ABSTRACT
In this chapter I discuss different ways of transcribing spoken interaction. Transcription is valuable not simply as a means of representing speech, but also as an analytical tool. It is often the slow, repetitive process of transcription and retranscription that begins to reveal to analysts aspects of an interaction that may be worth further investigation. At the same time, transcription is not a neutral exercise in which features of an interaction are objectively identified. Transcriptions necessarily correspond to a researcher’s interests and what they see as the analytical potential of their data, as well as their wider beliefs and values. It is in this sense that transcription is said to constitute both a representation and the beginnings of an interpretation of data. As Elinor Ochs has argued, in a now classic account of ‘Transcription as theory’, it is ‘a selective process reflecting theoretical goals and definitions’ (1979: 44). This point is illustrated by the different forms of transcription considered below. I shall discuss, in turn, conventions for transcribing spoken interaction; the representation of languages, varieties and styles; ways of laying out a transcript; and multimodal transcription.
