UC Santa Barbara

In this article, I address some of the issues for the analysis of categorial features of talk and texts raised by Stokoe's 'Moving forward with membership categorization analysis: Methods for systematic analysis'. I begin by discussing a number of points raised by Stokoe, relating to previous conversation analytic work that has addressed categorial matters; the implicit distinction in her article between 'natural' and 'contrived' data; and ambiguity with respect to the (possible) relevance of categories, in particular practices or utterances. I then discuss how my own previous work could be located in light of Stokoe's discussion of debates and divergences between conversation analysis (CA) and membership categorization analysis (MCA), and argue that being bound by the integrity of the data on which an analysis is based (Schegloff, 2005) should take precedence over attempting to characterize the analysis as exemplifying either a CA- or MCA-based approach. I conclude by calling for a commitment to doing analysis, and pointing to the value of the resources Stokoe offers in this regard. © The Author(s) 2012.