| Chapter 1: | Introduction |
|
|
notion formally. For some researchers, cohesion has been taken to be the defining feature of coherence whereby links or ties between parts of the text act to ‘bind’ it into a single whole and create ‘cohesive harmony’ (Hasan, 1984, p. 218). Others have argued that this view of coherence fails to take into account the dynamic nature of discourse and the interactive process with the interlocutor's mind and background schemata (e.g., McCagg, 1990).
The need to define and operationalise coherence formally is an important task, however, since the term is often employed in applied linguistic contexts to characterise texts and assess language proficiency. The IELTSTM speaking test 2, for example, has coherence as one of its descriptors of performance and describes it as “the ability…to link ideas and language together to form coherent, connected speech” (IELTS, 2002, p. 14). Research to date, though, has mostly been concerned with native texts, often produced by skilled writers or speakers, where the linkage of ideas and construction of connected speech is relatively good. Studies on the lack of coherence and disturbances of the type exemplified in example 1.1 above, which are more common in non-native discourse, have been sparse, and there has been little evidence of the systematic cataloguing of the miscues that lead to incoherence.
Research questions
The first aim of this study is to produce an operationalised definition of coherence, which will enable the identification and cataloguing of discourse miscues to be undertaken in a corpus of non-native extended spoken discourse. The second aim is to determine if the manipulation of these miscues can lead to an improvement in coherence as perceived by native listeners. In order to achieve these goals, the research will be divided into two parts: a textual analysis and an evaluative study. The textual analysis will attempt to answer the following broad research question: