Four Models for Saving Endangered Languages: How Academia, Technology, Art, and Communities Divide the Work

Endangered language preservation is not one discipline but several, and the organizations doing the work rarely resemble each other. A university initiative at the CUNY Graduate Center promotes the scientific study of endangered languages in research and teaching. A language services company founded in 2021 pushes underrepresented languages into modern technological ecosystems while creating fair employment for Indigenous linguists. A Stanford collaboration with Pratt Institute's Communication Design department applies design methods to documentation and revitalization. And in China, poets and singer-lyricists perform in She, Hani, and Mulao to keep cultural identity alive through art. Understanding how these models differ—and where they overlap—is the clearest way to grasp the modern landscape of language preservation.

Introduction: Diverse Approaches to Language Preservation

The initiatives documented here share a sense of urgency but diverge sharply in their methods. The Design for Endangered Languages project at Stanford's digital humanities lab frames the stakes plainly: it responds to what it calls "the urgent call to document and describe endangered languages before they disappear," based on a survey of over two hundred Indigenous communities around the world. That urgency is the common thread. What differs is the instrument each model chooses—scholarship, employment and digital infrastructure, visual and communication design, or performance and song.

These are not interchangeable strategies. Academic research produces documentation and trained specialists. Technological integration determines whether a language can function in the digital environments where modern life happens. Artistic work reaches audiences that scholarly archives never will. Recognizing which model a given initiative belongs to tells the reader what that initiative can realistically accomplish, and what it cannot.

Academic Foundations: Research and Education

The academic model is exemplified by the Endangered Language Initiative at the CUNY Graduate Center, which aims to promote the scientific study of endangered languages in the areas of research and teaching. Its contribution is foundational rather than direct: it builds the scholarly capacity—courses, researchers, and institutional structures—on which other approaches depend. The initiative sits within CUNY's Linguistics Program and connects to related efforts such as the Endangered Language Alliance, positioning endangered language study within a broader academic ecosystem.

The strength of this model is rigor and continuity; its limitation is reach. Academic programs document and describe languages, but they do not by themselves put those languages into daily use, into software, or into popular culture. That gap is precisely where the other models begin.

Technological Integration: Digital Tools and Communication Design

The technological model takes two distinct forms in the evidence. The first is commercial and community-facing: Indigenius, founded in 2021, operates as a language service provider that promotes fair employment opportunities for Indigenous linguists and encourages the integration of underrepresented languages into modern technological ecosystems. Its founder's background in the non-profit sector informs an approach that fosters close collaboration between communities, organizations, and technology partners, with the stated aim of ensuring Indigenous knowledge is represented with accuracy, respect, and cultural integrity.

Indigenius also advocates ethical and sustainable localization practices—work that honors each language's identity while expanding its digital presence. The underlying logic is economic as much as cultural: if Indigenous linguists can earn fair wages localizing technology into their own languages, preservation becomes a livelihood rather than a volunteer burden.

The second form is design-driven. The Design for Endangered Languages project, a collaboration between Stanford and the Communication Design department at Pratt Institute, combines lexicographic and pedagogical methods from the humanities with communication design. The project parallels Dr. Sarah Ogilvie's Endangered Language Digital Humanities Project, and its stated goals include sharing methods, models, and software; raising the profile of existing documentation work; and critiquing innovative documentation methods. It also aims to determine the distribution and gauge the "health" of endangered languages worldwide—a diagnostic ambition that treats design as an instrument of measurement, not just presentation.

Artistic Expression: Poetry, Song, and Cultural Identity

The artistic model documented in the Versopolis Poetry Expo 25 project works from the opposite direction: instead of building infrastructure, it builds attachment. The project brings visibility to Indigenous-minority languages in China through bilingual and biliterate poets, arranged geographically across the country's regions. Their works, written and performed in both native languages and Mandarin Chinese, serve as what the project describes as powerful acts of cultural preservation.

Some of these figures reach beyond literary circles. Lan Yongxiao, a She (Shan Ha) artist from Zhejiang who turned from migrant work into a popular singing career, treats song as the vessel for ancestral stories. "If young people don't speak She, the cultural heritage of She can never be preserved," he emphasizes—framing language loss as an intergenerational transmission problem that music can address. Gebu, a Hani poet from Yunnan, performs live readings displaying poems in Hani using a newly Romanized writing system, underscoring the urgency of preservation. Tong Yu of Guangxi recites in Mulao, a language spoken by a small population in southern China, highlighting its ties to the larger Zhuang and Dong (Kam) families. Notably, the project deliberately showcases "small" languages often overshadowed by dominant minority languages such as Tibetan and Uyghur.

Comparative Framework of Language Preservation Models

The table below categorizes the documented initiatives by their core methodology, primary method, and key goal, allowing the four models to be compared directly.

Comparison of language preservation models by approach, method, and goal
Initiative / Model	Primary Method	Key Goal
Academic — CUNY Endangered Language Initiative	Scientific study of endangered languages through research and teaching	Advance scholarly knowledge and institutional capacity for preservation
Technological — Indigenius (founded 2021)	Language services, fair employment for Indigenous linguists, ethical localization	Integrate underrepresented languages into modern technological ecosystems while representing Indigenous knowledge accurately
Technological / Design — Stanford–Pratt Design for Endangered Languages	Combining lexicographic and pedagogical humanities methods with communication design	Maintain, preserve, and revitalize endangered cultures; gauge the distribution and "health" of endangered languages worldwide
Artistic — Chinese bilingual poetry project (Poetry Expo 25)	Poems, songs, and live reading performances in native languages and Mandarin	Foster cultural identity, preserve ancestral stories, and raise awareness of "small" languages

Read side by side, the models reveal a division of labor. The academic and design projects document and diagnose; Indigenius builds economic and digital infrastructure; the poets create the emotional and cultural pull that makes a language worth speaking to a younger generation. No single row can substitute for another.

Navigating the Future of Endangered Language Preservation

To effectively support endangered language preservation, consider initiatives that integrate academic rigor, technological innovation, artistic engagement, and strong community leadership. The evidence here suggests these are complementary rather than competing strategies: documentation without transmission fails, and enthusiasm without infrastructure fades. The measurable indicator to watch is increased visibility and accessibility of endangered languages through diverse platforms and community-led programs—whether that means more Indigenous linguists employed in localization work, more design-informed documentation tools shared across institutions, or more young people hearing She, Hani, or Mulao performed in public. Each new appearance of an endangered language on a stage, a screen, or a payroll is a small, countable act of preservation.