Paper short abstract:

The National Institute for Japanese Language and Linguistics (NINJAL) has been conducting projects to preserve endangered languages in Japan. We explore their histories and examine two ongoing projects that adopt global standard formats in digital humanities, such as TEI, IIIF, and Dublin Core.

Paper long abstract:

This presentation focuses on the preservation of endangered languages in Japan. Japan is home to several indigenous languages, including Ainu (Ainuic language family), Amami, Kunigami, Okinawan, Miyakoan, Yaeyaman, Yonaguni, Hachijo, and various Japanese dialects (Japonic). The National Institute for Japanese Language and Linguistics (NINJAL) manages the documentation and preservation of these languages.

There are five projects currently ongoing: NINJAL Digital Archive (NINDA), Open Multilingual Online Lexicon of Okinawan (OMOLO), Endangered Languages in Japan Database, Kotoba no Museum Portal, and Corpus of Japanese Dialects (COJADs). This presentation starts by exploring the history of these projects before examining the two projects that adopt global standard formats in digital humanities (DH).

NINDA, the digital archive of language resources from NINJAL, features images of rare documents, audio, and videos related to endangered languages and dialects in Japan. All the metadata is in Dublin Core, the de facto standard of metadata description in DH. Furthermore, all the images accompanying the documents, audio, and videos are provided through the International Image Interoperability Framework (IIIF) and displayed in IIIF viewers, such as the Universal Viewer and Mirador. Users can easily interoperate and reuse the items provided through the IIIF. Users can also easily zoom in and out of images and annotate them with a low burden on the local computer’s memory. This archive is built upon Omeka S, a content management system developed for digital archives that readily enables Dublin Core metadata descriptions and IIIF installation.

OMOLO is an online dictionary of Okinawan that utilizes TEI Lex-0, a dictionary-oriented subset of TEI, the de facto standard of text markup in DH. This is based on NINJAL’s publication of the dictionary in paper format, which we have now digitized. We annotated all the TEI Lex-0 meanings with English, Portuguese, and Spanish translations for the Okinawan diaspora community. We publish the TEI data publicly on an open license; thus, users can freely utilize it. We are also visualizing the data on a user-friendly website.

Thus, this presentation proposes two specific models to digitize resources related to the preservation of endangered languages in Japan utilizing the standard formats in DH.