This study explored the development of new texts by fluent non-native speakers of Chinuk Wawa, an endangered indigenous contact language of the Pacific Northwest United States. The texts were developed as part of the language and culture program of the Confederated Tribes of the Grand Ronde Community of Oregon for use in university-sponsored language classes. The collaborative process of developing 12 texts was explored through detailed revision analysis and interviews with the materials developers and other stakeholders. Fluent non-native speakers relied on collaboration, historical documentation, reference materials, grammatical models, and their own intuitions and cultural sensibilities to develop texts that would be both faithful to the speech of previous generations and effective for instruction. The texts studied were stories and cultural information developed through research-based composition, translation from interlinear and narrative English in ethnographic sources, and editing of transcribed oral narrative. The revision analysis identified points of discussion in the lexical development and grammatical standardization of the language. The preferred strategy for developing new vocabulary was use of language-internal resources such as compounding although borrowing and loan translation from other local Native languages were also sometimes considered appropriate. The multifunctionality of the lexicon and evidence of dialectal and idiolectal usage problematicized the description of an “ideal” language for pedagogical purposes. Concerns were also expressed about detailed grammatical modeling due to potential influence on non-native speaker intuitions and the non-utility of such models for revitalization goals. Decisions made in the process of developing texts contributed to the development of a written form of Chinuk Wawa that would honor and perpetuate the oral language while adapting it for the requirements of inscription. The repeated inclusion of discourse markers and the frequent removal of nominal reference brought final versions of texts closer to oral style, while inclusion of background information and the avoidance of shortened pronouns and auxiliaries customized the presentation for a reading audience. The results of this study comprise a sketch of one aspect of the daily work of language revitalization, in which non-native speakers shoulder responsibility for the growth of a language and its transfer to new generations of speakers.