Repair in the lab hour: second language interactions between Korean TAs and native English-speaking students

textLab hours in the science/engineering departments of U.S. universities provide a special conversational event in which a significant number of nonnative (NNS) graduate students, as teaching assistants (TAs), work with native English-speaking (NS) students to instruct content matters. NNS TAs frequently engage in spontaneous, one-on-one interactions with NS students. Previous research has not addressed specifically which aspects of the NNS TAs’ verbal and non-verbal actions contribute to reaching a shared understanding. Through the microethnographic analysis of verbal and nonverbal realizations of repair, I examined videotaped one-on-one interactions of five Korean TAs interacting with twelve NS students. I also analyzed the interactions of four NS TAs communicating with nine NS students in order to look at how nativeness and nonnativeness affected the meaning negotiation process, i.e., “repair.” From this data I identified factors that were crucial to efficient repair, and more importantly, in which areas of the second language NNS TAs encountered the most difficulties. The most problematic aspects were in L2 discourse and pragmatics. For example, the NNS TAs often failed to enact proper instructorship, as their points did not entail follow-up explanations. Moreover, Korean TAs’ responses to NS students’ corrective actions were often indirect, irresponsive, or delayed. These pragmatic flaws complicated the NS students’ understanding, resulted in unsuccessful repair, or increased the tension between participants. Consequently, participant alignment became unstable such that NS students held speakership, and played a powerful role during repair. The students’ misunderstanding was prolonged, and meaning was shared only after extended talk. While making efforts to overcome their limited L2 competence, the NNS TAs employed various strategies. They used a Korean information structure, and used English discourse markers not in concert with English language norms (e.g., use of “kin’ of” as a filler). Their initiatives of trouble were often inefficient because the TAs overwhelmingly depended on unspecific devices. The ways in which meaning-making resources alongside language were coordinated affected all the levels of the repair sequences. The findings suggest that limited L2 competence is embodied in the ways in which entire interactional resources, rather than one resource separated from another, are coordinated.Curriculum and Instructio