Abstract

This commentary addresses the complexities of representation in sociolinguistic transcripts, considering the meaning potentials of different representational choices at the level of both ideology and identity. It considers the kinds of authenticity and evidence that are indexed by simplified versus more detailed transcripts, suggesting that a simplified transcript may give more direct access to elements of the original speaker's voice. Second, it discusses the role of transcripts as scholarly texts, and questions their use as sociolinguistic records suitable for reinterpretation by scholars who do not have access to the original recordings. Third, it emphasizes that the use of non-standard orthographies is an unreliable source of data on sociolinguistic variation, because few authors use non-standard forms in consistent and principled ways. Fourth, it proposes that translations may take different forms depending on whether they are efforts to be `faithful' to `authentic' speech in the target versus the source language. Finally, it reflects on the act of transcription as part of professional practice and identity among linguists.