This paper investigates the interactional organization of racism through participant production and uptake of explicit racial membership categories across a corpus of 50+ hours of audio/videorecorded interaction in three U.S. states. The discourse analysis examines one participant method for addressing "hearably racist" talk: echoing extreme versions of the problematic utterance to provide opportunities for repair work on inferable associations between membership categories and category-bound activities. Orienting to implicit inferential material as the source of trouble licenses participant account-seeking; treating the racism as a repairable downgrades its status as an overt instance of racism.
KeywordsRacism, membership categorization, extreme case formulations, repair, conversation analysis,Communication is proffered as the solution to a variety of social problems. Parents are told to "talk to your kids about drugs" 1 (Craig, 2005). Youths are encouraged to "use your voice" and "speak up" 2 in bullying situations (e.g., Bhat, 2008). "Just say no" 3 is a common strategy urged for refusing drugs or unwanted sexual advances (Kitzinger & Frith, 1999). And advice about responding to racist comments not only suggests there is an obligation to reply 4 , but that replies should convey disapproval and question speakers' reasoning 5 . Yet these well-intended admonitions often fail to account for how people actually interact and the multiple (at times contradictory) functions that talk serves. For example, Kitzinger and Frith (1999) showed that refusals are complex conversational actions accomplished by many tactics besides saying "no";and van Dijk (1992) illustrated how talk about race makes racist identities salient. Studies such as these demonstrate that the societal ideal of directly addressing social problems can be at odds with how and why talk unfolds in particular ways.This paper takes a discourse analytic approach to interpretations of racism in ordinary conversation. Two difficulties in pinning down "racism"-for participants and analysts-include(1) explicitly racist stances are rarely espoused, indeed, potentially racist discourse is often delicately introduced; and (2) "calling out" or otherwise obviously disaligning may be dispreferred, sanctionable, or face-threatening (e.g., van Dijk, 1992; Stokoe, 2015 [this issue];Whitehead, 2009; Whitehead, 2015 [this issue]). This paper analyzes examples in which participants explicitly name a racial category, associate it with a negatively-assessed activity, and they or their interlocutors orient to that as problematic.Drawing on conversation analytic membership categorization analysis (Stokoe, 2012;Bushnell, 2014) and grounded practical theory (Craig & Tracy, 1995), the analysis describes a practice deployed in response to naturally-occurring possibly-racist talk in recordings of private 4 face-to-face conversations. The practice involves jokingly taking the hearably racist utterance seriously/literally and reformulating it back to the speaker in such a way ...