Epistemic markers in university advisory sessions: Towards a local grammar of epistemicity

Abstract

This study is concerned with epistemic markers (linguistic expressions indicating different states of knowledge) in spoken academic discourse. It is based on ADVICe, a small corpus (160,000 running words) of spoken individual (one-to-one and one-to-two) academic interactions between university students and their lecturers (i.e. advisory sessions). The corpus was compiled as part of my PhD project at the University of Auckland (New Zealand) and is now available to the general linguistic community. The corpus was analysed using standard corpus linguistic and sociolinguistic methodologies as well as detailed discourse analysis. However, unlike most corpus-based sociolinguistic studies this study pays particular attention to inter-speaker variation in the data. The overall aim of the study is to contribute to the description of the local grammar of epistemicity in university advisory sessions in the tradition of Stubbs' programme. It provides a general framework for analysing epistemic phenomena - epistemic framing of discourse (EFD). It also deals with complex epistemic features such as epistemic accumulation, harmony and tension and provides an overview of typical epistemic structures in academic speech. The sociolinguistic part of the study analyses selected epistemic markers (I think, you know, sort of, kind of and adverbial epistemic markers) in dyadic situations (on-to-one academic sessions). It offers an insight into the variation of the use of these markers with particular attention to external factors such as the speaker's role, gender, age and native-speaker status. The results show a number of social differences in the use of these markers, however, the patterns are more subtle than typically reported in the literature. In order to complement the quantitative corpus-based investigations, the study also offers a detailed analysis of three transcript samples representing strong preference for, average use of and strong dispreference for epistemic markers. These three micro-studies offer an insight into the larger context in which epistemic markers are used as well as into the interplay between the individual epistemic forms. The study concludes with a discussion of two general models of epistemicity: the subjective and the intersubjective model. Evidence is reviewed suggesting that the intersubjective model of epistemicity more successfully explains epistemicity in interactional data such as the university advisory sessions. This model (which stands in contrast to the traditional subjective model) accounts for the functions of epistemic markers with reference to speaker-addressee relationship and the shared epistemic space.