| { |
| "paper_id": "Y00-1029", |
| "header": { |
| "generated_with": "S2ORC 1.0.0", |
| "date_generated": "2023-01-19T13:44:59.051533Z" |
| }, |
| "title": "JAPANESE HONORIFICATION IN AN HPSG FRAMEWORK", |
| "authors": [ |
| { |
| "first": "Melanie", |
| "middle": [], |
| "last": "Siegel", |
| "suffix": "", |
| "affiliation": { |
| "laboratory": "", |
| "institution": "University of the Saarland Postfach", |
| "location": { |
| "postCode": "151150 D-66041", |
| "settlement": "Saarbriicken", |
| "country": "Germany" |
| } |
| }, |
| "email": "siegel@dfki.de" |
| } |
| ], |
| "year": "", |
| "venue": null, |
| "identifiers": {}, |
| "abstract": "We present a solution for the representation of Japanese honorificational information in the HPSG framework. Basically, there are three dimensions of honorification. We show that a treatment is necessary that involves both the syntactic and the contextual level of information. The Japanese grammar is part of a machine translation system.", |
| "pdf_parse": { |
| "paper_id": "Y00-1029", |
| "_pdf_hash": "", |
| "abstract": [ |
| { |
| "text": "We present a solution for the representation of Japanese honorificational information in the HPSG framework. Basically, there are three dimensions of honorification. We show that a treatment is necessary that involves both the syntactic and the contextual level of information. The Japanese grammar is part of a machine translation system.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Abstract", |
| "sec_num": null |
| } |
| ], |
| "body_text": [ |
| { |
| "text": "The Verbmobil system is a machine translation system for German, English and Japanese dialogues. The szenario is a special sort of task-oriented dialogues: appointment scheduling' .", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Introduction", |
| "sec_num": "1" |
| }, |
| { |
| "text": "Spoken language encodes references to the social relation of the dialogue partners. The utterances can express social distance between addressee and speaker and third persons, who are mentioned. Honorifics can even express respect concerning entities of the world. Consider the following examples from Japanese, German The semantic content of these utterances is: 'When does it suit you?'. But there is an additional pragmatic content: The speaker expresses social distance concerning the addressee. This is expressed by the honorific pronouns Sie and vows in German and French. In the Japanese example it is expressed by the following attributes:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Introduction", |
| "sec_num": "1" |
| }, |
| { |
| "text": "\u2022 The honorific prefix go in front of tsugoo", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Introduction", |
| "sec_num": "1" |
| }, |
| { |
| "text": "\u2022 The honorific adjective yoroshii", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Introduction", |
| "sec_num": "1" |
| }, |
| { |
| "text": "A Japanese utterance with the same semantic content in for example -a family context could be:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "\u2022 The honorific copula deshoo", |
| "sec_num": null |
| }, |
| { |
| "text": "(4\\ (4) Information about honorification is -on the one hand -necessary for the description of syntactic phenomena like honorific agreement or relative sentences and -on the other hand -necessary for correct translation. In order to understand the whole meaning of the Japanese utterances it is important to represent the different honorific attributes in the analysis structure. The information can be used to resolve zero pronominalization and topicalized structures. It is even more important for the adequate generation of the Japanese utterances. In other investigations on zero pronoun resolution in task-oriented dialogues (cf. [Siege11997] ) we calculated that 23.9% of the zero pronominals can be solved using lexical pragmatic restrictions about honorification.", |
| "cite_spans": [ |
| { |
| "start": 4, |
| "end": 7, |
| "text": "(4)", |
| "ref_id": null |
| }, |
| { |
| "start": 635, |
| "end": 647, |
| "text": "[Siege11997]", |
| "ref_id": null |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "\u2022 The honorific copula deshoo", |
| "sec_num": null |
| }, |
| { |
| "text": "Honorifics in Japanese express the social relation of familiarity or distance between speaker, addressee and third persons. Consider the situation where the speaker asks if s/he returned a book to the addressee, 'did I return the book to you?'. In a familiar context (s)he would say:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Honorific Forms in Japanese", |
| "sec_num": "2" |
| }, |
| { |
| "text": "(5) hon book AKK kaeshita returned QUE", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Honorific Forms in Japanese", |
| "sec_num": "2" |
| }, |
| { |
| "text": "In a more formal situation with more social distance between speaker and addressee the utterance would be:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Honorific Forms in Japanese", |
| "sec_num": "2" |
| }, |
| { |
| "text": "(6) hgoonton ok Ngdm o-kaeshi-shimashita hon-returned-hon ka QUE", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Honorific Forms in Japanese", |
| "sec_num": "2" |
| }, |
| { |
| "text": "The book as belonging to the addressee is prefixed by the honorific go. The predicate gets the 'humble' extension shimashita and the question is expressed by ka.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Honorific Forms in Japanese", |
| "sec_num": "2" |
| }, |
| { |
| "text": "The social relationships that can be expressed are threefold: The first one is the relation between speaker and addressee, in the above example expressed by no and ka. The second one is the relation between the speaker and the subject of the utterance, in the above example expressed by the verbal form. The third one is the relation between the speaker and other arguments in the sentence. In the above example, book (hon) gets the honorific go prefix, because it is a book belonging to the addressee being honored by the speaker.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Honorific Forms in Japanese", |
| "sec_num": "2" |
| }, |
| { |
| "text": "Familiarity or distance between speaker and addressee can be expressed by verbal endings and/or the lexical choice of self-referring pronouns. Verbal endings encoding a relation of distance between speaker and addressee can be, for example, -masu, -mashita or -n-deshoo-ka. Those encoding a familiar relation can be, for example, -ru, -ta or -no. The choice of self-referring pronouns also depends on the gender of the speaker. A self-referring pronoun uttered by a woman in a familiar context could be watashi, a selfreferring pronoun uttered by a woman in a distant context could be watakushi. Parallel the appropriate self-referring pronoun for a man in a familiar context would be boku, one in a distant context would be watashi. I will call the relationship of honorifics concerning the relation between speaker and addressee AHON and give it a polarity [-AHON] for the plain form in a family context and [H-AHON] for the expressions in a context of social distance.", |
| "cite_spans": [ |
| { |
| "start": 859, |
| "end": 866, |
| "text": "[-AHON]", |
| "ref_id": null |
| }, |
| { |
| "start": 910, |
| "end": 918, |
| "text": "[H-AHON]", |
| "ref_id": null |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Honorific Forms in Japanese", |
| "sec_num": "2" |
| }, |
| { |
| "text": "The social relation between the speaker and a subject that is not referring to the speaker is expressed by the lexical choice of verbs, by the expression o-VERB-ni-naru, by the honorific prefix o/go at nouns referring to entities belonging to the subject and by the lexical choice of pronominals. I will call this relation between speaker and subject SHON. A relation of distance between speaker and subject (where the subject is the addressee or a third person) can be for example expressed by the verb irassharu (to go), while in a familiar situation the verb iku with the same semantic content is used. This is expressed by [+SHON] and [-SHON] , respectively. Possible referring expressions for the second and third person can be, for example, sochira and X-san in relations of distance and kimi or X-kun in relations of familiarity.", |
| "cite_spans": [ |
| { |
| "start": 627, |
| "end": 634, |
| "text": "[+SHON]", |
| "ref_id": null |
| }, |
| { |
| "start": 639, |
| "end": 646, |
| "text": "[-SHON]", |
| "ref_id": null |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Honorific Forms in Japanese", |
| "sec_num": "2" |
| }, |
| { |
| "text": "The third relation is the one between speaker and objects in the sentence (other than subject). I will call this relation OHON. It is expressed by the lexical choice of these entities and by the honorific prefixes o and go.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Honorific Forms in Japanese", |
| "sec_num": "2" |
| }, |
| { |
| "text": "The relationship of the speaker and the addressee can be one of three possible constellations:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Interaction of Different Kinds of Honorification in Japanese", |
| "sec_num": "3" |
| }, |
| { |
| "text": "1. The addressee is the subject of the utterance.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Interaction of Different Kinds of Honorification in Japanese", |
| "sec_num": "3" |
| }, |
| { |
| "text": "2. The speaker is the subject of the utterance.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Interaction of Different Kinds of Honorification in Japanese", |
| "sec_num": "3" |
| }, |
| { |
| "text": "When the addressee is the referent of the sentence subject, the relationships AHON and SHON must have the same polarity. In these cases, in a sentence with AHON there must also be SHON.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "A third person is the subject of the utterance.", |
| "sec_num": "3." |
| }, |
| { |
| "text": "In the situation, where the speaker is the subject of the utterance, (s)he uses humble forms of the verbs (a matter of lexical choice), if the AHON relation is a distant one. An example is mairu (to go). In this case, both relationships (SHON and AHON) are concerned.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "A third person is the subject of the utterance.", |
| "sec_num": "3." |
| }, |
| { |
| "text": "In many cases utterances contain multiple honorification as can be seen in the following example:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "A third person is the subject of the utterance.", |
| "sec_num": "3." |
| }, |
| { |
| "text": "(7) watakushi Aim o-denwa itashi-mashi-ta telephone do(hon)-hon-Past", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "A third person is the subject of the utterance.", |
| "sec_num": "3." |
| }, |
| { |
| "text": "The verbal stem itashi expresses subject honorification, the verbal ending mashi and the pronoun watakushi express addressee honorification.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "A third person is the subject of the utterance.", |
| "sec_num": "3." |
| }, |
| { |
| "text": "Japanese honorification undergoes different kinds of restrictions. The first kind to mention is called. 'pragmatic agreement' by [Pollard and Sag1994] . There must be agreement between the SHON honorification of the subject and the verb, as the following examples show:", |
| "cite_spans": [ |
| { |
| "start": 129, |
| "end": 150, |
| "text": "[Pollard and Sag1994]", |
| "ref_id": null |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "A third person is the subject of the utterance.", |
| "sec_num": "3." |
| }, |
| { |
| "text": "watashi ja sensei ni o-denwa I NOMprofessor DAT telephone itashi-mashi-ta do(humble-shon)-ahon-Past *sensei Afid M watashi ni o-denwa professor I DAT telephone itashi-mashi-ta do(humble-shoe)-ahon-Past (10) sensei watashi ni o-denwa professor NgaM I DAT telephone nasai-mashi-ta do(honorific-shon)-ahon-Past", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "A third person is the subject of the utterance.", |
| "sec_num": "3." |
| }, |
| { |
| "text": "The pronoun watashi can be used with a humble verb form, but not the honorific noun sensei. This must be used with a honorific verb form.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "A third person is the subject of the utterance.", |
| "sec_num": "3." |
| }, |
| { |
| "text": "Another kind of restriction concerns relative sentences as opposed to complement sentences. See the following examples from [Harada1976] ", |
| "cite_spans": [ |
| { |
| "start": 124, |
| "end": 136, |
| "text": "[Harada1976]", |
| "ref_id": null |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "A third person is the subject of the utterance.", |
| "sec_num": "3." |
| }, |
| { |
| "text": "Investigations of Japanese honorification have been made from the sociolinguistic, the grammatical and the machine translational viewpoint. For the sociolinguistic viewpoint see for example [Ide1986] , [Coulmas1987] , [Hori1986] 5 Japanese Honorification in HPSG [Pollard and Sag1994] analyze honorification as a pragmatic fact. They describe the problem as 'pragmatic agreement' and introduce a relation owe -honour in the BACKGR feature:", |
| "cite_spans": [ |
| { |
| "start": 190, |
| "end": 199, |
| "text": "[Ide1986]", |
| "ref_id": null |
| }, |
| { |
| "start": 202, |
| "end": 215, |
| "text": "[Coulmas1987]", |
| "ref_id": null |
| }, |
| { |
| "start": 218, |
| "end": 228, |
| "text": "[Hori1986]", |
| "ref_id": null |
| }, |
| { |
| "start": 263, |
| "end": 284, |
| "text": "[Pollard and Sag1994]", |
| "ref_id": null |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "{ B A C K G R RELAT ION owe -honour HONORER 1 HONORED 2 POLARITY 1/0 }", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "Still, the approach lacks for the fact that there are different kinds of honorification as we described above. It describes only subject honorification. [Green1997] elaborates the CONTEXT feature and introduces information about social ranking of the participants. We would, though, propose to leave the inference of the social relations to other components of, e.g., a machine translation system. The reason is that all necessary information is not always directly accessible in the analysis procedure. An example is given by [Coulmasl987]: The secretary in a company is asked by an employee, when the boss comes back. He In this example, we would represent the fact that the secretary honors the boss in the first example, but not in the second one. The interpretation of the complex social relations must be left to a module that has access to the information about the actual social relations of the participants in the context.", |
| "cite_spans": [ |
| { |
| "start": 153, |
| "end": 164, |
| "text": "[Green1997]", |
| "ref_id": null |
| }, |
| { |
| "start": 620, |
| "end": 622, |
| "text": "He", |
| "ref_id": null |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "To account for the fact that Japanese honorification has more dimensions, we propose the following CONTEXT feature structure: The C-INDS contain indices for speaker and addressee, as proposed by [Pollard and Sag1994] . The value of BACKGR is a difference list that sums up the occuring honorificational relations in the utterance. Each occuring relation gets classified into addr-honor-rel, subj-honor-rel and obj-honor-rel. The HONORER is co-indexed with the speaker in all cases here. This must be different in cases of indirect speech that we will describe later. The HONORED value is co-indexed with the addressee in C-INDS in the addr-honor-rel case, with the subject's CONTENT.INDEX value in the subj-honor-rel case and with the CONTENT.INDEX value of the argument that introduces the relation in the obj-honor-rel case.", |
| "cite_spans": [ |
| { |
| "start": 195, |
| "end": 216, |
| "text": "[Pollard and Sag1994]", |
| "ref_id": null |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "(15) C -INDS [1] [ SPEAKER [2] ADDRESSEE [3] { B A C K G", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "The relations all get a value of POLARITY, to account for the fact that there can be forms that are honorific, humble or neutral. A negative SHON polarity, e.g., reflects the situation where the speaker or a third person that socially belongs to the innner circle of the speaker is the subject of the utterance. [McGloin1987] describes this situation sociolinguistically as \"positive politeness\", because it expresses social closeness.", |
| "cite_spans": [ |
| { |
| "start": 312, |
| "end": 325, |
| "text": "[McGloin1987]", |
| "ref_id": null |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "The question is: how does the information enter into the BACKGR? Let us start with the obj-honorrel. This relation is encoded in the nouns that express honorification. The entry for o-uchiawase, e.g., contains:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "A verb that has restrictions on the honorification of its subject, contains the following in its lexical entry: If it happens to be the case that an entity with an obj-honor-rel in its BACKGR becomes the subject of the sentence, the mother node must get the subj-honor-rel from the predicate, identified with the index of that entity. It is, though, necessary to prove the honorificational restrictions on predicate and subject. Since this is a syntactic process, we decided on representing honorification on the syntactic level, too.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "CONTENTIKEY.ARG -1 [1] SPEAKER [2] 1 ADDRESSEE [3] C -", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "[Gunji1987} gives reasons for the syntactic approach. He describes in his JPSG-account of Japanese syntax honorification as a kind of agreement:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "\"Since Japanese does not have syntactic agreement phenomena such as number, person,etc., the honorification system is more or less a counterpart.\"", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "He introduces the feature HON as a head-feature (with values +/-), underlying the head feature principle. This accounts for the fact that the value of the HON-feature passes from the head to the mother node. Gunji's HON, though, is only a representation of subject honorification. Honorification concerning the addressee or objects is not considered. The values of SHUN can be either plus or minus, but neutral forms also exist.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "Therefore we expanded the syntactic part of the representation of honorification. The lexical entries get a HEAD feature called FORMAL:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "[FORMAL AHON +I-11 SHON +", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "Only the connection of representing honorification on the syntactic and contextual level makes it possible to account for all phenomena. The pure syntactic representation cannot account for the representation of honorificational relations between speaker and addressee, for OHON and for multiple honorifications, while the pure contextual representation cannot account for the syntactic restrictions on subjects and relative sentences. The CONTEXT level gives information about felicity of an utterance, while the CAT level gives information about syntactical correctness of an utterance. For honorification in Japanese, we need both. With the fundamental concept of HPSG, the sign, it is possible to incorporate both levels of linguistic analysis.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "Being a HEAD feature, the value of FORMAL is passed up from head daugthers to mother daughters. A honorific noun therefore contains the value SY NSEMI LOCAL1 CAT I HEAD FORMAL' SHONpius, as well as a verb with subject honorification. For Japanese, we set up the This principle accounts for the compatibility of the information on the syntactic (CAT) and contextual (CONTEXT) levels. While the agreement of subject and verb is checked on the syntactic level, the contextual level gets the information on subject honorification and links it to the semantic entities.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "Honorification concerning the addressee inside the sentence is seen as a purely syntactic restriction. As non-addressee-honorific and addressee-honorific verbs may combine, it is not useful to introduce the relation into the context during processing the sentence. The syntactic restriction is needed for relative sentences, as shown above. At the top-most node (utterance-type in our grammar of spoken language), the addr-honor-rel is introduced into the CONTEXT' BACKGR. Its polarity is co-indexed with the value of HEAD -DTRISYNSE11/11LOCALICATIHEADI FORMAL1AHON. The HONORER is coindexed with the speaker, the HONORED is co-indexed with the addressee. Also here it can be seen that it is meaningful to represent the honorification on both levels. Inside the sentence, it is a purely syntactic relation, but outside, it is a contextual relation.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "While the syntactic information goes up the tree via the head feature principle, the contextual information underlies different principles. The HPSG principle of contextual consistency ( [Pollard and Sag1994] ,p.333) says:", |
| "cite_spans": [ |
| { |
| "start": 187, |
| "end": 208, |
| "text": "[Pollard and Sag1994]", |
| "ref_id": null |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "\"The CONTEXT (BACKGR value of a given phrase is the union of the CONTEXT (BACKGR values of the daughters.\"", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "This must be modified for our approach, since the head-subject rule takes the CONTEXT (BACKGR value of its head daughter. It can be hold for all structures besides the head-subject rule and the utterance rule, as shown before.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "Let us take an example for multiple honorification, 7, that shall be repeated here as 17:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "(17) watakushi qa o-denwa itashi-mashi-ta INOM telephone do(hon)-hon-Past", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "The self-referring pronoun watakushi introduces an obj-honor-rel with POLARITY minus, where HONORER and HONORED are co-indexed with the speaker and the CONTENTIINDEX. This is passed up the tree in the head-complement structure of watakushi ga. At the same time, the values of HEAD1FORM AL are introduced: AHON plus and SHON minus. As particles are assumed to be heads (see [Siegell999] ), they must take their SHON value from their objects (which is defined in the lexical type of particles).", |
| "cite_spans": [ |
| { |
| "start": 373, |
| "end": 385, |
| "text": "[Siegell999]", |
| "ref_id": "BIBREF17" |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "The honorific form o-denwa itashi-mashi-ta introduces a subj-honor-rel in the context with POLAR-ITY minus. The HONORED is co-indexed with the subject's CONTENT1INDEX. The HEAD( FORMAL values are the same as the ones of watakushi. The principle of subject honorification sets up the restrictions for the predicate's subject. As this is found in watakushi ga, the HEAD1FORM AL values are unified and the subj-honor-rel is introduced. The utterance-rule introduces an addr-honor-rel with POLARITY plus, since the value of HEAD(FORMAL(AHON is plus.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "This was an example of the special case where the speaker is the subject. Another example with the addressee being the subject is:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "(18) anata you NOM o-denwa telephone AKK kudasai-mashi-ta do(shon)-ahon-Past", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "All three types of honorificational relations are introduced here: subject honorification by the addressee-refering pronoun anata, object honorification by the honorific noun o-denwa and addressee honorification by the -mashita ending of the verb. The polarity is plus in all cases.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Previous Approaches", |
| "sec_num": "4" |
| }, |
| { |
| "text": "The CONTEXT (BACKGR value passes up the tree, independend which daughter is the head of the phrase. It is even possible to represent the honorification in cases of embedded phrases. There can be more than one relation of obj-honor-rel in an utterance, as there can be more than one honored constituents. An effect for the machine translation system is that lexical pragmatic restrictions for zero pronominals can be directly accounted for in the analysis. They are essential to find referents for many zero pronominals, as is shown by [Metzing and Siege11994] . See for example:", |
| "cite_spans": [ |
| { |
| "start": 535, |
| "end": 559, |
| "text": "[Metzing and Siege11994]", |
| "ref_id": null |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Effects", |
| "sec_num": "6" |
| }, |
| { |
| "text": "ornachz shite-orimasu: study do-hon", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Effects", |
| "sec_num": "6" |
| }, |
| { |
| "text": "This is part of the structure for this utterance:", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Effects", |
| "sec_num": "6" |
| }, |
| { |
| "text": "The structure restricts the subject to one with a subject honorification with negative polarity. Thus, only the speaker or a person of the speaker's social group can be the antecedent of the subject.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Effects", |
| "sec_num": "6" |
| }, |
| { |
| "text": "Syntactic restrictions for relative sentences can easily be formulated in a way that only verbs with a non-addressee-honorific form can modify nouns.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Effects", |
| "sec_num": "6" |
| }, |
| { |
| "text": "We randomly chose 100 utterances from the Verbmobil corpus. Then we tagged these with expected values for SHON, OHON and AHON. The utterances contained 170 occurrences of honorification, with 99 AHON, 32 OHON and 39 SHON. We parsed the utterances and compared the human-made tagging with the parsing result. Then we calculated precision and recall in the following way: The sentence is ambiguous, because it allows a first interpretation where Sie is a third person plural pronoun and therefore refers to a group of people and a second interpretation where it is a polite second person singular pronoun and refers to a single person. Honorification in German thus introduces OHON with honorific pronouns, but no special treatment of subjects and no AHON relation. The agreement between subject and verb is a purely syntactic one. French honorification shows different habits in agreement (as is shown by [Pollard and Sag1994] ,p.96f.), but as well concerns only the OHON dimension.", |
| "cite_spans": [ |
| { |
| "start": 904, |
| "end": 925, |
| "text": "[Pollard and Sag1994]", |
| "ref_id": null |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Evaluation", |
| "sec_num": "7" |
| }, |
| { |
| "text": "number", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Evaluation", |
| "sec_num": "7" |
| }, |
| { |
| "text": "Honorification in Korean, as it is described by [Lee1996] , is distinct form Japanese honorification in one point: There are no neutral forms of NPs and VPs in respect to subject honorification.", |
| "cite_spans": [ |
| { |
| "start": 48, |
| "end": 57, |
| "text": "[Lee1996]", |
| "ref_id": null |
| } |
| ], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Evaluation", |
| "sec_num": "7" |
| }, |
| { |
| "text": "Our approach thus seems to work for different kind of languages that express honorification.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Evaluation", |
| "sec_num": "7" |
| } |
| ], |
| "back_matter": [ |
| { |
| "text": "The Japanese language has a complicated system to express the social relation between speaker, addressee and subject of an utterance. This relation is expressed by honorification. It concerns verbal forms, verbal conjugations, nominal prefixes and pronouns and undergoes syntactic, semantic, pragmatic and domainspecific restrictions.We have shown that for Japanese it is necessary to distinguish subject honorification, object honorification and addressee honorification and to introduce polarity for these. The number and kind of the dimensions is language-specific; German and French, for example, have only one dimension, while Korean and Japanese have three. In one utterance different dimensions of honorification can be expressed.We have given a treatment of honorifics in the HPSG framework that covers all three dimensions of Japanese honorifics and makes it possible to account for honorific agreement as well as restrictions in complement sentences and restrictions for zero pronominals. The approach allows an uniform treatment of honorific dimensions in different languages.", |
| "cite_spans": [], |
| "ref_spans": [], |
| "eq_spans": [], |
| "section": "Conclusions", |
| "sec_num": "9" |
| } |
| ], |
| "bib_entries": { |
| "BIBREF0": { |
| "ref_id": "b0", |
| "title": "HOflichkeit und soziale Bedeutung im Japanischen. Linguistische Berichte", |
| "authors": [ |
| { |
| "first": "Florian", |
| "middle": [], |
| "last": "Coulmas", |
| "suffix": "" |
| } |
| ], |
| "year": 1987, |
| "venue": "", |
| "volume": "107", |
| "issue": "", |
| "pages": "44--62", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Florian Coulmas. HOflichkeit und soziale Bedeutung im Japanischen. Linguistische Berichte, 107:44-62, 1987.", |
| "links": null |
| }, |
| "BIBREF1": { |
| "ref_id": "b1", |
| "title": "Identifying the referents of zero-pronouns in Japanese based on pragmatic constraint interpretation", |
| "authors": [ |
| { |
| "first": "Kohji", |
| "middle": [], |
| "last": "Dohsaka", |
| "suffix": "" |
| } |
| ], |
| "year": 1990, |
| "venue": "Proceedings of ECAI 1990", |
| "volume": "", |
| "issue": "", |
| "pages": "240--245", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Kohji Dohsaka. Identifying the referents of zero-pronouns in Japanese based on pragmatic con- straint interpretation. In Proceedings of ECAI 1990, pages 240-245, 1990.", |
| "links": null |
| }, |
| "BIBREF2": { |
| "ref_id": "b2", |
| "title": "The structure of context: The representation of pragmatic restrictions in HPSG", |
| "authors": [ |
| { |
| "first": "Georgia", |
| "middle": [ |
| "M" |
| ], |
| "last": "Green", |
| "suffix": "" |
| } |
| ], |
| "year": null, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Georgia M. Green. The structure of context: The representation of pragmatic restrictions in HPSG.", |
| "links": null |
| }, |
| "BIBREF3": { |
| "ref_id": "b3", |
| "title": "Studies in the Linguistic Sciences. Proceedings of the 5th annual meeting of the Formal Linguistics Society of the Midwest", |
| "authors": [], |
| "year": 1997, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "In James Yoon, editor, Studies in the Linguistic Sciences. Proceedings of the 5th annual meeting of the Formal Linguistics Society of the Midwest, 1997.", |
| "links": null |
| }, |
| "BIBREF4": { |
| "ref_id": "b4", |
| "title": "Hanaoka McGloin1976] Naomi Hanaoka McGloin. Negation", |
| "authors": [ |
| { |
| "first": "Takao", |
| "middle": [], |
| "last": "Gunji", |
| "suffix": "" |
| } |
| ], |
| "year": 1976, |
| "venue": "Japanese Generative Grammar", |
| "volume": "5", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Takao Gunji. Japanese Phrase Structure Grammar. Dordrecht: Reidel, 1987. [Hanaoka McGloin1976] Naomi Hanaoka McGloin. Negation. In Masayoshi Shibatani, editor, Japanese Genera- tive Grammar, volume 5 of Syntax and Semantics. New York, 1976.", |
| "links": null |
| }, |
| "BIBREF6": { |
| "ref_id": "b6", |
| "title": "Universals of linguistic politeness. quantitative evidence from Japanese and American english", |
| "authors": [ |
| { |
| "first": "Beverly", |
| "middle": [], |
| "last": "Hill", |
| "suffix": "" |
| }, |
| { |
| "first": "Sachiko", |
| "middle": [], |
| "last": "Ide", |
| "suffix": "" |
| }, |
| { |
| "first": "Shoko", |
| "middle": [], |
| "last": "Ikuta", |
| "suffix": "" |
| }, |
| { |
| "first": "Akiko", |
| "middle": [], |
| "last": "Kawasaki", |
| "suffix": "" |
| }, |
| { |
| "first": "Tsunao", |
| "middle": [], |
| "last": "Ogino", |
| "suffix": "" |
| } |
| ], |
| "year": 1986, |
| "venue": "Journal of Pragmatics", |
| "volume": "10", |
| "issue": "", |
| "pages": "347--371", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "et a/.1986] Beverly Hill, Sachiko Ide, Shoko Ikuta, Akiko Kawasaki, and Tsunao Ogino. Universals of linguis- tic politeness. quantitative evidence from Japanese and American english. Journal of Pragmatics, 10:347-371, 1986.", |
| "links": null |
| }, |
| "BIBREF7": { |
| "ref_id": "b7", |
| "title": "A sociolinguistic analysis of the Japanese honorifics", |
| "authors": [ |
| { |
| "first": "Motoko", |
| "middle": [], |
| "last": "Hori", |
| "suffix": "" |
| } |
| ], |
| "year": 1986, |
| "venue": "Journal of Pragmatics", |
| "volume": "10", |
| "issue": "", |
| "pages": "373--386", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Motoko Hori. A sociolinguistic analysis of the Japanese honorifics. Journal of Pragmatics, 10:373-386, 1986.", |
| "links": null |
| }, |
| "BIBREF8": { |
| "ref_id": "b8", |
| "title": "Subjectlessness and honorifics in Japanese: A case of textual construal", |
| "authors": [ |
| { |
| "first": "Motoko", |
| "middle": [], |
| "last": "Hori", |
| "suffix": "" |
| } |
| ], |
| "year": 1992, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Motoko Hori. Subjectlessness and honorifics in Japanese: A case of textual construal., 1992.", |
| "links": null |
| }, |
| "BIBREF9": { |
| "ref_id": "b9", |
| "title": "The background of Japanese sociolinguistics", |
| "authors": [ |
| { |
| "first": "Sachiko", |
| "middle": [], |
| "last": "Ide", |
| "suffix": "" |
| } |
| ], |
| "year": 1986, |
| "venue": "Journal of Pragmatics", |
| "volume": "10", |
| "issue": "", |
| "pages": "281--286", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Sachiko Ide. The background of Japanese sociolinguistics. Journal of Pragmatics, 10:281-286, 1986.", |
| "links": null |
| }, |
| "BIBREF10": { |
| "ref_id": "b10", |
| "title": "Japanese honorific systems", |
| "authors": [ |
| { |
| "first": "Akira", |
| "middle": [], |
| "last": "Ikeya", |
| "suffix": "" |
| } |
| ], |
| "year": 1983, |
| "venue": "Seoul Papers in Formal Grammar Theory. Proceedings of the 3rd Korean-Japanese Joint Workshop", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Akira Ikeya. Japanese honorific systems. In Seoul Papers in Formal Grammar Theory. Proceedings of the 3rd Korean-Japanese Joint Workshop. Seoul: Hanshin Publishing Company, 1983.", |
| "links": null |
| }, |
| "BIBREF11": { |
| "ref_id": "b11", |
| "title": "The Structure of Japanese Language", |
| "authors": [ |
| { |
| "first": "Susumo", |
| "middle": [], |
| "last": "Kuno", |
| "suffix": "" |
| } |
| ], |
| "year": 1973, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Susumo Kuno. The Structure of Japanese Language. Cambridge, Mass.: MIT Press, 1973.", |
| "links": null |
| }, |
| "BIBREF12": { |
| "ref_id": "b12", |
| "title": "An HPSG account of the Korean honorification system", |
| "authors": [ |
| { |
| "first": "Dong-Young", |
| "middle": [], |
| "last": "Lee", |
| "suffix": "" |
| } |
| ], |
| "year": 1996, |
| "venue": "Studies in HPSG", |
| "volume": "12", |
| "issue": "", |
| "pages": "165--190", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Dong-Young Lee. An HPSG account of the Korean honorification system. In Claire Grover and Enric Vallduvo, editors, Studies in HPSG, volume 12, pages 165-190. Centre for Cognitive Science, Univ.of Edinburgh., 1996.", |
| "links": null |
| }, |
| "BIBREF13": { |
| "ref_id": "b13", |
| "title": "The role of WA in negation", |
| "authors": [ |
| { |
| "first": "Naomi", |
| "middle": [], |
| "last": "Hanaoka Mcgloin", |
| "suffix": "" |
| } |
| ], |
| "year": 1987, |
| "venue": "Perspectives on Topicalization. The Case of Japanese 'WA'", |
| "volume": "", |
| "issue": "", |
| "pages": "165--184", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Naomi Hanaoka McGloin. The role of WA in negation. In John Hinds, Senko Kumiya Maynard, and Shoichi Iwasaki, editors, Perspectives on Topicalization. The Case of Japanese 'WA'., pages 165-184. Amsterdam: John Benjamins Publishing Company, 1987.", |
| "links": null |
| }, |
| "BIBREF14": { |
| "ref_id": "b14", |
| "title": "Zero pronoun processing: Some requirements for a Verbmobil system", |
| "authors": [ |
| { |
| "first": "D", |
| "middle": [], |
| "last": "Metzing", |
| "suffix": "" |
| }, |
| { |
| "first": "M", |
| "middle": [], |
| "last": "Siegel", |
| "suffix": "" |
| } |
| ], |
| "year": 1994, |
| "venue": "Verbmobil-Memo", |
| "volume": "46", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "[Metzing and Siege11994] D. Metzing and M. Siegel. Zero pronoun processing: Some requirements for a Verbmobil system. Verbmobil-Memo 46, Universitat Bielefeld, 1994.", |
| "links": null |
| }, |
| "BIBREF15": { |
| "ref_id": "b15", |
| "title": "Head-Driven Phrase Structure Grammar", |
| "authors": [ |
| { |
| "first": "C", |
| "middle": [], |
| "last": "Pollard", |
| "suffix": "" |
| }, |
| { |
| "first": "I", |
| "middle": [ |
| "A" |
| ], |
| "last": "Sag", |
| "suffix": "" |
| } |
| ], |
| "year": 1994, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "[Pollard and Sag1994] C. Pollard and I.A. Sag. Head-Driven Phrase Structure Grammar. Chicago: University of Chicago Press, 1994.", |
| "links": null |
| }, |
| "BIBREF16": { |
| "ref_id": "b16", |
| "title": "Die Maschinelle Ubersetzung aufgabenorientierter japanisch-deutscher Dialoge. LOsungen fiir Translation Mismatches", |
| "authors": [ |
| { |
| "first": "Melanie", |
| "middle": [], |
| "last": "Siegel", |
| "suffix": "" |
| } |
| ], |
| "year": 1997, |
| "venue": "", |
| "volume": "", |
| "issue": "", |
| "pages": "", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Melanie Siegel. Die Maschinelle Ubersetzung aufgabenorientierter japanisch-deutscher Dialoge. LOsungen fiir Translation Mismatches. Logos Verlag Berlin, 1997.", |
| "links": null |
| }, |
| "BIBREF17": { |
| "ref_id": "b17", |
| "title": "The syntactic processing of particles in Japanese spoken language", |
| "authors": [ |
| { |
| "first": "Melanie", |
| "middle": [], |
| "last": "Siegel", |
| "suffix": "" |
| } |
| ], |
| "year": 1999, |
| "venue": "Proceedings of the 13th Pacific Asia Conference on Language, Information and Computation", |
| "volume": "", |
| "issue": "", |
| "pages": "313--320", |
| "other_ids": {}, |
| "num": null, |
| "urls": [], |
| "raw_text": "Melanie Siegel. The syntactic processing of particles in Japanese spoken language. In Jhing-Fa Wang and Chung-Hsien Wu, editors, Proceedings of the 13th Pacific Asia Conference on Language, Information and Computation, pages 313-320, 1999.", |
| "links": null |
| } |
| }, |
| "ref_entries": { |
| "FIGREF1": { |
| "text": "Classification of honorifics by Haradaextracted from the dialogue and on the other hand restricts the interpretation of zero pronominals in the dialogue. This approach shows that the extraction of information about honorification from the dialogue is urgently needed for the interpretation of zero pronominals.", |
| "num": null, |
| "type_str": "figure", |
| "uris": null |
| }, |
| "FIGREF2": { |
| "text": "principle of subject honorification: In a honorific lexical structure, the FORM ALISHON value of the HEAD is identical to the FORM ALISHON value of the subject's HEAD and the polarity of the subj-honor-rel in BACKGR. The values of the subject's CONTENTIINDEX and the HONORED of the sub j-honor-rel in BACKGR are identical.", |
| "num": null, |
| "type_str": "figure", |
| "uris": null |
| }, |
| "TABREF2": { |
| "content": "<table><tr><td/><td/><td/><td>:</td><td/><td/><td/><td/></tr><tr><td colspan=\"2\">(11) Taro Taro</td><td>TOP</td><td>Hanako Hanako</td><td>t /1-</td><td>ki-mashi-ta come-hon-Past</td><td colspan=\"2\">to COMP</td><td>it-ta say-Past</td></tr><tr><td colspan=\"2\">(12) Taro r r 0</td><td>TOP</td><td>Hanako Hanako</td><td>m</td><td>ki-ta come-Past</td><td>to COMP</td><td colspan=\"2\">it-ta say-Past</td></tr><tr><td colspan=\"2\">(13) *Taro Taro</td><td>wa TOP</td><td>Hanako Hanako</td><td>NOM</td><td colspan=\"2\">ki-mashi-ta come-hon-Past</td><td>koto NOM</td><td>AKK</td></tr><tr><td/><td colspan=\"2\">shiranakat-ta not know-Past</td><td/><td/><td/><td/><td/></tr><tr><td>(14)</td><td>Taro Taro</td><td>wa TOP</td><td>Hanako Hanako</td><td>Ng(5M</td><td colspan=\"2\">ki-ta come-Past</td><td>koto NOM</td><td>AKK</td></tr><tr><td/><td colspan=\"2\">shiranakat-ta not know-Past</td><td/><td/><td/><td/><td/></tr></table>", |
| "text": "Complement sentences allow a honorific predicate, while relative sentences do not.", |
| "type_str": "table", |
| "num": null, |
| "html": null |
| }, |
| "TABREF3": { |
| "content": "<table><tr><td/><td>Holorifics</td></tr><tr><td colspan=\"2\">Propositional</td><td>Performative</td></tr><tr><td>honorifics</td><td/><td>honorifics</td></tr><tr><td>1 Subject</td><td>1 Object</td></tr><tr><td>honorifics</td><td>honorifics</td></tr><tr><td>o-hanasi ni nar-u,etc.</td><td/></tr></table>", |
| "text": ",[Hill et a/.1986] and[Hanaoka McGloin1976]. The authors state that honorification is an expression of the social distance or 'perceived distance' ([Hill et a/.1986]) between speaker and addressee and the belonging to a social group([Coulmas1987]). They investigate the relation between gender and the use of honorificational expressions ([Hori1986]). Examples for a grammatical investigation of Japanese honorification are[Ikeya1983],[Kuno1973],[Harada1976] and[Hori1992].[Hori1986] uses honorification for a definition of 'subject' in Japanese.[Kuno1973] classifies honorification concerning style and honorification concerning respect. In our approach, these classes are AHON and SHON, respectively. He shows that there are differences of grade in various expressions of honorification.[Harada1976] gives a classification of honorificational forms that at first sight seems complementary to ours. It can be seen in figure 1. A closer look shows that the difference is only a question of naming. Haradas 'Subject honorifics' is [-I-SHON] in our approach, the 'Object honorifics' is[-SHON] and the 'Performative honorifics' would correspond to our [+AHON]. What we call [OHON], turns into [SHON], if the entity is used as a subject in the utterance. [Ikeya1983] gives a GPSG account for honorification, where [+SHON] and [-SHON] (called OHON in his approach) are head features, with the head feature principle accounting for the agreement restrictions on subject honorification. [Gunji1987] also gives examples for syntactic restrictions on honorification and introduces HON as a head feature.The machine translational viewpoint is shown by[Dohsaka1990]. Dohsaka describes, how information about honorification can be used in the machine translation system to resolve zero pronominal references to human entities. He builds up a model of social relations during processing the dialogue, where the pragmatic relations honorification, speaker's point of view and territory of information is on the one hand", |
| "type_str": "table", |
| "num": null, |
| "html": null |
| }, |
| "TABREF8": { |
| "content": "<table><tr><td/><td/><td/><td colspan=\"2\">PRECISION RECALL</td></tr><tr><td>(19)</td><td/><td>AHON</td><td>1</td><td>1</td></tr><tr><td/><td/><td>SHON</td><td>1</td><td>0.86</td></tr><tr><td/><td/><td>OHON</td><td>1</td><td>0.79</td></tr><tr><td/><td/><td>HON</td><td>1</td><td>0.93</td></tr><tr><td/><td/><td colspan=\"3\">Table 1: Precision and recall</td></tr><tr><td>Precision =</td><td/><td colspan=\"2\">of correct assigned honorifications number of assigned honorifications</td></tr><tr><td colspan=\"4\">number of correct assigned honorifications</td></tr><tr><td>example shows:</td><td/><td/><td/></tr><tr><td>(20) g</td><td>sind are</td><td>nett nice</td><td/></tr></table>", |
| "text": "= number of honori fications in the corpusThe results can be seen in table 1.8 Honorification in Other LanguagesHonorification in German concerns only the relation between speaker and addressee, as the following", |
| "type_str": "table", |
| "num": null, |
| "html": null |
| } |
| } |
| } |
| } |
