Phonetic and Sequential Differences of Other-Repetitions in Repair Initiation

ABSTRACT

This article analyzes two different repair initiation practices that both utilize other-repetition. We call these framing and prefacing other-repetitions and show that they are treated as making different claims about the speakers’ depth of understanding of the prior talk. Framing repetitions repeat the turn-initial components of the prior turn with a particular “long and flat” phonetic pattern; prefacing repetitions consist of a minimal repetition of the final grammatical structures of the prior speaker’s talk, produced quietly and with a falling intonation contour. While framing repetitions are treated as displays of either a hearing or simple understanding problem, prefacing repetitions claim a more serious breakdown of understanding. Data are in British and American English.

Funding

Trevor Benjamin would like to acknowledge the support of the Center for Language and Cognition, University of Groningen, The Netherlands.

Notes

¹ For instance, some of our findings (for English) are congruent with Bolden (2009) for Russian, and we share the use of the term “prefacing repetitions,” but she reports finding 94 candidate instances in just 60 hours of conversation. There are other notable differences: Because our collections were built according to different sequential criteria, our prefacing repetitions are only comparable to Bolden’s category of “indicating information retrieval problems.” And even there, our findings are different—perhaps not unexpectedly, given the different grammatical and prosodic structures of the two languages under investigation.

² Note the embedded correction of “got” to “have”: Such practices have been discussed in detail in Jefferson (1987), and we will not discuss them further here.

³ Here the name of the street and the city name have been changed.

⁴ This is not the case for all verbs—e.g., sleep cannot take an object, and catch requires one.

⁵ Or even two turns, given the pause in Fragment 2. The use of pauses between the other-repetition and the continuation by Speaker B is discussed in the conclusion.

⁶ The duration and speech rate of prefacing repetitions was examined and compared with surrounding speech, but no patterns were found. See the following section for a discussion of the difficulties concerning what to measure and how to compare any measurements.

⁷ See the discussion in the next section regarding inconsistencies in the data prohibiting a fully systematic measurement of intensity.

⁸ Because Speaker B does not continue her own talk right away, the intensity of her subsequent turn (after talk from Speaker A) was measured.

⁹ Szczepek-Reed (2004) describes the use of level intonation at the end of turns in making the point that intonation may not always, or only, contribute to the signaling of turn-taking. We are sympathetic to this view and analyze the function of level intonation in these data as more concerned with marking out these repetitions as inviting a particular kind of completion than with simply signaling that the current speaker is done talking.

¹⁰ An anonymous reviewer has suggested that the differences in treatment (as a hearing versus understanding problem) might be better explained by grammatical (expressed versus implied constituents) rather than turn-constructional considerations. We have attempted several analyses based on constituency and subjecthood, but no commonality captured by a single grammatical term seems to account for the data set as a whole—not in the same way as “objecthood” seems to work for the prefacing repetitions.

¹¹ It should also be pointed out that Bolden’s findings for repeat prefacing as a practice are not necessarily directly comparable with ours regarding prefacing repetitions, for the reasons noted in footnote 1.

Additional information

Funding

Trevor Benjamin would like to acknowledge the support of the Center for Language and Cognition, University of Groningen, The Netherlands.