In multiethnic, multilingual, and hazard-prone regions, language is not only a medium of communication, but a critical determinant of inclusion and survival during emergencies. This study examines the intersection of language policy and emergency governance in the Garzê Tibetan Autonomous Prefecture in China, where linguistic diversity—primarily Tibetan, Mandarin Chinese, and the Sichuan dialect—poses challenges for effective disaster response. Drawing on a mixed-methods design comprising 13 focus group interviews (N = 125), ethnographic field notes, and a large-scale survey (N = 4137), the research identifies how generational, occupational, and ethnolinguistic factors shape linguistic accessibility and exacerbate risk during crises. The findings demonstrate the absence of an emergency language policy that accounts for the sociolinguistic realities on the ground, particularly the needs of elderly monolingual Tibetan speakers and the linguistic limitations of external rescue teams and volunteers. By bridging sociolinguistic theory with disaster management frameworks, this study calls for a paradigm shift toward linguistically inclusive emergency planning that recognizes language as a core element of risk communication. It proposes a multilayered policy response encompassing the institutionalization of language accessibility, human and technological capacity-building, cross-sector coordination, resource allocation, and cultural integration. Beyond immediate practice, the study underscores the need for further research on AI-mediated language tools and participatory policy design, ensuring that emergency language services not only overcome communication barriers, but also promote equity, trust, and resilience in multilingual, high-risk contexts.
