Or-prefaced third turn self-repairs in student questions

Highlights

• This study uses conversation analysis to examine student questions.

• Or-prefaced third turn self-repairs do not forward substantively equivalent options.

• Or-prefaced third turn self-repairs revise initial questions for correctness/relevance.

• Participants orient to preference structure and convey distinct interactional goals.

• Participants concurrently manage talk and bodily conduct to achieve interactional goals.

Abstract

Analyzing video-recordings of university-level writing conferences, this conversation analytic study identifies interactional practices through which participants orient to the preference structure of talk-in-interaction. In particular, the study examines students’ or-prefaced third turn self-repairs produced during the earliest moments of the teacher's possible dispreference projection (e.g., short pause, hesitation, and possibly, gaze shift). In addition to displaying their orientation to getting a preferred response via the or-prefaced self-repair, students show their orientation to the correctness and relevance of their initial question formulation. The analyses show the ways in which students as well as teachers establish their divergent institutional goals and approach the task at hand differently. This study has implications for conversation analytic work on turn design, repair, and action projection while broadening the scope of existing research on discourse markers and educational discourse.

Introduction

This study uses conversation analysis (CA) to show the ways in which students as well as teachers establish their divergent institutional goals and orient differently to the task at hand during writing conferences. In particular, the study examines students’ or-prefaced third turn self-repairs produced upon the possible projection of the teacher's dispreferred answer. By deploying the conjunction or, students may appear to be providing an equivalent alternative to their previously conveyed candidate understanding in the initial question. However, upon closer examination, the data show that students instead problematize the correctness or relevance of their question via the third turn self-repair. In particular, the current paper focuses on the cases in which students produce the self-repair at the earliest moments of the teacher's possible dispreference indication. Excerpt 1 illustrates the student's or-prefaced third turn self-repair (line 3):

It is important to note that the third turn is produced very early with reference to the possible projection of a dispreferred response – only after a 0.2 s pause (line 2). In fact, the student's or-prefaced self-repair turns out to be in an overlap with the teacher's preferred response conveyed by her nodding (Goodwin, 1980; among others). Although prefaced by the conjunction or, the student's third turn does not convey an equivalent alternative. Instead, it conveys her orientation to the possible non-correctness of the initial question formulation in line 1. The following schematic is provided to illustrate the target phenomenon examined in this paper:

Before analyzing examples of or-prefaced third turn self-repairs, I first discuss preference organization in social interaction and the structural importance of third turn repair, focusing on practices of dealing with the projection of a dispreferred response. I then briefly review the interactional functions of or and its use in framing alternative questions. Lastly, I consider the implications of the current study for educational interactions. The literature review is followed by the introduction of the current data and its analysis.

By examining the detailed practices of third turn self-repair, this study contributes to our knowledge of the preference structure of a sequence and how participants manage dispreference in interaction. Many action types project two alternative responses: one displays alignment with the FPP (First Pair Part) (e.g., agreement, granting, acceptance to assessments, requests, invitations respectively) and the other displays disalignment with the FPP (e.g., disagreement, refusal, rejection to assessments, requests, invitations respectively) (Pomerantz, 1984, Sacks, 1987, Schegloff, 2007, pp. 58–73). These different types of responses in adjacency pair sequences are not “symmetrical alternatives” (Schegloff & Sacks, 1973, p. 314); the SPP that aligns with and furthers the action launched by the FPP is structurally preferred (Schegloff and Sacks, 1973, Stivers and Robinson, 2006).

The preference for promoting alignment in social interaction (cf. Goffman, 1967, Goffman, 1983) manifests in the way the SPP is produced in a contiguous or a non-contiguous manner subsequent to the FPP (Sacks, 1987). That is, in a question–answer sequence, which is a prototypical example of an adjacency pair sequence, there exists “a strong preference for contiguity between question and answer, and for agreement between question and answer” (Sacks, 1987, p. 58). These two types of preferences interact with each other; aligning answers that advance the initiated course of action are produced contiguously to questions, while disaligning answers are produced with delay (pause, mitigation, elaboration, etc.) (Pomerantz, 1984, Sacks, 1987, Schegloff, 2007, pp. 58–73). Thus, one of the central features of producing a dispreferred SPP is its positioning; a preceding inter-turn gap (silence) and various delay markers (uh, well, etc.) often project a dispreferred SPP (Sacks, 1987, Schegloff, 1987, Schegloff, 2007, pp. 58–73).

It has been well-established in the CA literature that participants jointly orient toward completing a sequence in which the FPP and the SPP are in a preferred relationship; not only do recipients prefer their SPP to be in agreement with the FPP in a contiguous manner, but FPP speakers also modify their turns upon the projection of disagreeing answers (Davidson, 1984, Pomerantz, 1984, pp. 152–163; Sacks, 1987, pp. 63–65). Exploring various action types including an invitation, offer, request, or proposal, Davidson (1984) analyzes a range of practices by which speakers of FPPs deal with potential or actual disalignment. Sacks's (1987) and Pomerantz's (1984) observation of non-continuous disaligning SPPs shows that silence, turn beginning sentence prefaces (well, uh:), and filled silence (breathing or laughing, or both) are often treated as disalignment-implicative. Here is an excerpt from Pomerantz (1984, p. 77):

The question in line 1 exhibits the questioner's preference for a ‘yes’ answer. However, the response is delayed, resulting in a discontiguity in the sequence. Upon the lack of response and the ensuing 1.7 s pause, A re-asks the question, its formulation now projecting a contrasting preferred answer. The revised question in line 3 is built for a ‘no’ answer, and B's response in line 4 is in accordance with this preference now projected by the question.

Davidson (1984) expands on Pomerantz's (1975, p. 82) work on weak agreements (hm, uh huh, Mm hm, yeah), showing how such weak agreements may also be treated as rejection-implicative in an invitation/request sequence. Excerpt 3 is taken from Davidson (1984, p. 113):

After a weak agreement (line 2) and a micropause (line 3), A produces another version of the request. The more emphatic subsequent version shows that A treats the weak agreement as the projection of a dispreferred answer.

In addition, Schegloff (1997) notes that third turn repair may be a locus for the speaker's attention to “getting it right” whether in reference to the recipient's (mis)understanding or independent from it (pp. 38–39). The following excerpt found in my data shows how the student orients to the correctness of the question formulation upon the teacher's weak agreement in the second turn.

In the prior turns, the teacher has suggested that the student organize her data into different categories. After acknowledging the suggestion, the student produces a question as an upshot of the teacher's prior turn and proposes parts of speech as a candidate category: so: (1.0) u:m like as a verb or as a noun o:r (line 1). The turn-final or relaxes the preference organization of the turn, marking its action as problematic in some way (Lindstörm, 1997). The teacher provides confirmation in line 2. However, the confirmation is preceded with a 0.5 s pause and marked with a flat intonation contour with vowel lengthening, indicating the answerer's uncertainty regarding the response and projecting disagreement (Stivers and Heritage, 2001, Verdugo, 2005). In the following turn, the student produces the or-prefaced turn and proffers an alternative category referring to their prior discussion: Or is it better to like denotative and connotative,. In addition to proffering the alternative category, the student orients to this alternative as a correction of the prior: Or is it better.

Finally, there is a growing body of research on multimodality (e.g., lexico-grammar, prosody, bodily-visual practices) and how different modes of organization carry out a social action in concert (see Bavelas et al., 2002, Ford et al., 2012, Goodwin, 2000, Goodwin, 2003, Hayashi, 2003, Hayashi, 2005, Heath, 1982, Kendon, 1967, Kendon, 1985, Kendon, 1990). In particular, the role of gaze in turn-taking has received an increasing attention; speaker gaze can be used to signal the beginning and the end of the turn as well as the turn transition to the next speaker (Bavelas et al., 2002, Duncan and Fiske, 1977, Heath, 1986, Kendon, 1967, Lerner, 2003, Rossano, 2006, Stivers and Rossano, 2010), and recipient gaze can be used to signal mutual orientation (Goodwin, 1979, Goodwin, 1986) and repair initiation (Seo & Koshik, 2010). Adding to this body of knowledge, this study examines the participants’ use of gaze shift as a visual-behavioral cue to convey dispreference in conjunction with their use of verbal formulations.

In the above examples (excerpts 2–4), the questioner repairs the original question in the third turn after the clear forecasting of dispreference (e.g., long pause) or the completion of a SPP response, albeit weak. On the other hand, the target self-repair practices in this study operate during some of the earliest moments of possible dispreference indication. The self-repairs are produced in the wake of recipient behavior that projects a dispreferred SPP (e.g., short pause, hesitation, gaze shift, tongue click), thereby achieving virtual ‘third turn’ positioning. The implications of such a practice, which wavers on the edge of third turn repair, will be explored in the discussion section.

Focusing on the students’ use of or-prefaced third turns, this study also contributes to our knowledge of the interactional functions of discourse markers. In both traditional and discourse-based grammar texts, the English word or is usually described as a coordinating conjunction (e.g., Berk, 1999, Givón, 1993, Huddleston, 1988, Quirk et al., 1985, Schiffrin, 1987). Generally, or has been shown to mark an option (e.g. a or b) for the hearer and can be used exclusively by forcing a choice or inclusively by indicating both are legitimate choices¹ (Berk, 1999). Or, then, is a central element in posing an alternative question that “expect[s] as the reply one of two or more options presented in the question […]” (Quirk et al., 1985, p. 806). The following excerpt exemplifies a question sequence with two alternative choices joined by the conjunction or. It is taken from a telephone conversation between a brother and sister.

Joyce presents two alternative options within her question, and Stan chooses one of them in the following turn. Following the type-conforming answer (Raymond, 2003) provided by the recipient, the questioner produces a news receipt token (Heritage, 1984) and closes the sequence.

Jefferson (1986) further suggests that, in discourse, or functions as a colligator to disguise items that are counterposed as equivalent options. She argues that colligation is one of the practices for minimizing disagreement in interaction, as “co-class membership, this ‘belonging together,’ is not necessarily inherent in the items, but can be imparted to them by their co-list membership” (p. 3). The following excerpt illustrates how or is used as a colligator to disguise disagreement (adapted from Jefferson, 1986, p. 12):

In line 12, Bill uses the conjunction or and presents ‘quarter inch’ and ‘half inch’ as two equivalents, when in fact ‘half inch’ could stand as a counter item. He self-repairs to propose ‘half inch’ as a correction of the prior ‘quarter inch.’ The self-repair shows how the parties to interaction use a grammatical item to achieve a particular action, namely, the conjunction or is used to disguise disagreement, retroactively making the original understanding (line 11) an alternative understanding with two options (line 12). Expanding on our knowledge of how the grammatical item or is used to accomplish different interactional goals, this study examines how students display their orientation to the correctness of the initial question formulation via the or-prefaced third turn self-repair.

Most broadly, adopting CA as its research method, this study expands our knowledge on the nature of educational discourse and how it is accomplished by participants as situated practices (Koole, 2012, Vanderstraeten, 2001). Drawing on the observations made by George Simmel, Vanderstraeten (2001) notes that education is based on social interactions in which the participants mutually perceive each other. By examining how both teachers as well as students produce their turns and respond to each other in naturally occurring conversations, this study contributes to achieving a fuller understanding of educational discourse as situated practices. In particular, analyzing student-initiated sequences, the study furthers our knowledge on shared as well as divergent orientations of both participants in educational discourse.

Traditionally, studies of educational discourse have largely focused on teacher-initiated sequences (Freedman and Katz, 1987, Koshik, 2002, Koshik, 2005, Margutti, 2010, Mehan, 1979, Sinclair and Coulthard, 1975). In particular, the teacher-initiated three-turn sequence, often known as an IRE (Initiation-Response-Evaluation) sequence, has been well-identified as one of the basic structures of classroom discourse (Mehan, 1979, Sinclair and Coulthard, 1975). The teacher's evaluative third turn confirms that the initiating question is formulated and understood by the participants as a known-answer question. Based on the identification of the IRE sequence, many studies have examined the teacher's action in relation to classroom instruction and student learning (see Hall, 1998, Lee, 2007, Nassaji and Gordon, 2000, Young, 1992).

This study shifts the focus to student actions and explores what students orient to in the educational context. In particular, it aims to show how students contribute to accomplishing situated practices on a moment-by-moment basis. Not only do students initiate a question–answer sequence, they also revise their question formulation or retract the question in the third turn upon the possible projection of the teacher's problematic answer. That is, they convey their orientation to getting a preferred response as well as their orientation toward the correctness and/or relevance of their initial question formulation. Such student orientations often contrast with teacher orientations in that teachers often treat the students’ third turn as being redundant in some way. Teacher responses are primarily directed to the students’ initial position and to providing a more generalizable lesson.