Abstract

This article presents a study of corrective feedback and learner uptake (i.e., responses to feedback) in four immersion classrooms at the primary level. Transcripts totaling 18.3 hours of classroom interaction taken from 14 subject-matter lessons and 13 French language arts lessons were analyzed using a model developed for the study and comprising the various moves in an error treatment sequence. Results include the frequency and distribution of the six different feedback types used by the four teachers, in addition to the frequency and distribution of different types of learner uptake following each feedback type. The findings indicate an overwhelming tendency for teachers to use recasts in spite of the latter's ineffectiveness at eliciting student-generated repair. Four other feedback types-elicitation, metalinguistic feedback, clarification requests, and repetition-lead to This research was funded by grants from the Quebec Ministry of Education's research funding agency (Fonds pour la formation de chercheurs et l'aide à la recherche) and the Social Sciences and Humanities Research Council of Canada. We gratefully acknowledge the generous cooperation of the participating teachers and their students. We are also grateful to the following research assistants, who contributed to various phases of this study: Dawn Allen, Linda Corliss, Yakov Goldstein, Randall Halter, Thierry Karsenti, Lyne Laganière, Tamara Loring, Nina Padden, James Poirier, and Nicole Sabourin. Finally, we thank Patsy Lightbown, Nina Spada, and four anonymous reviewers for their helpful comments on earlier versions of this paper. Address correspondance to Roy Lyster, Department of Second Language Education, McGill University, Montreal, QC, H3A 1Y2; e-mail: cxrl@musica.mcgill.ca, or to Leila Ranta, e-mail: laranta@vax2.concordia.ca. ERROR CORRECTION AND COMMUNICATIVE LANGUAGE TEACHING The issue of how competent speakers react to learners' language errors takes on many guises depending on the disciplinary orientation of the researcher. It has been examined in terms of negative evidence by linguists (e.g., There are, however, a number of good state-of-the-art discussions of error treatment in classroom SLA In contrast to the experimental orientation of the studies that have examined the utility of error correction, the other questions posed by Hendrickson (concerning when, how, and which) have been addressed by observational studies of teachers in their own classrooms. 1 Thus, an early study by Of the larger scale investigations, Hamayan and Tucker (1980) examined specific aspects of teacher input provided to French immersion and French first language (L1) classes at the Grade 3 and Grade 5 levels in the Montreal area. One part of the study focused on teaching strategies, including error correction practices. They found a significant difference between the L1 and L2 teachers' correcting behavior: The French immersion teachers tended to explicitly correct the L2 learners more in the earlier grade than in the later grade, but the reverse pattern was found with the native speaker classes where more correction took place in the higher grade. Finally, in the context of adult French foreign language learning in Australia, Doughty (1994a) reported on a pilot study involving 6 hours of interaction in one class on three different occasions. She was interested in comparing L2 classroom learning to L1 acquisition in terms of the degree to which feedback is finely tuned. Student turns were coded for number of errors, and teacher turns were coded for the type of feedback (either clarification request, repetition, recast, expansion, or translation). The teacher provided feedback on 43% of the erroneous learner turns and tended to give feedback to learner turns that had only one error rather than those with many errors. The teacher also tended to repeat student utterances only when they were well formed. Doughty concluded, tentatively, that the feedback to L2 learners is predictable in the same way that the input for L1 learners is and that learners respond in a way that suggests that the information conveyed by the teacher is noticed. This result is in sharp contrast to Fanselow's findings and may highlight the expertise of this particular teacher and the expertise of these particular students rather than providing a generalization about L2 learning in the classroom. This review of existing studies concerning error treatment in communicative language teaching provides a sketchy picture at best. Further research that examines relationships among different variables in a variety of teaching contexts is necessary. In this paper, we propose to respond to this need by reporting on a study whose purpose is twofold: 1. To develop an analytic model comprising the various moves in an error treatment sequence in L2 classrooms. 2. To apply the model to a database of interaction in four primary-level French immersion classrooms with a view to documenting the frequency and distribution of corrective feedback in relation to learner uptake, that is, student responses to corrective feedback.