On Recruiting Conversation Analysis for Critical Realist Purposes

Abstract

In this paper I provide a summary and evaluation of some of the key features of Sims-Schouten, Riley and Willig's multi-level, `critical realist' approach to discourse analysis, as exemplified in their study of motherhood, childcare and female employment. I argue that (i) their analyses recruit and depend on arguments and techniques from the very perspectives they criticize, and (ii) those techniques are deployed in a somewhat ad hoc fashion. Consequently, I suggest that the authors fail to provide a distinctive or systematic operationalization of a critical realist discourse analysis. I end by arguing that if critical realists really want to understand what (purportedly extra-discursive) factors account for why participants say what they do, then they need to begin by adopting a more reflexive approach to their data, and pay serious attention to analysing the interview as an interview, and as an occasion for interaction in its own right.