Discontinuity in conversational speech : an investigation of some theoretical problems and their analysis
The occurrence of discontinuity in conversational speech raises a variety of theoretical problems for the study of verbal communication. According to the dominant explanatory models of language, discontinuities in the form of pauses, self-corrections, repeats, false starts, and the like are errors'of language performance. Consequently, when, in - conversational speech, discontinuities do occur, it is natural that questions should arise regarding (a) why they occur and (b) given that they do occur, how speakers and hearers are able to deal with them. It is argued that no coherent answers may be offered to these questions under the prevalent conceptualization of discontinuity. As a feature of verbal performance, discontinuity has been studied primarily by researchers from the related disciplines of psychology, sociology, ethnomethodology, and discourse analysis. Although the explanatory approaches differ greatly, there is an underlying presuppositional unity to them. Whether discontinuity is studied as an indication of the speaker's cognitive processing, as a feature of social differentiation, or as an essential aspect of the speaker and hearer's management of conversational interaction, it is always assumed that discontinuities occur as the result of speaker error. This assumption, in turn, is shown to be derived from an unrecognised written-language (or 'scriptist') bias in the study of verbal communication. This conceptual bias, it is argued, is a shared weakness in the otherwise differing theoretical approaches to the explanation of discontinuity in conversational speech. The theoretical discussion of the thesis is supplemented by an appendix containing a selection of transcribed examples of discontinuities compiled with the aid of a computer concordance program. These examples are used as comparative evidence throughout the central chapters of the thesis.