About
The Early Records Database of the Ainu Language (ERDAL; アイヌ語古記録データベース; Aynuitak Huskokampi Aomarepu) publishes early written records of the Ainu language in a form that people and machines can both read. Wordlists become structured data, each item a headword, the historical spelling and a modern Ainu form with its source, ready for dictionaries and corpora. Prose and travel accounts are edited against the holding institutions' facsimiles into a readable text. The transcriptions are the work of the contributors to the みんなで翻刻 project アイヌ関連資料 and, for the printed Meiji records, of the proofreaders on Japanese Wikisource; they can be read beside the images and downloaded as TEI, CSV and JSON.
How it began
The Ezo hōgen moshiogusa and the other early records that matter for the study of Ainu had been transcribed before, but never published in a form that is easy to get at. In April 2026 the project アイヌ関連資料 was opened on みんなで翻刻 with a call for help, and in under half a year its volunteers had transcribed most of the pages. This site keeps the promise made to them then: to publish their work as structured open data that serves research on the language.
What you can do here
Wordlists are split line by line into a headword and the historical spelling, with a modern Ainu form added, each carrying a certainty and a citation. The search runs both ways, so a record of 1792 can be read against the dictionaries of today, and the data goes straight into dictionaries and corpora. Prose and travel accounts can be read beside the facsimile, and a readable edited text, checked against the images, is being prepared for them. The record itself is kept as written; the edited text and the modern forms are separate editorial layers. The texts are available as TEI, and the wordlist data also as CSV and JSON.
Transcription status and sync
Page status (not started, in progress, transcribed) follows the record on みんなで翻刻, or on Japanese Wikisource for the printed Meiji records proofread there, where a page counts as transcribed once it is marked proofread or validated. Anyone can edit a transcription on either platform, and corrections reach this site at the next sync. Image-based partial transcriptions are marked in progress, with their extent and attribution shown on each page. Facsimile checks and interpretations are recorded separately.
Name and icon
The Ainu name Aynuitak Huskokampi Aomarepu strings together aynu itak (the Ainu language), husko kampi (old paper) and a=omare pu (a storehouse things are put into): a storehouse for the old papers of the language. Its initials spell AHA, and aha is also the Ainu name of the hog peanut, Amphicarpaea edgeworthii (Japanese ヤブマメ), whose underground beans were dug up and eaten (田村 1996, 千葉 2002 and others). The site icon is that bean.
Licence
The transcriptions are published under CC BY-SA 4.0, the licence that applies to transcription data on みんなで翻刻 and on Wikisource; when you use them, credit the contributors of the みんなで翻刻 project アイヌ関連資料, and the contributors of Japanese Wikisource for the records transcribed there. Image-based partial transcriptions are attributed to ERDAL under the same licence. The wordlist data and other files this site derives from the transcriptions carry the same licence. Facsimile images belong to their holding institutions and are shown from their IIIF servers under each institution's own terms, linked from each witness. Research tables carry separate source attribution and rights metadata.
Related sites
This site is one of the aynu.org family. The catalogue record for each source (author, date, holdings, earlier scholarship) is on db.aynu.org, and the two sites link to each other by a shared source identifier. A sister edition built the same way, bible.aynu.org, sets Batchelor's 1897 Ainu New Testament beside a modern-orthography text. The modern forms interpreted here are taken into the corpus search at corpus.aynu.org and the dictionaries on aynu.org, so that a historical spelling can be followed through to present-day usage. The transcription itself continues on the みんなで翻刻 project アイヌ関連資料 and, for the printed Meiji records, on Japanese Wikisource, and the images come from the holding institutions' servers, some of them through the Kokusho Database.