Speech acts in (inter)action: Repeated questions*

GABRIELE KASPER

1. Two ways of studying social actions

Social actions performed through language are the focal object of re-search on speech act realization (speech act research,1 for short) andconversation analysis, among other approaches to discourse that I willnot consider here. Yet because of their dierent disciplinary origins andhistories, the two approaches are interested in dierent aspects of socialaction mediated through language and, specically, through talk. Speechact research is primarily concerned with the semantic structures (con-ventions of means) and linguistic resources (conventions of form)2 bywhich speech acts are implemented (pragmalinguistics, the linguistic in-terface of pragmatics) and the social conditions for carrying out speechacts appropriately and eectively (sociopragmatics, the sociological in-terface of pragmatics). Conversation analysis is fundamentally concerned with the sequentialorganization of action in talk-in-interaction. These dierences3 in ana-lytical focus make dierent demands on data. Speech act researchersdraw on a large range of data types, including natural, elicited, interac-tional, non-interactional, and various forms of self-report data. As themain interest is in the conventions of means and form by which the focalspeech act is implemented, non-interactional (e.g., written questionnaire)data specically eliciting that speech act are considered an appropriatesource. And even when the data are interactional, it is standard analyticalpractice to isolate the focal speech act from its interactional environ-ment, submit its linguistic design to scrutiny, and relate the identiedmeaning and form conventions to discourse-external context factors. Thisapproach has been particularly common in cross-cultural and interlan-guage pragmatics research on speech act realization. It has yielded awealth of knowledge about speech act strategies and their implementa-tion in dierent languages, the relationship of indirectness and politeness,resources for mitigating and upgrading illocutionary force, and about

Intercultural Pragmatics 1-1 (2004), 125133 1612-295X/04/00010125 6 Walter de Gruyter

126 Gabriele Kasper

speech acts and their realizations as indices of social stance. Conversationanalysis, on the other hand, strictly connes its data to naturally occur-ring talk-in-interaction, with the mandate to examine actions as they un-fold in the sequential organization of the activity or analyzed segment.Rather than isolating speech acts from their sequential habitat and ana-lyzing them according to pre-established categories, the analytical prac-tice is to adopt an emic, interlocutors perspective by paying close atten-tion to the meanings that the co-participants make relevant to each otherthrough the details of their interactional conduct in the moment-by-moment unfolding of the interaction. With few exceptions (e.g., Golato 2003), speech act research and con-versation analysis have gone their separate ways, and to good eect. Butfor some pragmatic objects, drawing on both approaches in an inte-grated fashion may have analytical benets that are not available througheither speech act research or conversation analysis when deployed in-dependently. One such object that will serve as an illustrative case is therepetition of questions and requests functioning as question substitutes(Heritage & Roth 1995).

2. Repeated questions

Repetitions can accomplish a wide range of discourse-pragmatic actions(e.g., Johnstone 1994). Any interactional element is a potential candidatefor repetition, including illocutionary acts. The repeated actions consid-ered here are questions, formatted either as direct questions (e.g., wh-questions) or as question substitutes, which have the structure of requests(e.g., can you tell me, tell me). Extract (1) provides a straightfor-ward example, taken from an Oral Prociency Interview (see Appendixfor transcription conventions).

(1)06070809101112131415

I: !

C:I: !

C:

Mm. "Can you tell me about- what- you did over Golden Week?Pardon?>Tell me what you did< for Golden Week, >over Golden Week.assistant< of ca-, >cameraman

Speech acts in (inter)action: Repeated questions 127

In lines 9 and 10, the interviewer (I) repeats the question4 issued ini-tially. The repetition is prompted by an other-initiation of repair (8), byvirtue of which the candidate (C) displays non-comprehension of thepreceding directive. The subsequent version of the directive successfullyrepairs the initial problem, as evident from the relevant and extended an-swer turn that the candidate is now able to produce. Although the seg-ment appeared in a particular institutional setting, this sort of sequence isvery common in ordinary discourse and dierent types of institutionalsettings alike:

1. A asks a question (or issues another sort of directive).2. B has a problem in hearing or understanding the question. So instead of answering (as projected by the question), B (other-)initiates repair.3. A redoes (repairs) the question.4. B answers the question.

Such other-initiated and self-repaired sequences have been examinedextensively in the literature on repair (e.g., Scheglo 2000; Scheglo, Jef-ferson, & Sacks 1977). While this research provides a relevant backdropto the focus of this paper, my main concern here is the formatting of thequestion when it appears a second time aroundthe repeat. What is repeated when "Can you tell me about- what- you did overGolden Week? reappears as >Tell me what you did< for Golden Week,>over Golden Week.

128 Gabriele Kasper

a speech-act based analysis of requests (Blum-Kulka, House, & Kasper1989), the rst version ("Can you tell me about- what- you did overGolden Week?) falls into the category of conventionally indirect requestspecically, a query-preparatory or ability question format, whereasthe second (>Tell me what you did< for Golden Week, >over GoldenWeek.over Golden Week. Okay - excuse me- if you could tell me< wha-what did you do in Indonesia.(.) wha-huh?what did you do in Indonesia

Speech acts in (inter)action: Repeated questions

15 C:16 I:17 C:

129

!Indonesia?uh huh [what was your mission there what were you [doing [I had a meeting with (name)

The interviewers rst turn is composed of three directives. The rst, for-matted as a mitigated want statement (Id like you to think to back (.) toyer (.)trip to Indonesia), foreshadows the upcoming question and speciesthe topic (Scheglo 1980). The initial element of the next action, please,projects the illocutionary force of the upcoming action, while the tokencan simultaneously be heard as a politeness marker (House 1989) to mit-igate the locution-derivable directive ( please tell me about it). The pro-noun (it) anaphorically links to the topic yer trip to Indonesia, establishedin the preliminary directive.7 The third directive partially repeats the pre-ceding request. Here, the interviewer self-repairs the grammatical objectin please tell me about it from the pronominal it to an object clause (tellme what you did ), with emphasis on the verb, and in so doing pinpointsthe precise focus of her question. This repeated question directive pre-serves the directness level of the original version (locution-derivable) butomits the mitigating please and sharpens the question target, i.e., itslightly alters the propositional content. Shortly after the candidate beginshis narrative, the interviewer intervenes with a turn comprising four units:(1) a turn-initial okay, which can be heard both as a token receipting theprevious turn and as a discourse marker indexing the beginning of a newsequence; (2) an apologetic routine that is conventionally deployed to doan interruption (excuse me); (3) a question directive (if you could tell me),formatted as a conventionally indirect request with two syntactic miti-gators (a hypothetical if clause, could ) and (4) a question about the can-didates activities in Indonesia, designed as syntactically independent ofthe preceding question directive (what did you do in Indonesia not whatyou did in Indonesia). The rst three units, produced with acceleratedspeed, supply contrastive emphasis to the fourth unit, which carries thefocal point of the repeated question and is articulated with normal speed.Notably, the question directive displays a marked increase in politeness,indexing the interviewers interruption as a dispreferred action (Lerner1996). Prompted by the candidates other-initiation of repair (13), theinterviewer does an exact repetition (Cushing 1994), i.e., she repeatsthe question verbatim, including the emphasis on the innite verb (14)8.However the candidate responds to the repair completion with yet an-other repair initiation, this time requesting conrmation that he heardcorrectly the location specied by the interviewer (15). After conrmingthe candidates hearing, the interviewer produces two more versions ofthe question. These repeat questions (what was your mission there what

130 Gabriele Kasper

were you doing) preserve the illocution and pragmalinguistic format of thepreceding question (what did you do in Indonesia) but provide alternatereference expressions to the candidates activities in Indonesia. Once theinterviewer issued the question in a direct (wh-) question structure, thisformat remained stable throughout the repeat series. We have seen that the questions and question substitutes in the twoextracts exhibited dierent levels of directness (in decreasing directness):

direct questions (wh-)what did you do in Indonesiawhat was your mission therewhat were you doing

locution-derivable/directTell me what you did for Golden Week, over Golden Week.please tell me about ittell me what you did

conventionally indirect"Can you tell me about what you did over Golden Week?if you could tell me

Furthermore, two of the requests were mitigated with lexical and syntac-tic material ( please tell me about it; if you could tell me), whereas theothers were not. For speech act research, accounting for the observed dierences in thedesign of the repeated question directives is problematic. The standardexplanatory resource in a speech act realization framework is the con-stellation of social variables pre-existing the interaction (relative power,social distance, degree of imposition; Brown & Levinson, 1987). Morefundamental problems with unidirectional causal explanations aside, thediculty with this explanation is that it does not lend itself well to ex-plaining why in an ongoing interaction between co-participants in an os-tensibly stable social relationship, repeated actionsactions whose im-position ranking is (supposedly) constantexhibit dierences in theirpragmalinguistic formatting that are associated with dierent socio-pragmatic meanings. Whereas speech act research can locate these ana-lytical diculties but cannot solve them with its own methodological re-sources, conversation analysis provides the analytical strategies necessaryto account for the formats of repeated directives. The solution oered byconversation analysis is its trademark analytical practice, the close exam-ination of actions in the local, sequential context in which they are pro-duced. Drawing on speech act research and conversation analysis in acombined analytical framework makes it possible to connect the prag-

Speech acts in (inter)action: Repeated questions 131

malinguistic resources identied in earlier work on speech act realizationto the sequential contexts in which they are deployed. Repeated actionsare a particularly suitable object for such analysis.

Appendix

Transcription conventions(0.8)Time gap in tenths of a second(.)Brief time gapLatching of utterance segments[ ]Overlapping talk-Cut-o:Elongated sound.Falling intonation,Continuing intonation?Rising intonation"Marked rise of immediately following segmentUnder EmphasisDecreased volume>

Speech acts in (inter)action: Repeated questions 133

Drew, Paul. 1997. Open class repair initiators in response to sequential sources of troubles in conversation. Journal of Pragmatics 28: 69101.Gass, Susan M. 2003. Input and interaction. In Catherine J. Doughty and Michael H. Long (eds.), Handbook of Second Language Acquisition. Oxford: Blackwell. 224255.Golato, Andrea. 2003. Studying compliment responses: A comparison of DCTs and record- ings of naturally occurring talk. Applied Linguistics 24: 90121.Heritage, John C. and Andrew L. Roth. 1995. Grammar and institution: Questions and questioning in the broadcast news interview. Research on Language and Social Interaction 28: 160.House, Juliane. 1989. Politeness in English and German: The functions of please and bitte. In Shoshana Blum-Kulka, Juliane House, and Gabriele Kasper (eds.), Cross-cultural Pragmatics: Requests and Apologies. Norwood, NJ: Ablex. 96119.Johnstone, Barbara (ed.) 1994. Repetition in Discourse (Vols. 1 & 2). Norwood, NJ: Ablex.Lerner, Gene. 1996. Finding face in the preference structure of talk-in-interaction. Social Psychology Quarterly 59: 303321.Ross, Steven. 1995. Aspects of communicative accommodation in oral prociency interview discourse. Unpublished Ph.D. dissertation, University of Hawaii at Manoa.Scheglo, Emanuel A. 1980. Preliminaries to preliminaries: Can I ask you a question? So- ciological Inquiry 50: 104152. . 1984. On some questions and ambiguities in conversation. In J. Maxwell Atkinson and John Heritage (eds.), Structures of Social Action. Cambridge: Cambridge University Press. 2852. . 1988. Presequences and indirectness. Journal of Pragmatics 12: 5562. . 2000. When others initiate repair. Applied Linguistics 21: 205243.Scheglo, Emanuel A., Gail Jeerson, and Harvey Sacks, 1977. The preference for self- correction in the organization of repair in conversation. Language 53: 361382.