Negative Feedback And Learner Uptake:A Study Of Teacher-Student Dyadic Negotiation In College English Classrooms

In recent two decades, negative feedback has been a research focus in the field of SLA. An examination of the empirical studies investigating the relationship between negative feedback and learner uptake has revealed two problems with this line of research. One problem lies in that the existing classification frameworks of negative feedback have shown defects in its orientation and category types. Another problem pertains to the situation in which researchers generally examine the relationship between the two variables from a cognitive perspective, while few have touched upon the process of how negative feedback facilitates language acquisition.The purpose of the present study is two-fold:one is to build a classification framework for negative feedback which aims at both learners’erroneous language forms and content problems; the other is to follow a discourse-analytic pattern to examine the dynamic process of how negative feedback facilitates language acquisition.875 students from a university in Nanjing participated in the study. The teachers made recordings respectively in their classrooms whenever they had dyadic interactions with their students. The recordings were made over 11 weeks, and the database analyzed for the present study totaled 13 hours and 50 minutes. The recordings were transcribed and the data were analyzed by the researcher.The research findings of the present study are as follows:(1) In College English classrooms, learner errors or problems involve five major types:lexical errors, phonological errors, syntactic errors, content problems, and use of LI. Learners’lexical errors and phonological errors are more likely to solicit teachers’negative feedback than content problems. Syntactic errors are least likely to solicit teachers’negative feedback.(2) On the basis of the existing literature and the data for the present study, a new classification framework of negative feedback is formulated. The framework includes three general categories:error-induced negative feedback, content-induced negative feedback, and translational feedback, within which translational feedback is a newly set general category. In the new classification framework, three sub-categories are newly found:true-and-false feedback, repetition in meaning and translational feedback. However, four categories in the literature cannot be found in our data. They are metalinguistic feedback and three subcategories of recasts:isolated interrogative recast (with pragmatic hint), incorporated interrogative recast, and incorporated interrogative recast (with pragmatic hint).(3) In all negative feedback types, recasts are the most frequently used feedback type. Teachers usually provide recasts in declarative forms with no pragmatic hint, and the recasts are often in the isolated form.(4) Four general categories of learner uptake have been identified in our data: acknowledgement, repair, needs-repair, and repair in Chinese, and the fourth one is a new uptake type found in our data.(5) The amount of the negative feedback is not in positive proportion with the amount of learner repair. In terms of learner uptake after recasts, learners correct their errors in 41%of the time. Specifically, after isolated declarative recast, learners correct their errors in more than half of the situations; after incorporated declarative recast, learners correct only 1/5 of their errors.(6) Negotiation routines involving negative feedback and learner repair usually consist of 5 turns and 3 turns, and those consisting of over 6 turns are rarely used. Furthermore, one round of negative feedback may not result in learner repair; the repair may be the outcome of multiple rounds of negative feedback.(7) As only about half of the negotiation routines in our data can match Varonis and Gass’(1985) model of negotiation between NNSs, based on their model, a model of teacher-student dyadic negotiation on negative feedback and repair is set up. The model shows the downward movement from negative feedback to repair, and also presents multiple layers of the movement. Through the simple and the complex movements, learner repair is made after the teacher’s negative feedback and eventually facilitates language acquisition.