File size: 27,471 Bytes
2935d2a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b221a3d
2935d2a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
---
license:
  - other
license_details: >-
  Mixed. Every document carries its own license in the `license` and
  `license_family` columns (CC BY-SA 4.0/3.0/2.0, CC BY-NC 3.0, CC0 1.0,
  CC BY 4.0/3.0, CC BY-NC-SA 4.0, Project Gutenberg license, and a few
  library-specific rights statements). Filter by `license_family` before
  reuse.
language:
  - de
  - la
  - gml
language_details: deu, lat, gml (Middle Low German); Early New High German and New High German texts, tagged as deu
multilinguality:
  - monolingual
size_categories:
  - 1K<n<10K
pretty_name: Deutsches Textarchiv (DTA) Documents
source_datasets:
  - original
annotations_creators:
  - expert-generated
language_creators:
  - found
tags:
  - historical
  - historical-german
  - early-new-high-german
  - german
  - dta
  - deutsches-textarchiv
  - tei
  - tcf
  - fraktur
  - ocr
  - orthographic-normalization
  - spelling-normalization
  - cab
  - digital-humanities
  - 15th-century
  - 16th-century
  - 17th-century
  - 18th-century
  - 19th-century
  - 20th-century
configs:
  - config_name: documents
    default: true
    data_files:
      - split: train
        path: train-*.parquet
dataset_info:
  config_name: documents
  features:
    - name: id
      dtype: string
    - name: dta_id
      dtype: string
    - name: urn
      dtype: string
    - name: url
      dtype: string
    - name: title
      dtype: string
    - name: subtitle
      dtype: string
    - name: volume
      dtype: string
    - name: part
      dtype: string
    - name: authors
      list:
        - name: surname
          dtype: string
        - name: forename
          dtype: string
        - name: gnd
          dtype: string
    - name: date_published
      dtype: string
    - name: year
      dtype: int64
    - name: place_of_publication
      dtype: string
    - name: publisher
      dtype: string
    - name: edition
      dtype: string
    - name: num_pages
      dtype: int64
    - name: repository
      dtype: string
    - name: shelfmark
      dtype: string
    - name: typeface
      dtype: string
    - name: bibl
      dtype: string
    - name: genre_dtamain
      dtype: string
    - name: genre_dtasub
      dtype: string
    - name: genre_dwds1main
      dtype: string
    - name: genre_dwds1sub
      dtype: string
    - name: dta_corpus_labels
      list: string
    - name: language
      dtype: string
    - name: language_note
      dtype: string
    - name: license
      dtype: string
    - name: license_family
      dtype: string
    - name: num_images
      dtype: int64
    - name: num_tokens
      dtype: int64
    - name: num_types
      dtype: int64
    - name: num_characters
      dtype: int64
    - name: num_sentences
      dtype: int64
    - name: text
      dtype: string
    - name: text_source
      dtype: string
    - name: text_normalized
      dtype: string
  splits:
    - name: train
      num_examples: 5480
---

# Deutsches Textarchiv (DTA) Documents

This datasets hosts all documents from the Deutsches Textarchiv (DTA).

One row per work of the [Deutsches Textarchiv (DTA)](https://www.deutschestextarchiv.de/), built from the
official DTA dump `dta_komplett_2026-02-10` (TCF format). The dataset contains **5,480 documents** spanning
**1472 to 1987** with about **205M tokens** and **1.4B characters** of German text, with two text views per document:

* `text`: the historical, layout-faithful transcription (line breaks, long s `ſ`, combining diacritics, original spelling);
* `text_normalized`: the same text in modern orthography, produced by applying the DTA CAB orthography corrections
  (`operation="replace"`) to the token layer.

Rich bibliographic metadata comes directly from the embedded TEI header: authors with GND identifiers, original
publication date and place, publisher, holding library and shelfmark, typeface, four genre schemes, DTA subcorpus
labels and the per-work license. Licenses vary across the corpus, so please check the `license_family` column
before reuse (see [Licensing](#licensing)).

## Dataset structure

### Configs and splits

| Config | Split | Rows | Files |
|---|---|---:|---|
| `documents` (default) | `train` | 5,480 | `documents/train-0000[0-5].parquet` (6 shards, zstd-compressed Parquet) |

There is only one split, because the dataset is a corpus and not a benchmark.

### Loading

```python
from datasets import load_dataset

ds = load_dataset("histde/dta-documents", "documents", split="train")
print(ds[0]["title"], ds[0]["year"])

# Only permissively licensed works, as a streaming iterator
permissive = load_dataset("histde/dta-documents", "documents", split="train", streaming=True)
permissive = permissive.filter(lambda d: d["license_family"] in {"cc0-1.0", "cc-by-4.0", "cc-by-3.0"})
```

With Polars, straight from the Parquet shards:

```python
import polars as pl

df = pl.read_parquet("hf://datasets/histde/dta-documents/documents/train-*.parquet")
df.filter(pl.col("year") < 1700).select("id", "title", "license_family", "num_tokens")
```

### Schema

| Column | Type | Description | Example |
|---|---|---|---|
| `id` | `string` | DTA directory name, used as the primary document identifier (`DTADirName`). Falls back to the TCF file name | `goethe_faust01_1808` |
| `dta_id` | `string` | Numeric DTA identifier (`DTAID`) | `200006` |
| `urn` | `string` | Persistent URN of the work | `urn:nbn:de:kobv:b4-200006-1` |
| `url` | `string` | Landing page of the work at deutschestextarchiv.de | `https://www.deutschestextarchiv.de/343016` |
| `title` | `string` | Main title(s) from the TEI title statement. Multiple main titles are joined with `\n` | `Iphigenie auf Tauris` |
| `subtitle` | `string` | Subtitle(s), joined with `\n`. Null if absent | `Ein Schauspiel` |
| `volume` | `string` | Volume title (text of `<title type="volume">`). Null if absent | `Dritter Theil` |
| `part` | `string` | Part title (text of `<title type="part">`). Null if absent | `Die sentimentalischen Dichter` |
| `authors` | `list<struct>` | Authors as `{surname, forename, gnd}` structs. `gnd` is the GND/PND URI, may be null. Empty list for anonymous works | `[{"surname": "Reideburg", "forename": "Christoph von", "gnd": "http://d-nb.info/gnd/128862580"}]` |
| `date_published` | `string` | Raw publication date string of the original print, as given in the header | `1642` |
| `year` | `int64` | First four-digit number parsed from `date_published`. Null if none found | `1642` |
| `place_of_publication` | `string` | Place of publication of the original print | `Breslau` |
| `publisher` | `string` | Publisher name(s), joined with `\n` | `Georgius Baumann` |
| `edition` | `string` | Edition statement (text, or the `n` attribute if the element has no text) | `2. Auflage` |
| `num_pages` | `int64` | Page count of the original source (`measure type="pages"`) | `72` |
| `repository` | `string` | Holding library of the digitized copy | `Universitätsbibliothek Breslau` |
| `shelfmark` | `string` | Shelfmark of the copy at the repository | `Universitätsbibliothek Breslau, 4 A 277/11 / 343016` |
| `typeface` | `string` | Typeface of the original print | `Fraktur` |
| `bibl` | `string` | Short bibliographic citation string | `Reideburg, Christoph von: Kurtze Anleitung: Wie die jetzige böse Zeit/ ... Breslau, 1642.` |
| `genre_dtamain` | `string` | Main genre, DTA scheme | `Gebrauchsliteratur` |
| `genre_dtasub` | `string` | Sub genre, DTA scheme | `Leichenpredigt` |
| `genre_dwds1main` | `string` | Main genre, DWDS scheme | `Belletristik` |
| `genre_dwds1sub` | `string` | Sub genre, DWDS scheme | `Prosa` |
| `dta_corpus_labels` | `list<string>` | DTA (sub)corpus membership labels (`DTACorpus` class codes) | `["ready", "core"]` |
| `language` | `string` | ISO 639-3 code of the primary language | `deu` |
| `language_note` | `string` | Human-readable language label from the header | `(Früh-)Neuhochdeutsch` |
| `license` | `string` | Per-work license URL exactly as given in the header. Varies across the corpus | `http://creativecommons.org/licenses/by-sa/3.0/de/` |
| `license_family` | `string` | Normalized license identifier derived from `license` (http/https, `/de/`, `deed` variants collapsed) | `cc-by-sa-3.0` |
| `num_images` | `int64` | Number of page images (DTA extent measure) | `72` |
| `num_tokens` | `int64` | Number of tokens (DTA extent measure) | `11515` |
| `num_types` | `int64` | Number of word types (DTA extent measure) | `3440` |
| `num_characters` | `int64` | Number of characters (DTA extent measure) | `78795` |
| `num_sentences` | `int64` | Sentence count from the `full/` sentence layer. Null if the work has no `full/` file | `842` |
| `text` | `string` | Historical transcription. Layout-faithful `<text>` layer from `simple/` (line breaks, long s `ſ`, combining diacritics), or space-joined tokens if only `full/` exists | `Kurtze Anleitung:\nWie die jetzige boͤſe Zeit/ darinnen zwar fuͤr ſich ſelbſt/ ...` |
| `text_source` | `string` | Origin of `text`: `layout` (from `simple/`) or `tokens` (reconstructed from the `full/` token layer) | `layout` |
| `text_normalized` | `string` | Modern orthography: `full/` token layer with CAB `replace` corrections applied, space-joined. Null if the work has no `full/` file | `Kurze Anleitung : Wie die jetzige böse Zeit / darinnen zwar für sich selbst / ...` |

### Example row

```json
{
  "id": "343016",
  "dta_id": "200006",
  "urn": "urn:nbn:de:kobv:b4-200006-1",
  "url": "https://www.deutschestextarchiv.de/343016",
  "title": "Kurtze Anleitung: Wie die jetzige böse Zeit/ darinnen zwar für sich selbst/ nichts/ alß eytel Klag/ Ach/ vnd Weh regieret",
  "authors": [{"surname": "Reideburg", "forename": "Christoph von", "gnd": "http://d-nb.info/gnd/128862580"}],
  "date_published": "1642",
  "year": 1642,
  "place_of_publication": "Breslau",
  "publisher": "Georgius Baumann",
  "repository": "Universitätsbibliothek Breslau",
  "typeface": "Fraktur",
  "genre_dtamain": "Gebrauchsliteratur",
  "genre_dtasub": "Leichenpredigt",
  "dta_corpus_labels": ["ready", "aedit"],
  "language": "deu",
  "language_note": "(Früh-)Neuhochdeutsch",
  "license": "http://creativecommons.org/licenses/by-sa/3.0/de/",
  "license_family": "cc-by-sa-3.0",
  "num_tokens": 11515,
  "num_sentences": 842,
  "text": "Kurtze Anleitung:\nWie die jetzige boͤſe Zeit/ darinnen zwar fuͤr ſich ſelbſt/\nnichts/ alß eytel Klag/ Ach/ vnd Weh regieret; ...",
  "text_source": "layout",
  "text_normalized": "Kurze Anleitung : Wie die jetzige böse Zeit / darinnen zwar für sich selbst / nichts / als eitel Klage / Ach / und Weh regieret ; ..."
}
```

## Dataset creation

### Source data

The DTA is a reference corpus of printed German from the late 15th to the early 20th century, curated by the
Berlin-Brandenburg Academy of Sciences and Humanities (BBAW). Texts were transcribed by double keying from
first editions wherever possible, and are published with TEI headers, sentence and token layers and
orthographic normalization from the CAB tool chain.

This dataset was built from the complete TCF dump `dta_komplett_2026-02-10`, which comes in two variants:

* `simple/` (5,107 files): CMDI metadata and the layout-faithful historical text (`<text>` layer);
* `full/` (5,471 files): CMDI metadata and the annotation layers (tokens, sentences, orthography).

The union of both variants gives 5,481 works. One work (`ford_pitty_1633`) is excluded because its TCF file is
not well-formed XML.

### Processing

* `text` comes from `simple/` where available (`text_source = "layout"`, 5,105 docs). For the 375 works only
  present in `full/`, it is reconstructed by space-joining the token layer (`text_source = "tokens"`), so
  line breaks and the original whitespace are lost there.
* `text_normalized` is the `full/` token layer with CAB `replace` corrections applied (multi-token corrections
  replace the whole span), space-joined. It is null for the 10 works only present in `simple/`.
* Metadata is deliberately kept un-flattened: authors are a list of structs, all four genre schemes are
  separate columns, the raw license URL is kept next to a normalized `license_family`.
* `num_tokens`, `num_types`, `num_characters`, `num_images` and `num_pages` are the extent measures from the
  DTA header, not recomputed values.

The conversion script and the statistics script are available in the
[dta-experiments](https://github.com/histde/dta-experiments) repository.

## Statistics

All tables below were produced with `dta_dataset_stats.py` from the released Parquet shards.

### Overview

| statistic | value |
|---|---|
| Documents | 5,480 |
| Documents with normalized text (full/ layer) | 5,470 (99.8%) |
| Documents with at least one author | 3,104 (56.6%) |
| Distinct authors (by GND id) | 1,317 |
| Year range | 1472-1987 (1 unknown) |
| Median year | 1836 |
| Tokens (header measure) | 204,582,701 |
| Types (header measure) | 33,741,287 |
| Characters (header measure) | 1,423,995,435 |
| Characters in `text` column | 1,380,752,264 |
| Characters in `text_normalized` column | 1,370,639,606 |
| Sentences (full/ layer) | 10,376,423 |
| Page images | 761,995 |
| Median tokens per document | 10,394 |
| Median pages (images) per document | 16 |

### Temporal distribution (50-year bins)

| period | docs | sentences | tokens | characters | token share |
|---|---:|---:|---:|---:|---:|
| 1450–1499 | 15 | 1,899 | 177,258 | 997,389 | 0.1% |
| 1500–1549 | 17 | 10,469 | 308,471 | 1,915,965 | 0.2% |
| 1550–1599 | 98 | 139,035 | 3,572,336 | 23,618,803 | 1.7% |
| 1600–1649 | 422 | 433,762 | 9,757,100 | 66,033,889 | 4.8% |
| 1650–1699 | 691 | 1,118,361 | 23,338,594 | 159,938,400 | 11.4% |
| 1700–1749 | 435 | 1,143,293 | 24,779,141 | 171,116,209 | 12.1% |
| 1750–1799 | 583 | 1,566,582 | 30,933,941 | 214,031,382 | 15.1% |
| 1800–1849 | 1,618 | 2,000,919 | 40,798,688 | 284,536,219 | 19.9% |
| 1850–1899 | 1,143 | 3,086,868 | 55,792,116 | 393,248,560 | 27.3% |
| 1900–1949 | 444 | 737,908 | 13,114,628 | 95,203,044 | 6.4% |
| 1950–1999 | 13 | 137,150 | 2,005,488 | 13,321,270 | 1.0% |
| unknown | 1 | 177 | 4,940 | 34,305 | 0.0% |
| **total** | **5,480** | **10,376,423** | **204,582,701** | **1,423,995,435** | **100%** |

### Text source

| text_source | docs | docs % | tokens | tokens % |
|---|---:|---:|---:|---:|
| `layout` | 5,105 | 93.2% | 185,164,389 | 90.5% |
| `tokens` | 375 | 6.8% | 19,418,312 | 9.5% |

### Language

| language | docs | docs % | tokens | tokens % |
|---|---:|---:|---:|---:|
| `deu` | 5,477 | 99.9% | 204,523,746 | 100.0% |
| `lat` | 2 | 0.0% | 56,714 | 0.0% |
| `gml` | 1 | 0.0% | 2,241 | 0.0% |

### Genre (DWDS main)

| genre_dwds1main | docs | docs % | tokens | tokens % |
|---|---:|---:|---:|---:|
| Zeitung | 2,036 | 37.2% | 17,152,284 | 8.4% |
| Gebrauchsliteratur | 1,718 | 31.4% | 58,283,320 | 28.5% |
| Wissenschaft | 953 | 17.4% | 87,148,247 | 42.6% |
| Belletristik | 773 | 14.1% | 41,998,850 | 20.5% |

<details>
<summary>Genre (DWDS sub, top 10)</summary>

| genre_dwds1sub | docs | docs % | tokens | tokens % |
|---|---:|---:|---:|---:|
| unknown | 2,036 | 37.2% | 17,152,284 | 8.4% |
| Leichenpredigt | 336 | 6.1% | 3,517,225 | 1.7% |
| Zeitschrift | 217 | 4.0% | 1,697,733 | 0.8% |
| Theologie | 214 | 3.9% | 10,486,746 | 5.1% |
| Brief | 204 | 3.7% | 84,325 | 0.0% |
| Roman | 198 | 3.6% | 13,211,036 | 6.5% |
| Gesellschaft | 165 | 3.0% | 3,852,413 | 1.9% |
| Novelle | 126 | 2.3% | 3,031,211 | 1.5% |
| Lyrik | 125 | 2.3% | 5,090,960 | 2.5% |
| Gelegenheitsschrift:Tod | 116 | 2.1% | 103,280 | 0.1% |
| *other (123 values)* | 1,743 | 31.8% | 146,355,488 | 71.5% |

</details>

<details>
<summary>Genre (DTA main)</summary>

| genre_dtamain | docs | docs % | tokens | tokens % |
|---|---:|---:|---:|---:|
| unknown | 3,468 | 63.3% | 69,372,311 | 33.9% |
| Fachtext | 743 | 13.6% | 78,106,238 | 38.2% |
| Belletristik | 561 | 10.2% | 33,626,235 | 16.4% |
| Gebrauchsliteratur | 536 | 9.8% | 22,981,508 | 11.2% |
| Wissenschaftliche Abhandlungen in Form gedruckter Briefe | 50 | 0.9% | 89,564 | 0.0% |
| Abhandlungen in Zeitschriften, Sammelbänden etc. | 36 | 0.7% | 145,342 | 0.1% |
| Ankündigungen, Berichtigungen und kurze Nachrichten | 29 | 0.5% | 25,687 | 0.0% |
| Berliner Akademiereden/-schriften und andere Reden | 17 | 0.3% | 71,027 | 0.0% |
| Rezensionen | 7 | 0.1% | 8,514 | 0.0% |
| Pariser Akademiereden/-schriften | 6 | 0.1% | 16,728 | 0.0% |
| Vorworte und andere Beiträge Humboldts in Schriften anderer Autoren, Lexikonartikel | 6 | 0.1% | 20,581 | 0.0% |
| Gelegenheitsschrift | 5 | 0.1% | 49,906 | 0.0% |
| Journalismus | 5 | 0.1% | 54,829 | 0.0% |
| Albumblätter | 3 | 0.1% | 812 | 0.0% |
| Gutachten | 3 | 0.1% | 6,339 | 0.0% |
| Sonderdrucke und andere Grenzfälle zu selbständig erschienenen Schriften | 2 | 0.0% | 3,358 | 0.0% |
| Wissenschaft | 2 | 0.0% | 3,124 | 0.0% |
| Sachliteratur | 1 | 0.0% | 598 | 0.0% |

</details>

<details>
<summary>Genre (DTA sub, top 10)</summary>

| genre_dtasub | docs | docs % | tokens | tokens % |
|---|---:|---:|---:|---:|
| unknown | 3,628 | 66.2% | 69,896,619 | 34.2% |
| Leichenpredigt | 334 | 6.1% | 3,495,630 | 1.7% |
| Roman | 163 | 3.0% | 11,534,905 | 5.6% |
| Prosa | 128 | 2.3% | 10,291,868 | 5.0% |
| Lyrik | 114 | 2.1% | 4,731,157 | 2.3% |
| Drama | 82 | 1.5% | 2,330,599 | 1.1% |
| Recht | 79 | 1.4% | 9,743,726 | 4.8% |
| Philosophie | 76 | 1.4% | 5,426,303 | 2.7% |
| Historiographie | 46 | 0.8% | 7,221,703 | 3.5% |
| Medizin | 46 | 0.8% | 6,013,804 | 2.9% |
| *other (86 values)* | 784 | 14.3% | 73,896,387 | 36.1% |

</details>

### Typeface

| typeface | docs | docs % | tokens | tokens % |
|---|---:|---:|---:|---:|
| Fraktur | 4,478 | 81.7% | 164,623,059 | 80.5% |
| Antiqua | 427 | 7.8% | 35,856,371 | 17.5% |
| Schwabacher | 338 | 6.2% | 724,934 | 0.4% |
| Handschrift | 218 | 4.0% | 1,315,690 | 0.6% |
| Jean-Paul-Fraktur | 9 | 0.2% | 1,409,611 | 0.7% |
| Current | 2 | 0.0% | 1,963 | 0.0% |
| Rotunda | 2 | 0.0% | 313,225 | 0.2% |
| Antiqua (Schreibmaschine) | 1 | 0.0% | 34,176 | 0.0% |
| Antiqua, Kursive | 1 | 0.0% | 18,883 | 0.0% |
| Antqiua | 1 | 0.0% | 3,527 | 0.0% |
| Frakur | 1 | 0.0% | 10,103 | 0.0% |
| Textur | 1 | 0.0% | 270,599 | 0.1% |
| unknown | 1 | 0.0% | 560 | 0.0% |

Values are taken verbatim from the headers, including the two typos (`Antqiua`, `Frakur`).

<details>
<summary>DTA corpus labels (a document can carry several)</summary>

| label | docs | docs % | tokens | tokens % |
|---|---:|---:|---:|---:|
| `ready` | 5,480 | 100.0% | 204,582,701 | 100.0% |
| `core` | 1,478 | 27.0% | 130,114,948 | 63.6% |
| `china` | 1,149 | 21.0% | 111,416,242 | 54.5% |
| `mkhz1` | 625 | 11.4% | 3,996,672 | 2.0% |
| `nrhz` | 530 | 9.7% | 4,290,483 | 2.1% |
| `aedit` | 335 | 6.1% | 3,508,098 | 1.7% |
| `mts` | 320 | 5.8% | 18,006,989 | 8.8% |
| `lefevre` | 319 | 5.8% | 476,390 | 0.2% |
| `dtae` | 244 | 4.5% | 21,475,420 | 10.5% |
| `mkhz2` | 222 | 4.1% | 2,169,371 | 1.1% |
| `correspondent` | 204 | 3.7% | 925,755 | 0.5% |
| `tevo` | 202 | 3.7% | 5,682,395 | 2.8% |
| `sanders-briefe` | 190 | 3.5% | 61,890 | 0.0% |
| `hab` | 184 | 3.4% | 10,256,959 | 5.0% |
| `augsburgerallgemeine` | 181 | 3.3% | 2,761,778 | 1.3% |
| `avh` | 172 | 3.1% | 443,668 | 0.2% |
| `wikisource` | 169 | 3.1% | 5,787,543 | 2.8% |
| `sbb_funeralschriften` | 112 | 2.0% | 77,981 | 0.0% |
| `tdef` | 109 | 2.0% | 940,772 | 0.5% |
| `novellenschatz` | 87 | 1.6% | 1,643,336 | 0.8% |
| `blumenbach` | 33 | 0.6% | 2,688,412 | 1.3% |
| `epoetics` | 23 | 0.4% | 3,074,730 | 1.5% |
| `frauenstudium` | 22 | 0.4% | 134,198 | 0.1% |
| `psyleko` | 19 | 0.3% | 2,075,500 | 1.0% |
| `ntsm` | 11 | 0.2% | 32,139 | 0.0% |
| `avhkv` | 9 | 0.2% | 636,539 | 0.3% |
| `briefjeanpaul` | 9 | 0.2% | 1,409,611 | 0.7% |
| `gutenberg_org` | 8 | 0.1% | 325,940 | 0.2% |
| `gutzkow` | 7 | 0.1% | 478,246 | 0.2% |
| `dwds1` | 5 | 0.1% | 465,688 | 0.2% |
| `gei` | 2 | 0.0% | 57,474 | 0.0% |
| `gutenberg_de` | 2 | 0.0% | 116,184 | 0.1% |
| `gwb` | 2 | 0.0% | 80,666 | 0.0% |
| `urmel` | 2 | 0.0% | 1,068 | 0.0% |
| `zbk` | 2 | 0.0% | 48,127 | 0.0% |
| `greflinger` | 1 | 0.0% | 9,079 | 0.0% |
| `grenzboten` | 1 | 0.0% | 116,208 | 0.1% |

The `core` label marks the curated DTA core corpus (1,478 works, 63.6% of all tokens), which is the
balanced selection by genre and period. The other labels denote extension corpora (DTAE) contributed by
partner projects.

</details>

<details>
<summary>Place of publication (top 10)</summary>

| place | docs | docs % | tokens | tokens % |
|---|---:|---:|---:|---:|
| Berlin | 604 | 11.0% | 30,616,946 | 15.0% |
| Köln | 532 | 9.7% | 4,682,997 | 2.3% |
| Leipzig | 425 | 7.8% | 41,950,384 | 20.5% |
| Hamburg | 330 | 6.0% | 5,514,413 | 2.7% |
| Augsburg | 256 | 4.7% | 4,581,351 | 2.2% |
| Altstrelitz | 172 | 3.1% | 57,702 | 0.0% |
| unknown | 160 | 2.9% | 980,933 | 0.5% |
| München | 153 | 2.8% | 3,434,662 | 1.7% |
| Frankfurt (Main) | 141 | 2.6% | 11,430,873 | 5.6% |
| Königsberg | 133 | 2.4% | 926,843 | 0.5% |
| *other (246 values)* | 2,574 | 47.0% | 100,405,597 | 49.1% |

</details>

### Most frequent authors (top 10)

| author | GND | docs | tokens |
|---|---|---:|---:|
| Humboldt, Alexander von | http://d-nb.info/gnd/118554700 | 183 | 1,504,828 |
| Sanders, Daniel | http://d-nb.info/gnd/119242044 | 173 | 69,412 |
| Dach, Simon | http://d-nb.info/gnd/11852321X | 112 | 77,981 |
| N. N. | | 51 | 1,882,041 |
| Blumenbach, Johann Friedrich | http://d-nb.info/gnd/116208503 | 35 | 2,814,553 |
| Sattler, Basilius | http://d-nb.info/gnd/116974788 | 26 | 352,365 |
| Goethe, Johann Wolfgang von | http://d-nb.info/gnd/118540238 | 24 | 1,064,748 |
| Herder, Johann Gottfried von | http://d-nb.info/gnd/118549553 | 23 | 690,276 |
| Grimm, Jacob | http://d-nb.info/gnd/118542257 | 17 | 2,441,309 |
| Fontane, Theodor | http://d-nb.info/gnd/118534262 | 16 | 1,262,140 |

## Licensing

The DTA publishes every work under its own license, and this dataset keeps that information per row in the
`license` (raw URL from the header) and `license_family` (normalized) columns. There is **no single license**
for the whole dataset. Note that only `cc0-1.0` is a public domain dedication; all `cc-by-*` licenses carry
binding conditions (attribution, share-alike, or non-commercial use), and about 28% of the documents are
restricted to non-commercial use (`cc-by-nc-3.0`, `cc-by-nc-sa-4.0`, `noc-nc-1.0`, `out-of-copyright-nc`).

| license family | docs | docs % | tokens | tokens % | license URLs |
|---|---:|---:|---:|---:|---|
| `cc-by-sa-4.0` | 2,644 | 48.2% | 139,433,878 | 68.2% | http://creativecommons.org/licenses/by-sa/4.0/<br>http://creativecommons.org/licenses/by-sa/4.0/deed.de/<br>https://creativecommons.org/licenses/by-sa/4.0/<br>https://creativecommons.org/licenses/by-sa/4.0/deed.de |
| `cc-by-nc-3.0` | 1,559 | 28.4% | 20,418,259 | 10.0% | http://creativecommons.org/licenses/by-nc/3.0/<br>http://creativecommons.org/licenses/by-nc/3.0/de<br>http://creativecommons.org/licenses/by-nc/3.0/de/ |
| `cc-by-sa-3.0` | 623 | 11.4% | 15,287,129 | 7.5% | http://creativecommons.org/licenses/by-sa/3.0/<br>http://creativecommons.org/licenses/by-sa/3.0/de<br>http://creativecommons.org/licenses/by-sa/3.0/de/<br>https://creativecommons.org/licenses/by-sa/3.0/<br>https://creativecommons.org/licenses/by-sa/3.0/de/ |
| `cc0-1.0` | 370 | 6.8% | 16,938,595 | 8.3% | http://creativecommons.org/publicdomain/zero/1.0/ |
| `cc-by-4.0` | 215 | 3.9% | 8,189,539 | 4.0% | http://creativecommons.org/licenses/by/4.0/<br>http://creativecommons.org/licenses/by/4.0/deed.de<br>https://creativecommons.org/licenses/by/4.0/<br>https://creativecommons.org/licenses/by/4.0/deed.de |
| `cc-by-sa-2.0` | 51 | 0.9% | 3,058,054 | 1.5% | http://creativecommons.org/licenses/by-sa/2.0/de<br>http://creativecommons.org/licenses/by-sa/2.0/de/ |
| `gutenberg` | 9 | 0.2% | 344,316 | 0.2% | https://web.archive.org/web/20180927123034/http://www.gutenberg.org/wiki/Gutenberg:The_Project_Gutenberg_License |
| `cc-by-3.0` | 3 | 0.1% | 571,346 | 0.3% | http://creativecommons.org/licenses/by/3.0/de/ |
| `cc-by-nc-sa-4.0` | 2 | 0.0% | 19,672 | 0.0% | https://creativecommons.org/licenses/by-nc-sa/4.0/<br>https://creativecommons.org/licenses/by-nc-sa/4.0/deed.de |
| `mdz-copyright` | 1 | 0.0% | 24,964 | 0.0% | http://mdz.bib-bvb.de/copyright.htm |
| `noc-nc-1.0` | 1 | 0.0% | 125,859 | 0.1% | http://rightsstatements.org/vocab/NoC-NC/1.0/ |
| `nug-kkn` | 1 | 0.0% | 55,702 | 0.0% | http://www.deutsche-digitale-bibliothek.de/lizenzen/nug-kkn/ |
| `out-of-copyright-nc` | 1 | 0.0% | 115,388 | 0.1% | https://www.europeana.eu/portal/de/rights/out-of-copyright-non-commercial.html |
| **total** | **5,480** | **100%** | **204,582,701** | **100%** | |

To select a subset that fits your use case:

```python
ds_sa = ds.filter(lambda d: d["license_family"] in {"cc0-1.0", "cc-by-4.0", "cc-by-3.0", "cc-by-sa-4.0", "cc-by-sa-3.0", "cc-by-sa-2.0"})
```

## Considerations for using the data

* **Coverage is uneven.** Roughly 60% of the tokens date from 1750 to 1899; the 15th and 16th centuries are
  thinly covered. Newspapers (`Zeitung`) dominate the document count but make up only 8% of the tokens, and
  a single project (Alexander von Humboldt's writings) contributes 183 short documents.
* **Historical content.** The texts reflect the views, language and stereotypes of their time, including
  religious, political and colonial-era writing. They are provided for research on historical German and
  should not be taken as statements of fact or acceptable opinion.
* **Normalization is automatic.** `text_normalized` comes from the CAB tool chain and is not manually
  corrected. Expect residual errors, especially for very early texts, Latin passages, and proper names. The
  normalized text is also whitespace-tokenized (punctuation is separated by spaces), whereas `text` keeps the
  original layout.
* **`text` is not uniform.** 375 documents have their historical text reconstructed from the token layer and
  lack line breaks; check `text_source` if layout matters to you.
* **Metadata quirks.** Genre fields are null for a large share of the works in the DTA scheme (use the DWDS
  scheme for complete main-genre coverage), typeface strings contain a few typos, and `year` is parsed
  heuristically from the free-text `date_published`.

## Citation

Please cite the DTA when using this dataset:

```bibtex
@misc{dta2026,
  author       = {{Berlin-Brandenburgische Akademie der Wissenschaften}},
  title        = {Deutsches Textarchiv. Grundlage für ein Referenzkorpus der neuhochdeutschen Sprache},
  year         = {2026},
  address      = {Berlin},
  howpublished = {Herausgegeben von der Berlin-Brandenburgischen Akademie der Wissenschaften},
  url          = {https://www.deutschestextarchiv.de/}
}
```

## Acknowledgements

A big thank you to the whole team of the Deutsches Textarchiv at the Berlin-Brandenburg Academy of Sciences
and Humanities! Also many thanks to all the partner projects, libraries and archives that contributed texts to the DTA
extension corpora.

This dataset only converts an official dump into Parquet for convenient use with the Hugging Face ecosystem.

## AI disclosure

This dataset card was drafted with Claude Fable 5 (claude-fable-5) based on the statistics produced by
`dta_dataset_stats.py`, and reviewed by the dataset author. The conversion and statistics scripts carry their
own AI disclosure blocks, following the rules in the [ai-disclosure](https://github.com/stefan-it/ai-disclosure) repository.