Multi-Linear Transcription and Analysis of Oral Discourse

Abstract

This chapter discusses the components in a multi-linear transcription and explains their significance for discourse analysis. The essential procedures for orthographic transcriptions of disordered discourse production are first presented. In particular, specific considerations needed in preparing a transcript (such as the layout and levels of annotations) and an orthographic transcription in English versus in any other non-English languages are discussed. The values of including information on segmental and supra-segmental information, discourse features, and gestural characteristics to be encoded in parallel with the orthographic information are then explored. This is followed by the introduction of several web-accessible archiving software or computational tools for transcription of spoken discourse, including Child Language Data Exchange System (CHILDES), Systematic Analysis of Language Transcripts (SALT), EUDICO Linguistic Annotator (ELAN), Praat, and Python and the Natural Language Toolkit (NLTK). Finally, the clinical applications and research values of multi-linear transcriptions are explained and justified. Readers’ understanding will be enhanced by the carefully selected examples of multi-linear transcripts using typical or conventional transcription symbols for unimpaired as well as disordered language samples.