The last twenty years have witnessed an explosion of research on issues of language endangerment, with the emergence of documentary linguistics and the growth of language revitalization programs, resulting in changes in methodologies and in subfields within linguistics. The present article assesses this work in terms of its impact, focusing on documentation corpora, their contents, and how data are collected, archived, and used. The push to document the “last” fluent speakers has resulted in gaps in the current research, such as a general lack of documentation of variation, few studies of the kinds of change that take place during language shift and attrition, and few studies of the newer forms of language which emerge as the result of revitalization.