Beginning to Respond: Well-Prefaced Responses to Wh-Questions

Abstract

This article reports on the occurrence of well within an analytically delimited sequential environment: turn-initial position in the second pair-part position of adjacency pair sequences launched by a wh-question. We show that these well-prefaces operate as general alerts that indicate nonstraightforwardness in responding, and we compare this form of alert to others that operate in talk-in-interaction. We conclude by addressing the relationship of answering to responding, and by considering the relationship of well-prefacing to preference organization.

Notes

The work reported here had its beginning in the Conversation Analysis Advanced Study Institute held at UCLA in the summer of 2004 (CA ASI 2004), in which the authors were the instructors. The early phases of the work benefited substantially from the contributions of the participants in the Institute: Steven Bloch, Fabienne Chevalier, Helen Duah, Marja-Liisa Helasvuo, Shuya Kushida, Geraldine Leydon, Richard Ogden, Rebecca Shaw, Susan Speer, Elizabeth Stokoe, Cecilia Varcasia, and Hansun Zhang Waring; Christopher Koenig rendered invaluable technical support. A much earlier version of part of this text was presented at the 2004 Meeting of the National Communication Association, Chicago, IL, and a somewhat revised version at the 2008 Meeting of the American Sociological Association, Boston, MA.

¹Digitized audio or video clips of data discussed in this article may be accessed at: http://www.sscnet.ucla.edu/soc/faculty/schegloff/

²For those who may be unfamiliar with this aspect of American culture, in clubs that serve alcohol, it is a common practice to separate the underaged, who cannot legally be served alcohol, from those who can be served by checking their identification and directing them to separate parts of the club.

³An anonymous referee points out that, as this single TCU turn shows, nonstraightforwardness can be quite straightforward in form. This is an important observation because it shows that a well-prefaced response need not be an indicator of trouble in answering. Also, see Extract 12 for a case of straightforward responding where there is some trouble in answering.

⁴We examined some 150 other-initiations of repair that took or included the form of a wh-question; not a single one met with a well-prefaced response. However, of other-initiations that took the form “What do you mean,” “What do you mean by that,” or “What do you mean + [element of prior turn],” roughly half received well-prefaced responses. We cannot take this matter up here; the curious reader might have a look at Schegloff, 1997b, pp. 522–524, and note 3, p. 543, where the use of this format to do “challenging” is taken up.

⁵This data extract is taken up in Schegloff, 2005, pp. 451–452.

⁶This sequence is taken up in greater detail in Schegloff 1972, pp. 106–114. The usage here and in Extract 18 declares explicitly what Heritage describes as the “epistemic gradient” embodied in question formats (see Heritage, 2007).

⁷Moreover, we have left unquestioned the matter of whether it is a response. However, the following extract suggests that a recipient can be confronted with the question, Is the talk by a recipient that follows a turn-initial well in the next turn after a wh-question “doing responding” to it? In this extract Louise is a roughly 12-year-old girl, whose bedtime routine with her mother is to lie together in her bed and talk about the day's events, or anything else.

Super Seedy (Schegloff, 1997a)

At line 3 we have a next turn that is from its outset not a response to the prior turn; indeed, it is not a response to the prior turn, which is a wh-question, and this is something that its recipient will have to figure out in making sense of the well. The well with which it starts is only incidentally after a wh-question, and is not operating sequence organizationally; sequence organizationally Mom's wh-question at line 2 implements a “go-ahead” response to Louise's pre-telling at line 1, but Louise's turn at line 3 does not progress the sequence that her prior turn had projected and with which Mom's turn had aligned. Rather, Louise first makes a pass at doing a third turn repair (Schegloff, 1997a) on “today.”

As this extract shows, recipients can design next turns after questions not only as nonanswers but as nonresponses. The proper scope for analyzing and assessing what well-prefaced turns are doing with respect to the prior turns that they follow entails that we need first (because recipients need first) to establish that the target turn is designed to do responding to the prior. However, as all of our initial 10 specimens do have ensuing turns that do responding, as do virtually all of the other 40 specimens that we have examined, we will leave this undertaking for another occasion, and in this section we turn directly to the issue of whether the turns that are responsive to the turn that is the locus of a wh-question do answering or not.

⁸Arguably why has a type-conformity response token—namely because, but this will remain a conjecture until empirical analysis establishes the import of its presence or absence in a responsive turn (if there is, in fact, an import). Nevertheless, as with yes and no, an answerer can claim to be launching a type-conforming answer with a turn-initial because.

⁹See the discussion of Extract 7 on p. 102 of this article.