\nthe FHA, the Herreras request injunctive, declaratory, and\\nmonetary relief. Certainly the Herreras’ request that the\\ncourt enjoin the City from closing the motel and evicting the\\nHerreras from their personal residence would enjoin directly\\nthe state action. Furthermore, the Supreme Court has “held\\nthat Younger applies to requests for declaratory relief\\nbecause ‘ordinarily a declaratory judgment will result in\\nprecisely the same interference with and disruption of state\\nproceedings that the longstanding policy limiting injunctions\\nwas designed to avoid.’” Gilbertson, 381 F.3d at 971\\n(quoting Samuels v. Mackell, 401 U.S. 66, 72 (1971)).\\nBecause the request for declaratory relief would have “the\\nsame practical impact as injunctive relief on a pending state\\nproceeding as a result of the preclusive effect of the federal\\ncourt judgment,” Gilbertson, 381 F.3d at 975, Younger\\nabstention is also appropriate as to such relief.\\n\\n 2\\n\\n The Herreras’ request for monetary relief is not so\\nstraightforward. Our court has recognized that “Younger\\nprinciples apply to actions at law . . . because a\\ndetermination that the federal plaintiff’s . . . rights have been\\nviolated would have the same practical effect as a\\ndeclaration or injunction on pending state proceedings.” Id.\\nat 968. “[T]o determine whether the federal plaintiff is\\nentitled to damages,” the district court often first decides\\n\\f20 HERRERA V. CITY OF PALMDALE\\n\\nwhether a violation of the plaintiff’s civil rights has\\noccurred, imposing the same intrusion as a declaratory\\njudgment by the federal court. Id. at 979–80.\\n\\n a\\n\\n We first consider the Herreras’ claims for damages under\\n§ 1983 for alleged violations of the First, Fifth, and\\nFourteenth Amendments and the Contract Clause of the\\nConstitution, and under the FHA—in other words, all of\\ntheir damages claims except those concerning the Fourth\\nAmendment. Relief on such claims requires the district\\ncourt to determine first whether violations of their civil\\nrights have occurred in the course of the state enforcement\\nproceeding, which would create a federal court judgment\\nwith preclusive effect over the ongoing state action. See\\nGilbertson, 381 F.3d at 978 (“Preclusion rules may be\\nrelevant to determining the practical effect of a federal\\ncourt’s relief.”). For example, a holding by the district court\\nthat the Herreras are entitled to damages under § 1983\\nbecause City officials violated their Fifth Amendment rights\\nwould include a determination that the state action\\nconstituted a taking of property without just compensation.\\nPlainly, such determination that the state proceeding is itself\\nunconstitutional would interfere with the ongoing state\\nenforcement action in the same way as would a declaratory\\njudgment by the federal court. Id. at 979–80. Thus, like the\\nclaims for declaratory relief, resolution of the claims for\\ndamages under § 1983 for violations of the First, Fifth, and\\nFourteenth Amendments and the Contract Clause of the\\nConstitution and the FHA would collectively “frustrate the\\nstate’s interest in administering its judicial system, cast a\\nnegative light on the state court’s ability to enforce\\nconstitutional principles, and put the federal court in the\\nposition of prematurely or unnecessarily deciding a question\\n\\f HERRERA V. CITY OF PALMDALE 21\\n\\nof federal constitutional law.” Id. at 980. Success by the\\nHerreras on such claims would invalidate the code\\nenforcement proceeding, and Younger abstention is\\ntherefore appropriate as to such claims.\\n\\n b\\n\\n Finally, we consider separately the Herreras’ claims for\\ndamages under § 1983 for alleged violations of the Fourth\\nAmendment. Cf. AmerisourceBergen Corp. v. Roden, 495\\nF.3d 1143, 1147 (9th Cir. 2007) (considering only whether\\nthe district court properly abstained under Younger from\\nCount I, where the parties conceded that the district court\\nproperly dismissed Count II); Joseph A. ex rel. Corrine\\nWolfe v. Ingram, 275 F.3d 1253, 1272 (10th Cir. 2002)\\n(considering federal claim seeking enforcement of numerous\\nprovisions of a consent decree and noting that “[a] provision-\\nby-provision Younger analysis appears prudent, however,\\nfor the fact that one provision may not be enforceable in light\\nof Younger does not necessarily warrant voiding the entire\\nconsent decree . . . or dismissing the entire action”).\\nAlthough the allegations of Fourth Amendment violations\\narise from investigations conducted in the course of the state\\nenforcement action, we fail to see how determinations on\\nsuch claims by the federal court would intrude in the ongoing\\nstate nuisance proceeding in the impermissible way that the\\nHerreras’ other damages claims would.\\n\\n The Fourth Amendment claims arise from the\\ndefendants’ search of the motel and subsequent entry onto\\nthe property to enforce the abatement proceedings, rather\\nthan from a challenge to the state proceeding as a whole or\\nthe state’s allegedly discriminatory motivation in initiating\\nsuch action. A ruling in favor of the Herreras on such claims\\nwould presumably not invalidate the basis for the code-\\nviolation enforcement proceedings, and the Fourth\\n\\f22 HERRERA V. CITY OF PALMDALE\\n\\nAmendment claims themselves are not at issue in such\\nproceedings. See AmerisourceBergen, 495 F.3d at 1151–52\\n(holding that Younger abstention from a breach of contract\\nclaim was improper where federal jurisdiction over the claim\\n“would not have enjoined or in any way impeded the\\nongoing litigation” in state court where there was no\\n“counterclaim in state court proceedings attempting to\\nenforce” such contract claim). We may not abstain based on\\na mere “potential for conflict.” Id. at 1151. Thus, unlike a\\ndetermination that the civil proceeding itself is\\nconstitutionally deficient, a determination that a Fourth\\nAmendment violation occurred and that the Herreras are\\nentitled to monetary damages would not “have the same\\npractical effect as a declaration or injunction on pending\\nstate proceedings.” Gilbertson, 381 F.3d at 968.\\n\\n We thus conclude that the district court erred in\\nabstaining from the § 1983 damages claim alleging\\nviolations of the Fourth Amendment. We recognize that\\nsuch decision raises the possibility of piecemeal litigation.\\nAlthough related to the same sequence of events, the relief\\nsought based on alleged Fourth Amendment violations\\nsimply does not meet our court’s requirement that the relief\\nhave the practical effect of enjoining the state court\\nproceeding. See ReadyLink, 754 F.3d at 758. Recalling our\\n“oblig[ation] to decide cases within the scope of federal\\njurisdiction,” Sprint, 571 U.S. at 72, this claim must be\\nsevered and the district court shall consider it on the merits\\nupon remand. 3\\n\\n\\n 3\\n The Herreras also argue that the district court erred in abstaining\\nbecause an exception to Younger abstention for bad faith or other\\nextraordinary circumstances applies. See Middlesex, 457 U.S. at 435;\\nSan Jose Silicon Valley Chamber of Commerce Political Action Comm.\\n\\f HERRERA V. CITY OF PALMDALE 23\\n\\n IV\\n\\n In sum, we are satisfied that the district court abstained\\nproperly in every aspect, except with respect to the allegedly\\nunreasonable search, which must be severed from the other\\nclaims. On remand, the district court should consider the\\nHerreras’ claims for damages under § 1983 for alleged\\nviolations of the Fourth Amendment. 4\\n\\n The order of the district court is AFFIRMED IN PART\\nAND REVERSED AND REMANDED IN PART. The\\nparties shall bear their own costs on appeal.\\n\\n\\n\\n\\nv. City of San Jose, 546 F.3d 1087, 1092 (9th Cir. 2008). However,\\nbecause the Herreras failed to raise this argument below, we consider\\nthis argument waived on appeal. See In re Mercury Interactive Corp.\\nSec. Litig., 618 F.3d 988, 992 (9th Cir. 2010).\\n 4\\n In light of our disposition in this case affirming on abstention\\ngrounds the district court’s decision as to nearly all the Herreras’ claims,\\nwe need not reach the City’s alternative argument that we should affirm\\nthe district court’s dismissal of the claims based on the Herreras’ alleged\\nfailure to state a claim of municipal liability under Monell v. Department\\nof Social Services of City of New York, 436 U.S. 658 (1978). On remand,\\nthe district court may consider such argument with respect to the\\nHerreras’ claim for damages under § 1983 for alleged violations of the\\nFourth Amendment.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}}],"truncated":false,"partial":false},"paginationData":{"pageIndex":12,"numItemsPerPage":100,"numTotalItems":31175,"offset":1200,"length":100}},"jwt":"eyJhbGciOiJFZERTQSJ9.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sImlhdCI6MTc3ODc4NDk1OCwic3ViIjoiL2RhdGFzZXRzL2VpdGFucy1wdS9jbC1jb2RpbmctaW5wdXQiLCJleHAiOjE3Nzg3ODg1NTgsImlzcyI6Imh0dHBzOi8vaHVnZ2luZ2ZhY2UuY28ifQ.IX4MHk3y90J8f0BOls-D-WaDcXO_fnJWXZeQhJxrVh7tjTz-TA0Wbyd-SNaGMzI4qwId_tuV7wYGv9ONSaM0BQ","displayUrls":true,"splitSizeSummaries":[{"config":"default","split":"train","numRows":31175,"numBytesParquet":1629387251}]},"discussionsStats":{"closed":0,"open":1,"total":1},"fullWidth":true,"hasGatedAccess":true,"hasFullAccess":true,"isEmbedded":false,"savedQueries":{"community":[],"user":[]}}">
id int64 | name string | name_abbreviation string | decision_date string | docket_number string | court string | casebody string | judge1 string | judge2 string | judge3 string | exactly_three_caps int64 | extra_judges string | judge_designations string | opinions_count int64 | majority_count int64 | dissent_count int64 | concurrence_count int64 | other_count int64 | primary_citation string | reporter string | volume null | resource_url string | precedential_status string | 010combined_count int64 | 020lead_count int64 | 030concurrence_count int64 | 035concurrenceinpart_count int64 | 040dissent_count int64 | year int64 | month int64 | messages list |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
4,600,971 | Rafael Herrera-Garcia v. William P. Barr | 2019-03-18 | 18-3196 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Before KANNE, ROVNER, and BARRETT, Circuit Judges.", "parties": "", "opinions": [{"author": "BARRETT, Circuit Judge", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________\nNos. 18-1511 & 18-3196\nRAFAEL ... | KANNE | ROVNER | BARRETT | 1 | {"KANNE": ", Circuit", "ROVNER": ", Circuit", "BARRETT": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4600971/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,600,972 | Elena Hernandez v. Marque Medicos Fullerton, LLC | 2019-03-18 | 18-1789 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Before WOOD, Chief Judge, and SYKES and SCUDDER, Circuit Judges.", "parties": "", "opinions": [{"author": "Diane S. Sykes", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________\nNo. 18-1789\nIN RE:... | WOOD | SYKES | SCUDDER | 1 | {"WOOD": ", Chief", "SYKES": ", Circuit", "SCUDDER": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4600972/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,600,986 | Elena Hernandez v. Marque Medicos Fullerton, LLC | 2019-03-18 | 18-1789 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Before WOOD, Chief Judge, and SYKES and SCUDDER, Circuit Judges.", "parties": "", "opinions": [{"author": "Diane S. Sykes", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________\nNo. 18-1789\nIN RE:... | WOOD | SYKES | SCUDDER | 1 | {"WOOD": ", Chief", "SYKES": ", Circuit", "SCUDDER": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4600986/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,601,075 | Rafael Herrera-Garcia v. William P. Barr | 2019-03-18 | 18-1511 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Before KANNE, ROVNER, and BARRETT, Circuit Judges.", "parties": "", "opinions": [{"author": "BARRETT, Circuit Judge", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________\nNos. 18-1511 & 18-3196\nRAFAEL ... | KANNE | ROVNER | BARRETT | 1 | {"KANNE": ", Circuit", "ROVNER": ", Circuit", "BARRETT": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4601075/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
8,443,933 | IN RE: Elena HERNANDEZ, Debtor-Appellant. | In re Hernandez | 2019-03-18 | No. 18-1789 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Scudder, Sykes, Wood", "parties": "", "opinions": [{"author": "", "type": "020lead", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opinions/8415364/", "author_raw": ""}]} | SCUDDER | SYKES | WOOD | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/8443933/ | Published | 0 | 1 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,601,074 | Ja'Lin WILLIAMS, Plaintiff-Appellant, v. NORFOLK SOUTHERN CORPORATION and Norfolk Southern Railway Company, Defendants-Appellees. | Ja'Lin Williams v. Norfolk Southern Corporation | 2019-03-19 | 18-2517 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Flaum, Barrett, Scudder", "parties": "", "opinions": [{"author": "BARRETT, Circuit Judge", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________\nNo. 18-2517\nJA’LIN WILLIAMS,\n ... | FLAUM | BARRETT | SCUDDER | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4601074/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,601,253 | UNITED STATES of America, Plaintiff-Appellee, v. Donald WANJIKU, Defendant-Appellant. | United States v. Donald Wanjiku | 2019-03-19 | 18-1973 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Rovner, Sykes, Barrett", "parties": "", "opinions": [{"author": "Ilana Kara Diamond Rovner", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\nNo. 18-1973\n\nUNITED STATES OF AMERICA,\n Pl... | ROVNER | SYKES | BARRETT | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4601253/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,601,254 | Ja'Lin Williams v. Norfolk Southern Corporation | 2019-03-19 | 18-2517 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Before FLAUM, BARRETT, and SCUDDER, Circuit Judges.", "parties": "", "opinions": [{"author": "BARRETT, Circuit Judge", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________\nNo. 18-2517\nJA’LIN WILLIAMS... | FLAUM | BARRETT | SCUDDER | 1 | {"FLAUM": ", Circuit", "BARRETT": ", Circuit", "SCUDDER": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4601254/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,601,308 | United States v. Donald Wanjiku | 2019-03-19 | 18-1973 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Before ROVNER, SYKES, and BARRETT, Circuit Judges.", "parties": "", "opinions": [{"author": "Ilana Kara Diamond Rovner", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\nNo. 18-1973\n\nUNITED STATES OF AMERICA,\n ... | ROVNER | SYKES | BARRETT | 1 | {"ROVNER": ", Circuit", "SYKES": ", Circuit", "BARRETT": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4601308/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,601,833 | UNITED STATES of America, Plaintiff-Appellee, v. Andres GARCIA, Defendant-Appellant. | United States v. Andres Garcia | 2019-03-20 | 18-1735 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Rovner, Hamilton, Brennan", "parties": "", "opinions": [{"author": "", "type": "020lead", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opinions/9824722/", "author_raw": ""}, {"author": "", "type": "040dissent", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/... | ROVNER | HAMILTON | BRENNAN | 1 | {} | 4 | 0 | 2 | 0 | 2 | null | https://www.courtlistener.com/api/rest/v4/clusters/4601833/ | Published | 1 | 1 | 0 | 0 | 1 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,601,840 | United States v. Andres Garcia | 2019-03-20 | 18-1735 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Before ROVNER, HAMILTON, and BRENNAN, Circuit Judges.", "parties": "", "opinions": [{"author": "David Frank Hamilton", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________\nNo. 18-1735\nUNITED STATES... | ROVNER | HAMILTON | BRENNAN | 1 | {"ROVNER": ", Circuit", "HAMILTON": ", Circuit", "BRENNAN": ", Circuit"} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4601840/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,601,905 | United States v. Andres Garcia | 2019-03-20 | 18-1735 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Before ROVNER, HAMILTON, and BRENNAN, Circuit Judges.", "parties": "", "opinions": [{"author": "David Frank Hamilton", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________\nNo. 18-1735\nUNITED STATES... | ROVNER | HAMILTON | BRENNAN | 1 | {"ROVNER": ", Circuit", "HAMILTON": ", Circuit", "BRENNAN": ", Circuit"} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4601905/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,602,845 | United States v. Geraldo Colon | 2019-03-22 | 18-1233 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Before EASTERBROOK, BARRETT, and SCUDDER, Circuit Judges.", "parties": "", "opinions": [{"author": "SCUDDER, Circuit Judge", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________ \nNo. 18‐1233 \nUNIT... | EASTERBROOK | BARRETT | SCUDDER | 1 | {"EASTERBROOK": ", Circuit", "BARRETT": ", Circuit", "SCUDDER": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4602845/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,602,880 | United States v. Geraldo Colon | 2019-03-22 | 18-1233 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Before EASTERBROOK, BARRETT, and SCUDDER, Circuit Judges.", "parties": "", "opinions": [{"author": "SCUDDER, Circuit Judge", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________ \nNo. 18‐1233 \nUNIT... | EASTERBROOK | BARRETT | SCUDDER | 1 | {"EASTERBROOK": ", Circuit", "BARRETT": ", Circuit", "SCUDDER": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4602880/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
8,443,940 | United States v. Geraldo COLON | United States v. Colon | 2019-03-22 | No. 18-1233 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Barrett, Easterbrook, Scudder", "parties": "", "opinions": [{"author": "", "type": "020lead", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opinions/8415371/", "author_raw": ""}]} | BARRETT | EASTERBROOK | SCUDDER | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/8443940/ | Published | 0 | 1 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,603,400 | ST. JOAN ANTIDA HIGH SCHOOL INC., Plaintiff-Appellant, v. MILWAUKEE PUBLIC SCHOOL DISTRICT, Defendant-Appellee. | St. Joan Antida High School In v. Milwaukee Public School Distri | 2019-03-25 | 18-1673 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Sykes, Barrett, Eve", "parties": "", "opinions": [{"author": "", "type": "020lead", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opinions/9888193/", "author_raw": ""}, {"author": "", "type": "040dissent", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opinio... | SYKES | BARRETT | 0 | {} | 4 | 0 | 2 | 0 | 2 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603400/ | Published | 1 | 1 | 0 | 0 | 1 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,603,539 | St. Joan Antida High School In v. Milwaukee Public School Distri | 2019-03-25 | 18-1673 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Before SYKES, BARRETT, and ST. EVE, Circuit Judges.", "parties": "", "opinions": [{"author": "Diane S. Sykes", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________ \nNo. 18‐1673 \nST. JOAN ANTID... | SYKES | BARRETT | ST EVE | 1 | {"SYKES": ", Circuit", "BARRETT": ", Circuit", "ST EVE": ", Circuit"} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603539/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,603,822 | Norma L. COOKE, Plaintiff-Appellee, Cross-Appellant, v. JACKSON NATIONAL LIFE INSURANCE COMPANY, Defendant-Appellant, Cross-Appellee. | Norma Cooke v. Jackson National Life Insuran | 2019-03-26 | 18-3527 & 18-3583 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Easterbrook, Barrett, Stadtmueller", "parties": "", "opinions": [{"author": "Frank Hoover Easterbrook", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________\n\nNos. 18-3527 & 18-3583\nNORM... | EASTERBROOK | BARRETT | STADTMUELLER | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603822/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,603,935 | Norma Cooke v. Jackson National Life Insuran | 2019-03-26 | 18-3583 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Before EASTERBROOK and BARRETT, Circuit Judges, and STADTMUELLER, District Judge.", "parties": "", "opinions": [{"author": "Frank Hoover Easterbrook", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n _____... | EASTERBROOK | BARRETT | STADTMUELLER | 1 | {"EASTERBROOK": ", Circuit", "BARRETT": ", Circuit", "STADTMUELLER": ", District"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603935/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,604,321 | UNITED STATES of America, Plaintiff-Appellee, v. Alandous BRIGGS, Defendant-Appellant. | United States v. Alandous Briggs | 2019-03-27 | 18-1415 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Bauer, Hamilton, Barrett", "parties": "", "opinions": [{"author": "BARRETT, Circuit Judge", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________\nNo. 18-1415\nUNITED STATES OF AMERICA,\n ... | BAUER | HAMILTON | BARRETT | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4604321/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,604,373 | United States v. Alandous Briggs | 2019-03-27 | 18-1415 | U.S. Court of Appeals for the Seventh Circuit | {"judges": "Before BAUER, HAMILTON, and BARRETT, Circuit Judges.", "parties": "", "opinions": [{"author": "BARRETT, Circuit Judge", "type": "010combined", "text": "In the\n\n United States Court of Appeals\n For the Seventh Circuit\n ____________________\nNo. 18-1415\nUNITED STATES ... | BAUER | HAMILTON | BARRETT | 1 | {"BAUER": ", Circuit", "HAMILTON": ", Circuit", "BARRETT": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4604373/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,595,514 | Jesse CAMPBELL, Individually and on Behalf of All Others Similarly Situated, Plaintiff - Appellant v. TRANSGENOMIC, INC.; Paul Kinnon; Precipio, Inc., Defendants - Appellees | Jesse Campbell v. Transgenomic, Inc. | 2019-03-01 | 18-2198 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Benton, Beam, Erickson", "parties": "", "opinions": [{"author": "BENTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-2198\... | BENTON | BEAM | ERICKSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4595514/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,595,517 | Clemente Avelino PEREIDA Petitioner v. William P. BARR, Attorney General of the United States Respondent | Clemente Avelino Pereida v. William P. Barr | 2019-03-01 | 17-3377 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Erickson, Beam, Grasz", "parties": "", "opinions": [{"author": "BEAM, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-3377\n ... | ERICKSON | BEAM | GRASZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4595517/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,596,479 | UNITED STATES of America, Plaintiff - Appellee v. Craig A. SUTTON, Defendant - Appellant | United States v. Craig Sutton | 2019-03-05 | 17-3195 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Gruender, Kelly, Grasz", "parties": "", "opinions": [{"author": "KELLY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-319... | GRUENDER | KELLY | GRASZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4596479/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,596,480 | Perfecto MARTIN MARTIN; Maynor Martin-Vicente, Petitioners v. William P. BARR, Attorney General of the United States of America, Respondent | Perfecto Martin Martin v. William P. Barr | 2019-03-05 | 18-1059 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Loken, Melloy, Erickson", "parties": "", "opinions": [{"author": "LOKEN, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1059\... | LOKEN | MELLOY | ERICKSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4596480/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,597,035 | UNITED STATES of America Plaintiff-Appellee v. Mark PALMER Defendant-Appellant United States of America Plaintiff-Appellee v. Samuel Leinicke Defendant-Appellant | United States v. Mark Palmer | 2019-03-06 | 18-1365; 18-1367 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Gruender, Wollman, Shepherd", "parties": "", "opinions": [{"author": "WOLLMAN, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1365\n ___________________________\n\n Un... | GRUENDER | WOLLMAN | SHEPHERD | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4597035/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,597,038 | UNITED STATES of America Plaintiff - Appellee v. Kalolo Nathaniel IU Defendant - Appellant | United States v. Kalolo Nathaniel Iu | 2019-03-06 | 17-3237 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Shepherd, Kelly, Stras", "parties": "", "opinions": [{"author": "SHEPHERD, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-3237\n... | SHEPHERD | KELLY | STRAS | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4597038/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,597,039 | UNITED STATES of America Plaintiff - Appellee v. Jayme Nathaniel WALKER Defendant - Appellant | United States v. Jayme Walker | 2019-03-06 | 18-1355 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Gruender, Kelly, Grasz", "parties": "", "opinions": [{"author": "", "type": "020lead", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opinions/9888108/", "author_raw": ""}, {"author": "", "type": "030concurrence", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4... | GRUENDER | KELLY | GRASZ | 1 | {} | 4 | 0 | 0 | 2 | 2 | null | https://www.courtlistener.com/api/rest/v4/clusters/4597039/ | Published | 1 | 1 | 1 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,597,041 | UNITED STATES of America, Plaintiff-Appellee v. Armando REYES-RAMIREZ, Defendant-Appellant | United States v. Armando Reyes-Ramirez | 2019-03-06 | 18-1143 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Loken, Melloy, Erickson", "parties": "", "opinions": [{"author": "LOKEN, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1143\n ... | LOKEN | MELLOY | ERICKSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4597041/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,597,043 | Robert FORD Movant - Appellant v. UNITED STATES of America Respondent - Appellee | Robert Ford v. United States | 2019-03-06 | 17-2206 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Smith, Loken, Gruender", "parties": "", "opinions": [{"author": "SMITH, Chief Judge", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-2206\n ... | SMITH | LOKEN | GRUENDER | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4597043/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,597,044 | D.M., a Minor, BY BAO XIONG, the Mother, Legal Guardian, and Next Friend of D.M.; Z.G., a Minor, by Joel Greenwald, the Father, Legal Guardian, and Next Friend of Z.G. Plaintiffs - Appellants v. MINNESOTA STATE HIGH SCHOOL LEAGUE; Bonnie Spohn-Schmaltz, in Her Official Capacity as President of the Board of Directors fo... | D.M. v. Minn. State High Sch. League | 2019-03-06 | 18-3077 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Loken, Melloy, Erickson", "parties": "", "opinions": [{"author": "MELLOY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-3077... | LOKEN | MELLOY | ERICKSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4597044/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,597,618 | UNITED STATES of America, Plaintiff - Appellee v. Carlos ROMAN, Defendant - Appellant | United States v. Carlos Roman | 2019-03-07 | 17-3582 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Smith, Melloy, Stras", "parties": "", "opinions": [{"author": "SMITH, Chief Judge", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-3582\n ... | SMITH | MELLOY | STRAS | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4597618/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,597,624 | Barton ROBERTS, Plaintiff - Appellant v. Sergeant KOPEL, Et Al., Defendants - Appellees | Barton Roberts v. Sergeant Kopel | 2019-03-07 | 17-3579 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Smith, Loken, Gruender", "parties": "", "opinions": [{"author": "LOKEN, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-3579\n... | SMITH | LOKEN | GRUENDER | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4597624/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,598,090 | Wadith Stockinger NADER; Stacey Nichole Nader, Plaintiffs - Appellants v. CITY OF PAPILLION; Sarpy County, Nebraska; Bryan Svajgl; Benjamin Iversen; Scott A. Lyons; L. Kenneth Polikov; Jennifer Miralles, Defendants - Appellees | Wadith Nader v. City of Papillion | 2019-03-08 | 18-1402 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Loken, Melloy, Erickson", "parties": "", "opinions": [{"author": "MELLOY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1402\n ... | LOKEN | MELLOY | ERICKSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4598090/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,598,092 | UNITED STATES of America, Plaintiff - Appellee v. Mark A. BECKHAM, Defendant - Appellant | United States v. Mark Beckham | 2019-03-08 | 18-1406 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Gruender, Wollman, Shepherd", "parties": "", "opinions": [{"author": "SHEPHERD, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1... | GRUENDER | WOLLMAN | SHEPHERD | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4598092/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,598,093 | UNITED STATES of America, Plaintiff - Appellee, v. $579,475.00 IN U.S. CURRENCY, Defendant, LNG Express, Inc., Claimant - Appellant. | United States v. $579,475.00 in U.S. Currency | 2019-03-08 | 16-3646 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Smith, Wollman, Loken, Colloton, Gruender, Benton, Shepherd, Kelly, Erickson, Grasz, Stras", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ______________... | SMITH | WOLLMAN | LOKEN | 2 | COLLOTON; GRUENDER; BENTON; SHEPHERD; KELLY; ERICKSON; GRASZ; STRAS | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4598093/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||
4,598,095 | Tamela MUIR, Plaintiff - Appellee v. DECATUR COUNTY, IOWA; Ben Boswell, Sheriff, Defendants - Appellants | Tamela Muir v. Decatur County, Iowa | 2019-03-08 | 18-1057 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Loken, Melloy, Erickson", "parties": "", "opinions": [{"author": "MELLOY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1057\n ... | LOKEN | MELLOY | ERICKSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4598095/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,598,755 | UNITY HEALTHCARE, Plaintiff - Appellant v. Alex M. AZAR, II, Secretary, U.S. Department of Health and Human Services, Defendant - Appellee St. Anthony Regional Hospital, Plaintiff - Appellant v. Alex M. Azar, II, Secretary, U.S. Department of Health and Human Services, Defendant - Appellee Lakes Regional Healthcare, Pl... | Unity HealthCare v. Alex M. Azar II | 2019-03-12 | 18-1316; 18-1703; 18-1704 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Benton, Beam, Erickson", "parties": "", "opinions": [{"author": "ERICKSON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1316\n ... | BENTON | BEAM | ERICKSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4598755/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,599,064 | COR CLEARING, LLC, Plaintiff - Appellant v. CALISSIO RESOURCES GROUP, INC.; Signature Stock Transfer, Inc.; TD Ameritrade Clearing, Inc.; National Financial Services, LLC; Scottrade, Inc.; E-Trade Clearing, LLC, Defendants - Appellees | COR Clearing, LLC v. Calissio Resources Group, Inc. | 2019-03-13 | 17-3556 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Loken, Erickson, Magnuson", "parties": "", "opinions": [{"author": "LOKEN, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-3556\n... | LOKEN | ERICKSON | MAGNUSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4599064/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,599,775 | COUNTRY PREFERRED INSURANCE COMPANY Plaintiff - Appellee v. Christopher LEE; Sandra Lee Defendants - Appellants | Country Preferred Insurance v. Christopher Lee | 2019-03-14 | 18-2096 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Benton, Melloy, Kelly", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-2096\n ... | BENTON | MELLOY | KELLY | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4599775/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,600,834 | UNITED STATES of America Plaintiff - Appellee v. Marvin MEUX Defendant - Appellant | United States v. Marvin Meux | 2019-03-18 | 18-1416 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Benton, Melloy, Kelly", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1416\n ... | BENTON | MELLOY | KELLY | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4600834/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,600,835 | UNITED STATES of America Plaintiff - Appellee v. Cornelius COLEMAN, Also Known as Cornbread Defendant - Appellant | United States v. Cornelius Coleman | 2019-03-18 | 18-2400 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Benton, Melloy, Kelly", "parties": "", "opinions": [{"author": "KELLY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-2400\n ... | BENTON | MELLOY | KELLY | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4600835/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,600,841 | Adrian BRYANT, Plaintiff - Appellee v. JEFFREY SAND COMPANY, Defendant - Appellant | Adrian Bryant v. Jeffrey Sand Company | 2019-03-18 | 18-2297 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Benton, Melloy, Kelly", "parties": "", "opinions": [{"author": "KELLY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-2297\n ... | BENTON | MELLOY | KELLY | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4600841/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,601,480 | UNITED STATES of America, Plaintiff - Appellee, v. William Earl MORRIS, Defendant - Appellant. | United States v. William Morris | 2019-03-20 | 17-2979 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Wollman, Colloton, Benton", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-297... | WOLLMAN | COLLOTON | BENTON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4601480/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,601,481 | UNITED STATES of America, Plaintiff - Appellant, v. Stirling Michael HEATON, Defendant - Appellee. | United States v. Stirling Michael Heaton | 2019-03-20 | 17-3314 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Wollman, Colloton, Benton", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-... | WOLLMAN | COLLOTON | BENTON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4601481/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,602,282 | UNITED STATES of America, Plaintiff-Appellee v. Casey FOGG, Defendant-Appellant | United States v. Casey Fogg | 2019-03-21 | 18-1676 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Loken, Erickson, Magnuson", "parties": "", "opinions": [{"author": "LOKEN, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1676\n... | LOKEN | ERICKSON | MAGNUSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4602282/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,602,283 | Milagro Del Blanco DE GUEVARA, Et Al. Petitioners v. William P. BARR, Attorney General of the United States Respondent | Milagro Blanco De Guevara v. William P. Barr | 2019-03-21 | 18-1080 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Loken, Melloy, Erickson", "parties": "", "opinions": [{"author": "LOKEN, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1080\... | LOKEN | MELLOY | ERICKSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4602283/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,602,284 | Edwin VAN DORN, Plaintiff - Appellant v. Scott HUNTER, Defendant - Appellee | Edwin Van Dorn v. Scott Hunter | 2019-03-21 | 18-2053 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Loken, Melloy, Erickson", "parties": "", "opinions": [{"author": "ERICKSON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-2053\... | LOKEN | MELLOY | ERICKSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4602284/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,602,285 | DAKOTA, MINNESOTA & EASTERN RAILROAD CORPORATION and Soo Line Railroad Company D/B/A Canadian Pacific Railway, Plaintiffs-Appellees v. INGRAM BARGE COMPANY, Defendant-Appellant | Dakota, Minnesota & E. Railway v. Ingram Barge Company | 2019-03-21 | 18-2143 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Benton, Melloy, Kelly", "parties": "", "opinions": [{"author": "KELLY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-2143\n ... | BENTON | MELLOY | KELLY | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4602285/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,602,286 | CEDAR RAPIDS BANK AND TRUST COMPANY, Plaintiff - Appellee v. MAKO ONE CORPORATION, Et Al., Defendant - Appellant | Cedar Rapids Bank & Trust Co. v. Mako One Corporation | 2019-03-21 | 18-1298 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Benton, Beam, Erickson", "parties": "", "opinions": [{"author": "ERICKSON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1298\n... | BENTON | BEAM | ERICKSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4602286/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,602,287 | Augustine PACHECO; Vicki Hansen, for Themselves and Others Similarly-Situated Plaintiffs - Appellees v. HONEYWELL INTERNATIONAL INC. Defendant - Appellant | Augustine Pacheco v. Honeywell International Inc. | 2019-03-21 | 18-1006; 18-1294 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Loken, Melloy, Erickson", "parties": "", "opinions": [{"author": "LOKEN, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1006\n ... | LOKEN | MELLOY | ERICKSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4602287/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,602,847 | Walter ENGELHARDT, Plaintiff - Appellant v. QWEST CORPORATION; Tim Buchholz, Defendants - Appellees | Walter Engelhardt v. Qwest Corporation | 2019-03-22 | 17-2492 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Smith, Loken, Gruender", "parties": "", "opinions": [{"author": "SMITH, Chief Judge", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-2492\n ... | SMITH | LOKEN | GRUENDER | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4602847/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,602,851 | UNITED STATES of America, Plaintiff - Appellee v. Dashown Raymond KEYS, Defendant - Appellant | United States v. Dashown Keys | 2019-03-22 | 18-2183 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Loken, Erickson, Magnuson", "parties": "", "opinions": [{"author": "LOKEN, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-2183\n... | LOKEN | ERICKSON | MAGNUSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4602851/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,602,852 | Ronald C. BUCKLER, Plaintiff - Appellant v. UNITED STATES of America, Defendant - Appellee | Ronald Buckler v. United States | 2019-03-22 | 17-2567 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Benton, Melloy, Grasz", "parties": "", "opinions": [{"author": "MELLOY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-2567\n... | BENTON | MELLOY | GRASZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4602852/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,602,853 | Dennis Thomas THOMPSON, Plaintiff - Appellant, v. COMMISSIONER OF SOCIAL SECURITY ADMINISTRATION, Defendant - Appellee. | Dennis Thomas Thompson v. Nancy A. Berryhill | 2019-03-22 | 17-2111 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Wollman, Colloton, Benton", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-... | WOLLMAN | COLLOTON | BENTON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4602853/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,602,855 | Corrie NELSON, Plaintiff - Appellant v. USABLE MUTUAL INSURANCE COMPANY, Doing Business as Arkansas Blue Cross and Blue Shield, Defendant - Appellee | Corrie Nelson v. USAble Mutual Insurance Co. | 2019-03-22 | 18-1439 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Benton, Melloy, Kelly", "parties": "", "opinions": [{"author": "MELLOY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1439\n ... | BENTON | MELLOY | KELLY | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4602855/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,603,246 | UNITED STATES of America, Plaintiff - Appellee v. Pradis Lee BENTON, Defendant - Appellant | United States v. Pradis Benton | 2019-03-25 | 18-1016 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Gruender, Kelly, Grasz", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1016\n ... | GRUENDER | KELLY | GRASZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603246/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,603,248 | PRAIRIE LAND HOLDINGS, L.L.C., Petitioner v. FEDERAL AVIATION ADMINISTRATION, Respondent | Prairie Land Holdings, L. L.C. v. F.A.A. | 2019-03-25 | 18-2234 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Benton, Melloy, Kelly", "parties": "", "opinions": [{"author": "", "type": "020lead", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opinions/9888189/", "author_raw": ""}, {"author": "", "type": "040dissent", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opin... | BENTON | MELLOY | KELLY | 1 | {} | 4 | 0 | 2 | 0 | 2 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603248/ | Published | 1 | 1 | 0 | 0 | 1 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,603,250 | Anmarie CALGARO, Plaintiff - Appellant, v. ST. LOUIS COUNTY; Linnea Mirsch, Individually and in Her Official Capacity as Director of St. Louis County Public Health and Human Services ; Fairview Health Services, a Minnesota Nonprofit Corporation; Park Nicollet Health Services, a Nonprofit Corporation; St. Louis County S... | Anmarie Calgaro v. St. Louis County | 2019-03-25 | 17-2279 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Wollman, Colloton, Benton", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-227... | WOLLMAN | COLLOTON | BENTON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603250/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,604,104 | UNITED STATES of America, Plaintiff - Appellee v. Julius Lamon JONES, Defendant - Appellant | United States v. Julius Jones | 2019-03-27 | 17-2944 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Smith, Wollman, Grasz", "parties": "", "opinions": [{"author": "SMITH, Chief Judge", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-2944\n ... | SMITH | WOLLMAN | GRASZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4604104/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,604,597 | WAL-MART STORES EAST, LP, Doing Business as Walmart Distribution Center # 7035, Petitioner v. Alexander ACOSTA, Secretary of Labor, Respondent | Secretary of Labor v. Walmart Stores East | 2019-03-28 | 17-2647 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Colloton, Beam, Grasz", "parties": "", "opinions": [{"author": "GRASZ, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-2647\n ... | COLLOTON | BEAM | GRASZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4604597/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,605,133 | UNITED STATES of America, Plaintiff - Appellee, v. Anthony Leon WAITS, Defendant - Appellant. United States of America, Plaintiff - Appellee, v. Jacqueline D. Mills, Defendant - Appellant. | United States v. Anthony Waits | 2019-03-29 | 17-3399; 17-3762 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Colloton, Beam, Grasz", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-3399\n ___________________________\n\n Un... | COLLOTON | BEAM | GRASZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4605133/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,605,134 | UNITED STATES of America, Plaintiff - Appellee, v. Ampelio HERNANDEZ, Defendant - Appellant. United States of America, Plaintiff - Appellee, v. Oscar Espinoza Lopez, Defendant - Appellant. United States of America, Plaintiff - Appellee, v. Raymundo Moreno Lopez, Also Known as Brother, Defendant - Appellant. | United States v. Ampelio Hernandez | 2019-03-29 | 17-2753; 17-2754; 17-2755 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Colloton, Gruender, Grasz", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-2753\n ___________________________\n\n United St... | COLLOTON | GRUENDER | GRASZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4605134/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,605,135 | Oralia JUAREZ-CORONADO, Now Known as Oralia Magali Marroquin, Petitioner v. William P. BARR, Attorney General of the United States, Respondent Immigration Law Professors, Interested Party - Amicus on Behalf of Petitioner | Oralia Juarez-Coronado v. William P. Barr | 2019-03-29 | 17-3780 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Shepherd, Kelly, Stras", "parties": "", "opinions": [{"author": "KELLY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-378... | SHEPHERD | KELLY | STRAS | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4605135/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,605,137 | Joseph HARDEN, Petitioner - Appellee, v. Jeff NORMAN, Respondent - Appellant. | Joseph Harden v. Michael Bowersox | 2019-03-29 | 17-2067 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Colloton, Gruender, Grasz", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-... | COLLOTON | GRUENDER | GRASZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4605137/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,605,138 | Donald MORGAN, Plaintiff - Appellee v. Michael ROBINSON, Washington County Sheriff, an Individual, Defendant - Appellant Washington County, Nebraska, Defendant | Donald Morgan v. Michael Robinson | 2019-03-29 | 17-1002 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Smith, Wollman, Loken, Colloton, Gruender, Benton, Shepherd, Kelly, Erickson, Grasz, Stras", "parties": "", "opinions": [{"type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | SMITH | WOLLMAN | LOKEN | 2 | COLLOTON; GRUENDER; BENTON; SHEPHERD; KELLY; ERICKSON; GRASZ; STRAS | {} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4605138/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||
4,605,140 | Chad Menter HILL, Appellant v. James L. SNYDER, Appellee | Chad Menter Hill v. James L. Snyder | 2019-03-29 | 17-3572 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Shepherd, Kelly, Stras", "parties": "", "opinions": [{"author": "KELLY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-3572\n... | SHEPHERD | KELLY | STRAS | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4605140/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,595,544 | Vicente Crawford v. A. B. Won Pat Intl. Airport | 2019-03-01 | 17-16942 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Sidney R. Thomas, Chief Judge, Susan P. Graber, Circuit Judge, and Leslie E. Kobayashi,* District Judge.", "parties": "", "opinions": [{"author": "KOBAYASHI, District Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nVICEN... | SIDNEY R THOMAS | SUSAN P GRABER | LESLIE E KOBAYASHI | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4595544/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,595,545 | Kyle Rodney v. Timothy Filson | 2019-03-01 | 17-15438 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Ronald M. Gould and Marsha S. Berzon, Circuit Judges, and Rosemary Márquez, * District Judge.", "parties": "", "opinions": [{"author": "Opinion by Judge Márquez", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n KYLE J. RODNEY,... | RONALD M GOULD | MARSHA S BERZON | ROSEMARY MÁRQUEZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4595545/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,595,921 | Warren Iopa v. Saltchuk-Young Brothers, Ltd. | 2019-03-04 | 17-70415 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Richard C. Tallman, Jay S. Bybee, and N. Randy Smith, Circuit Judges.", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n WARREN K. IOPA, Claimant, No. 1... | RICHARD C TALLMAN | JAY S BYBEE | N RANDY SMITH | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4595921/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,595,922 | United States v. Antonio Gilton | 2019-03-04 | 16-10109 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: J. Clifford Wallace, M. Margaret McKeown, and Jay S. Bybee, Circuit Judges.", "parties": "", "opinions": [{"type": "010combined", "text": "FILED\n FOR PUBLICATION\n MAR 04 2019\n UNITED STATES COU... | J CLIFFORD WALLACE | M MARGARET MCKEOWN | JAY S BYBEE | 1 | {} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4595922/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,595,923 | Christopher Spreitz v. Charles Ryan | 2019-03-04 | 09-99006 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Richard A. Paez, Marsha S. Berzon, and Richard C. Tallman, Circuit Judges.", "parties": "", "opinions": [{"author": "PAEZ, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nCHRISTOPHER J. SPREITZ, No... | RICHARD A PAEZ | MARSHA S BERZON | RICHARD C TALLMAN | 1 | {} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4595923/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,596,569 | Christopher Spreitz v. Charles Ryan | 2019-03-04 | 09-99006 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Richard A. Paez, Marsha S. Berzon, and Richard C. Tallman, Circuit Judges.", "parties": "", "opinions": [{"author": "PAEZ, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nCHRISTOPHER J. SPREITZ, No... | RICHARD A PAEZ | MARSHA S BERZON | RICHARD C TALLMAN | 1 | {} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4596569/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,597,698 | Vijayakumar Thuraissigiam v. Usdhs | 2019-03-07 | 18-55313 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: A. Wallace Tashima, M. Margaret McKeown, and Richard A. Paez, Circuit Judges.", "parties": "", "opinions": [{"author": "TASHIMA, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nVIJAYAKUMAR THURAISSIGIAM, ... | A WALLACE TASHIMA | M MARGARET MCKEOWN | RICHARD A PAEZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4597698/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,597,699 | Dw Aina Le'a Development, LLC v. State of Hawaii Land Use Comm | 2019-03-07 | 17-16280 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Richard C. Tallman, Jay S. Bybee, and N. Randy Smith, Circuit Judges.", "parties": "", "opinions": [{"author": "Jay S. Bybee, Presiding Judge", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nDW AINA LE‘A ... | RICHARD C TALLMAN | JAY S BYBEE | N RANDY SMITH | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4597699/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,597,700 | Danilo Mairena v. William Barr | 2019-03-07 | 15-72833 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Ronald M. Gould and Mary H. Murguia, Circuit Judges, and Carol Bagley Amon, ** District Judge.", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n DANILO ALBERTO MAIRENA, AKA ... | RONALD M GOULD | MARY H MURGUIA | CAROL BAGLEY AMON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4597700/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,598,128 | Thomas Anderson v. State Farm Mutual Auto. Ins. | 2019-03-08 | 15-35981 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: M. Margaret McKeown and Michelle T. Friedland, Circuit Judges, and Susan R. Bolton, * District Judge.", "parties": "", "opinions": [{"author": "McKEOWN, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n PATRICIA... | M MARGARET MCKEOWN | MICHELLE T FRIEDLAND | SUSAN R BOLTON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4598128/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,598,508 | Ipc (Usa), Inc. v. Kathryn Ellis | 2019-03-11 | 17-60081 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: William A. Fletcher and Jay S. Bybee, Circuit Judges, and Larry A. Burns, * Chief District Judge.", "parties": "", "opinions": [{"author": "BURNS, Chief District Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nIN RE PETT... | WILLIAM A FLETCHER | JAY S BYBEE | LARRY A BURNS | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4598508/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,598,813 | Victoria Dieringer v. Cir | 2019-03-12 | 16-72640 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: M. Margaret McKeown and Richard A. Paez, Circuit Judges, and Cynthia A. Bashant, * District Judge.", "parties": "", "opinions": [{"author": "PAEZ, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n VICTORIA E. DI... | M MARGARET MCKEOWN | RICHARD A PAEZ | CYNTHIA A BASHANT | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4598813/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,598,814 | United States v. Kenneth Door | 2019-03-12 | 17-30165 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Sidney R. Thomas, Chief Judge, and M. Margaret McKeown and Morgan Christen, Circuit Judges.", "parties": "", "opinions": [{"author": "CHRISTEN, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AME... | SIDNEY R THOMAS | M MARGARET MCKEOWN | MORGAN CHRISTEN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4598814/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,598,815 | United States v. David Prien-Pinto | 2019-03-12 | 18-30055 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: William A. Fletcher and Jay S. Bybee, Circuit Judges, and Larry A. Burns,** District Judge.", "parties": "", "opinions": [{"author": "BYBEE, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n UNITED STATES OF A... | WILLIAM A FLETCHER | JAY S BYBEE | LARRY A BURNS | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4598815/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,598,816 | Franks Landing Indian Cmty. v. National Indian Gaming Comm'r. | 2019-03-12 | 17-35368 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: N. Randy Smith and Morgan Christen, Circuit Judges, and Robert E. Payne,* District Judge.", "parties": "", "opinions": [{"author": "CHRISTEN, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nFRANK’S LANDIN... | N RANDY SMITH | MORGAN CHRISTEN | ROBERT E PAYNE | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4598816/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,599,148 | homeaway.com, Inc. v. City of Santa Monica | 2019-03-13 | 18-55367 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Mary M. Schroeder and Jacqueline H. Nguyen, Circuit Judges, and Michael H. Simon, * District Judge.", "parties": "", "opinions": [{"author": "NGUYEN, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nHOMEAWAY.COM, ... | MARY M SCHROEDER | JACQUELINE H NGUYEN | MICHAEL H SIMON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4599148/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,599,149 | Wildearth Guardians v. Heather Provencio | 2019-03-13 | 17-17373 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: MICHAEL DALY HAWKINS, MILAN D. SMITH, JR., and ANDREW D. HURWITZ, Circuit Judges.", "parties": "", "opinions": [{"author": "M. SMITH, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nWILDEARTH GUARDIANS; GRAND ... | MICHAEL DALY HAWKINS | MILAN D SMITH JR | ANDREW D HURWITZ | 1 | {"MICHAEL DALY HAWKINS": ", Circuit", "MILAN D SMITH JR": ", Circuit", "ANDREW D HURWITZ": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4599149/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,599,150 | Mickey Fowler v. Tracy Guerin | 2019-03-13 | 16-35052 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Ronald M. Gould and Sandra S. Ikuta, Circuit Judges, and John R. Tunheim, * Chief District Judge.", "parties": "", "opinions": [{"author": "Judge for the District of Minnesota, sitting by designation", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR ... | RONALD M GOULD | SANDRA S IKUTA | JOHN R TUNHEIM | 1 | {} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4599150/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,599,151 | Duanna Knighton v. Cedarville Rancheria of Npi | 2019-03-13 | 17-15515 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: A. Wallace Tashima and Milan D. Smith, Jr., Circuit Judges, and Lawrence L. Piersol, * District Judge.", "parties": "", "opinions": [{"author": "Opinion by Judge Piersol", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n DUANNA K... | A WALLACE TASHIMA | MILAN D SMITH JR | LAWRENCE L PIERSOL | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4599151/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,599,152 | Advanced Building & Fabricatio v. Curtis Ayers | 2019-03-13 | 17-16618 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Milan D. Smith, Jr. and Jacqueline H. Nguyen, Circuit Judges, and Jane A. Restani, * Judge.", "parties": "", "opinions": [{"author": "NGUYEN, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nADVANCED BUILDING & ... | MILAN D SMITH JR | JACQUELINE H NGUYEN | JANE A RESTANI | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4599152/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,603,420 | homeaway.com, Inc. v. City of Santa Monica | 2019-03-13 | 18-55367 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Mary M. Schroeder and Jacqueline H. Nguyen, Circuit Judges, and Michael H. Simon, * District Judge.", "parties": "", "opinions": [{"author": "NGUYEN, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nHOMEAWAY.COM, ... | MARY M SCHROEDER | JACQUELINE H NGUYEN | MICHAEL H SIMON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603420/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
9,022,991 | ADVANCED BUILDING & FABRICATION, INC., a California Corporation Robert Honan v. CALIFORNIA HIGHWAY PATROL John Wilson, and Curtis Ayers | Advanced Bldg. & Fabrication, Inc. v. Cal. Highway Patrol | 2019-03-13 | No. 17-16618 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Nguyen", "parties": "", "opinions": [{"author": "", "type": "020lead", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opinions/9016229/", "author_raw": ""}]} | NGUYEN | NGUYEN | 0 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/9022991/ | Published | 0 | 1 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,603,416 | United States v. Sha-Ron Haines | 2019-03-14 | 17-10059 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: J. Clifford Wallace and Michelle T. Friedland, Circuit Judges, and Lynn S. Adelman, * District Judge.", "parties": "", "opinions": [{"author": "ADELMAN, District Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED S... | J CLIFFORD WALLACE | MICHELLE T FRIEDLAND | LYNN S ADELMAN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603416/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,603,417 | United States v. Jonathan Read | 2019-03-14 | 17-10439 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Michael Daly Hawkins, Milan D. Smith, Jr., and Andrew D. Hurwitz, Circuit Judges.", "parties": "", "opinions": [{"author": "Opinion by Judge Hawkins", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, ... | MICHAEL DALY HAWKINS | MILAN D SMITH JR | ANDREW D HURWITZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603417/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,603,418 | Michael McNeil v. Sherwood School District 88j | 2019-03-14 | 17-35500 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Raymond C. Fisher and Consuelo M. Callahan, Circuit Judges, and Cathy Ann Bencivengo,* District Judge.", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nMICHAEL MCN... | RAYMOND C FISHER | CONSUELO M CALLAHAN | CATHY ANN BENCIVENGO | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603418/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,603,419 | Martin Valdez, Jr. v. W. Montgomery | 2019-03-14 | 16-56845 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Ronald M. Gould, Jacqueline H. Nguyen, and John B. Owens, Circuit Judges.", "parties": "", "opinions": [{"author": "GOULD, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nMARTIN LEYVA VALDEZ, JR., ... | RONALD M GOULD | JACQUELINE H NGUYEN | JOHN B OWENS | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603419/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,603,414 | Vht, Inc. v. Zillow Group, Inc. | 2019-03-15 | 17-35587 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: M. Margaret McKeown, William A. Fletcher, and Ronald M. Gould, Circuit Judges.", "parties": "", "opinions": [{"author": "McKEOWN, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nVHT, INC., a Delaware corporatio... | M MARGARET MCKEOWN | WILLIAM A FLETCHER | RONALD M GOULD | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603414/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,603,415 | United States v. Kyle Korte | 2019-03-15 | 18-50051 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Dorothy W. Nelson, Consuelo M. Callahan, and John B. Owens, Circuit Judges.", "parties": "", "opinions": [{"author": "OWENS, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, ... | DOROTHY W NELSON | CONSUELO M CALLAHAN | JOHN B OWENS | 1 | {} | 2 | 0 | 0 | 1 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603415/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,603,411 | Rebecca Morris v. California Physicians' Service | 2019-03-18 | 17-55878 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Mary M. Schroeder and Jacqueline H. Nguyen, Circuit Judges, and Thomas J. Whelan,* District Judge.", "parties": "", "opinions": [{"author": "SCHROEDER, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nREBECCA MORR... | MARY M SCHROEDER | JACQUELINE H NGUYEN | THOMAS J WHELAN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603411/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,603,412 | Marcelo Martinez-Cedillo V.william P. Barr | 2019-03-18 | 14-71742 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "", "parties": "", "opinions": [{"author": "THOMAS, Chief Judge", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n MARCELO MARTINEZ-CEDILLO, No. 14-71742\n AKA Marcelo Martinez,\n Petitioner, ... | 0 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603412/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||||||
4,603,413 | Ann Garcia v. Salvation Army | 2019-03-18 | 16-16827 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Mary M. Schroeder and Paul J. Watford, Circuit Judges, and Edward R. Korman, * District Judge.", "parties": "", "opinions": [{"author": "KORMAN, District Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nANN GARCIA, ... | MARY M SCHROEDER | PAUL J WATFORD | EDWARD R KORMAN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603413/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,603,410 | Phonsavanh Phongmanivan v. Ron Haynes | 2019-03-19 | 16-36018 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Sandra S. Ikuta and Morgan Christen, Circuit Judges, and Jennifer Choe-Groves,** Judge.", "parties": "", "opinions": [{"author": "The Honorable Jennifer Choe-Groves, Judge for the United", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH C... | SANDRA S IKUTA | MORGAN CHRISTEN | JENNIFER CHOE-GROVES | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603410/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,603,407 | William Herrera v. City of Palmdale | 2019-03-20 | 17-55761 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Diarmuid F. O’Scannlain and Sandra S. Ikuta, Circuit Judges, and George Carem Steeh, * District Judge.", "parties": "", "opinions": [{"author": "Opinion by Judge O'Scannlain", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nWIL... | DIARMUID F O'SCANNLAIN | SANDRA S IKUTA | GEORGE CAREM STEEH | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4603407/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 3 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.