Methods and Results in Language Documentation using Literary Digital Editions

This project applies the international, scholarly standards of textual encoding developed by the Text Encoding Initiative (TEI) to the longest and most complete pre-1492 book to survive the conquest. Other digital projects represent the text in manuscript facsimiles, a format that is inaccessible to non-experts, cannot easily be used for language learning, and privileges the colonial-era act of writing rather than the original oral tradition. In contrast, this project’s five-item digital collection of historical and modern texts, videos, and translations, developed by Indigenous scholars in collaboration with non-Native faculty and student researchers, uses encoding tools to reveal colonial interventions and highlight Indigenous scholars’ corrections to orthography, morphosyntax, and vocabulary. These tools include referencing string analytical attributes to tag characters, places, and objects in the text, the element to preserve poetic orality and align concepts according to morphological parallelism and phrase-final markers, and annotations, managed through a relational database of 880 cultural topics, to present competing scholarly interpretations of textual passages, phonetics, and syntax. The collection is housed on a platform that uses an open-source Static Site Generator framework to avoid the long loading time of server-based websites, paired with a Progressive Web Application framework to provide mobile readers with an app-like experience that does not sacrifice phone storage. This digital humanities approach to historical textual analytics allows a variety of users, including researchers, teachers, students, and interested members of the public, to represent the corpus and code in ways that cannot be disseminated through print. In so doing it creates a new tool for language learning, textual analysis, and linguistic study by native speakers, heritage learners, and non-Native researchers and students.

Colonial archives present a range of methodological challenges for scholars. To address this well-known problem, scholars must develop methods that disentangle colonial interventions from Indigenous linguistic, historical, and cultural data. This project offers one such solution. In collaboration with Indigenous scholars from the community, this project uses open source, standards-compliant tools to build a digital collection of five versions of the longest and most complete pre-1492 book of the Americas. By showing where Indigenous and non-Native authors record vocabularies, syntactic structures, and morphologies in different ways and at different moments, this project generates a corpus and methodology that can be applied to other linguistic and historical contexts, including future educational efforts. Because there is little data on the effectiveness of digital materials on Indigenous language acquisition for heritage learners and non-heritage speakers, this project creates important materials for future research.

There are 28 Mayan languages, all of which exist in different states of endangerment. For this project, linguists Ajpub' Pablo García Ixmatá (Tz'utujil) and Miguel Óscar Chan Dzul (Yukatek) formed research teams to survey the unique language learning needs of their communities. Those surveys led the researchers to create different kinds of texts. The team working in Guatemala is creating an annotated, scholarly edition based on the one manuscript to survive the Spanish invasion. Their footnotes contain linguistic information and explanations that can be used by teachers in Guatemala's nationwide program of bilingual and intercultural education. In contrast, the Yukatek team is creating short, animated videos that kids can watch on their phones. Students, native speakers, and faculty at the Universidad de Oriente, a Maya-serving institution, collaborated on the script, recording, and illustration of each video. The UVA team builds the digital infrastructure to host all of the materials.

Languages are living ecosystems that connect us across time and space. They are critical repositories of knowledge, care, and ways of being. When Maya children stop speaking their languages, those connections and forms of knowledge can be lost. This project aims to promote youth language learning by designing dynamic, interactive materials that meet kids where they are -- online.

Our project is funded for another two years (https://www.nsf.gov/awardsearch/showAward?AWD_ID=2109679), but during the first year of grant activity we have completed the encoding of two editions of the Popol Wuj (Colop 1999 and Briceño Chel et al. 2012); built a platform to display a variety of texts; finished 90% of characters, 100% of backgrounds, and voice tracks for chapters 1-2 of audiovisual materials in Yukatek Maya; 5) completed the script and storyboard for chapter 3 of audiovisual materials in Yukatek Maya; 6) completed 80% of chapter 1 (with group edits), 50% of chapters 2-4 (with edits), and 50% of chapter 5 (no edits) of the K'iche' edition; 7) begun early work on chapters 3-4 of the kids' edition in K'iche'.

Our repositories of code are available here: https://github.com/Multepal. Our database of Mesoamerican cultural topics, which includes 476 unique topics, 207 bibliographic sources, 104 critical annotations, and a web of 1969 relationships between these entities, is available here (https://multepal.spanitalport.virginia.edu/) and linked within each encoded text. Symposia that we organized online during COVID are available here: https://www.youtube.com/watch?v=PMRjGUxAwuQ (2021) and https://www.youtube.com/watch?v=OsSrwo_tbsk (2022).

"it helped me connect with my culture in new ways" -- María Cristina Pech Can (Yukatek Maya), undergraduate student in linguistics at the Universidad de Oriente who collaborated on the script for videos 1-3

Linguistics, history, literature, Indigenous Studies, digital humanities