Recast And Second Language Development

Recasts have figured prominently in resent Second Language Acquisition (SLA) research, with studies documenting significant advantages for learners exposed to this type of negative feedback. Rod Ellis (2006) claims that recasts currently constitute an area of inquiry of great interest to SLA. The purpose of foreign language teaching lies not only in enabling learners to communicate in the target language but also in instructing learners to use it accurately. By means of providing corrective feedback, the relationship between communication and accuracy of language can be balanced. Moreover, compared with other corrective feedbacks, recasts are the most frequently used in classroom instruction. Research into recast, therefore, would be highly significant and practical. In the present study, we intentionally look at the issue of recast in terms of the acquisition of English relative clauses. The reason is that although relative clauses are universal in English, we can see from our observation of the acquisition of relative clauses by Chinese learners that they can not be processed without difficulties in communicative performance.The present study is carried out under the guidance of some theoretical viewpoints.1. Recasts not only offer implicit negative feedback, but also provide positive evidence.2. The implicit nature of recasts may make it difficult for learners to notice target forms provided by recasts and affect the uptake of them.3. Because of the disadvantages of recasts in use, many researchers explore effective types of recasts, some of which have been discussed in previous studies such as enhanced salience of positive evidence, negotiation plus recast, etc. However, few studies have investigated the relationship between different types of recasts and acquisition to date.This thesis is organized from the perspective of effects of different types of recasts on the acquisition of English relative clauses. Thirty students from two classes of English major in the same grade of a university in southern China were divided into four groups. Students of three groups were assigned to the three experimental conditions—one received corrective feedback in the form of Recast-only, one in the form of Stressed recast and the other in the form of Negotiation+recast. And a group of students was treated as the control group which received no feedback in the interaction. The experiment is designed to address the following three questions:1. Does exposure to input with recasts lead to greater L2 development than exposure to input without any feedback?2. Does exposure to recasts in which the corrected items are stressed lead to greater L2 development than exposure to recasts only?3. Does exposure to the negotiation-recast type feedback lead to greater L2 development than exposure to recasts-only or stressed recast? Among the three types of recast, which one is the most effective?A series of tests and treatment including sentence-combination and spot-difference are designed in the procedure of pretest, treatment, immediate posttest and delayed posttest. Judging from the analysis of the gathered evidence, the first question has been confirmed completely. That is to say, recasts can lead to more improvement of grammatical learning than input without corrective feedback. As for the second and third questions, the type of negotiation+recast is the most effective among the three types of recasts whereas recast-only is the least effective because the explicitness of the form of recast may influence learners'notice to the target form.Some pedagogical implications are put forward for the correction of learners'errors and focus-on-form instruction in the thesis. In addition, some questions, for instance, longitudinal study of effects of different types of recasts and time for recasting in classroom instruction, are suggested for further study.