# Kurdîzan vocabulary — sources and reuse

This list contains 2,000 distinct Kurmanji headwords with selected English and Turkish meanings, arranged into four study stages of 500 entries. These are editorial learning stages, not certified CEFR levels or measurements of Kurdish word frequency. Synonyms are separate words; spelling variants and obvious inflected duplicates are filtered. Meanings are not exhaustive and may vary by context and region.

## Dictionaries

Adapted from the Kurdish-English and Kurdish-Turkish Ferheng / FreeDict dictionaries, edition 1.2, by Erdal Ronahi and the contributors credited in their TEI headers. Copyright © 1999–2017 their respective contributors.

- https://github.com/freedict/fd-dictionaries/tree/master/kur-eng
- https://github.com/freedict/fd-dictionaries/tree/master/kur-tur
- https://freedict.org/

The dictionaries and the modified dictionary data in `words.json` are distributed under GNU GPL version 2 or later. See [the full license](./GPL-2.0.txt). The JSON is the editable source form of the published dictionary list. There is no warranty. Kurdîzan's changes (2026-10-11) include headword selection, normalization, bilingual gloss corrections, additional everyday words, and assignment to learning stages. This data notice does not relicense unrelated application code.

## Ordering

The first 500 words are an editorial foundation list. Later stages use English gloss familiarity and headword length as rough ordering aids, with editorial adjustments for everyday and specialist terms. This proxy does not establish actual Kurdish frequency or guarantee the difficulty of every individual word. Educator review is recommended before using the stages as a formal syllabus.

The ordering aid is Hermit Dave's FrequencyWords English 2018 list, derived from OpenSubtitles. That separate source file is licensed CC BY-SA 4.0, unchanged in `data/vocabulary/en-frequency.txt` in the project:

- https://github.com/hermitdave/FrequencyWords
- https://github.com/hermitdave/FrequencyWords/blob/master/content/2018/en/en_50k.txt
- https://creativecommons.org/licenses/by-sa/4.0/

The application does not display or distribute English occurrence counts in the dictionary. Full upstream dictionary snapshots, editorial inputs and the deterministic build script are kept in the project's `data/vocabulary/` and `scripts/build-vocabulary.py` for maintenance.

## Grammatical gender (2026-10-12)

Additional noun-gender facts were checked against the Kurdish and English Wiktionary entries, using the [Kaikki Kurmancî extract](https://kaikki.org/kuwiktionary/Kurmanc%C3%AE/index.html) and [Northern Kurdish extract](https://kaikki.org/dictionary/Northern%20Kurdish/index.html). The checked-in `data/vocabulary/gender-sources.json` records the individual source pages; `gender.tsv` is the editable annotation table. Wiktionary contributors are credited on those pages and their histories; Wiktionary content is available under [CC BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/). Only grammatical facts and attribution links were retained, not dictionary definitions, examples or translations.

`m` means masculine, `f` feminine, and `mf` either gender according to referent, meaning or regional usage. Labels describe noun usages; a verb gloss beginning with “to” does not inherit the gender of a nominalized homograph. Non-nouns and nouns without verified gender remain unlabeled; absence of a label never means neuter. Existing editorial annotations are retained. Spelling variants and ambiguous senses require review rather than guessing from a word ending.

Six additional noun-gender facts were verified in Michael L. Chyet’s *Kurdish-English Dictionary* (printed pages 19, 75, 299, 555 and 761) and Baran Rizgar’s *Learn Kurdish* (printed page 228). Individual PDF page links are recorded in `gender-sources.json`. Only gender facts were transcribed, not definitions or examples.

## Example sentences (2026-10-16)

Each word has three example sentences in `examples-1.json` … `examples-4.json`, with English and Turkish translations. They were written for Kurdîzan and are not copied from the dictionaries above, Wiktionary or any other source. They follow the selected gloss of each word and have not yet been reviewed by a native-speaker specialist; see `data/vocabulary/examples/README.md` for entries to check first.

## Word classes and transitivity (2026-10-16)

All 2,000 entries now carry word-class annotations for their selected meanings. These are editorial annotations stored in `data/vocabulary/grammar.tsv`, not part-of-speech fields supplied by the FreeDict snapshots. Existing noun-gender annotations inform noun classifications; words with multiple uses can carry multiple classes. Verb transitivity describes Kurdish usage and is not inferred automatically from English or Turkish translations. Uncertain transitivity is displayed as “not verified.” The annotations have not received independent native-speaker review, and regional or additional senses may differ.