英语课堂中口语纠错面面观(txt+pdf+epub+mobi电子书下载)

作者：蒋景阳

出版社：浙江大学出版社

格式: AZW3, DOCX, EPUB, MOBI, PDF, TXT

英语课堂中口语纠错面面观试读：

负反馈是指教师对于学习者含有错误的话语的反馈。其目的在于提高语言学习者对自己的口语输出和目的语之间差异的注意程度，提高口语的表达能力和准确度。半个世纪以来，负反馈受到了人们的广泛关注，不仅因为它是第二语言课堂中重要的教师行为，而且它与语言教学和语言习得理论密切相关，同时也与第二课堂教学的有效性有关。

但是，负反馈以及它的具体效果一直是人们争论的焦点。先天论者强调普遍语法和语言习得机制，认为学习者语法系统的变化是正面的语言证据的结果，负反馈的作用极为有限，它无法改变中介语的语法系统。和先天论者持相同观点的自然法也认为负反馈只能影响语言学习而不能影响语言习得。他们认为课堂内大量的纠错行为是无效的。

但也有很多学者认为负反馈是必要的、有效的。他们认为教师的反馈可以给学习者提供大量的语言输出的机会，有助于学习者更多地进行语法方面的加工，从而提高语言运用的准确性。“交互假设”认为错误是可以在自然的互动中通过纠错反馈得到纠正的。许多认知心理学理论也支持纠错反馈的作用。例如Schmidt的“注意假设”认为为了学习语言形式，“注意”是必不可少的，它是把“输入”转变成“吸收”的充分必要条件，而负反馈能够促使学习者有意识地注意语言形式，促进中介语的发展。

蒋景阳博士的此项研究围绕着二语课堂中教师针对学生的口语错误提供负反馈的过程展开。通过课堂的实况录音、转写、标注和分析，以及比较控制班和实验班的口试成绩、语法成绩等，来检验在英语作为外语的课堂教学中，非刻意负反馈在以意义为重点的口语活动中的有效性。也就是说，她探讨了课堂内针对学生口语错误的教师的负反馈是否能够提高学生的口语精确度和语言能力。该研究针对目前大学英语课堂学生口语错误的分布情况、教师提供负反馈现状，以及是否该纠错、如何纠错等问题得出了较为可信的结论和启示，是国内较为少见的针对负反馈的系统研究，对于一线教师的教学能提供很有价值的参考。

该书是蒋景阳老师在她的博士学位论文的基础上修改充实而成的。作为她博士论文的指导教师，我认为她的这篇论文不但选题有价值，研究方法合理，写作规范，而且也反映了她扎实的英语语言功底，是一项质量较高的研究成果。束定芳2011年7月26日于上海自序　Preface

此书是在我博士论文的基础上修改完成的。在论文的撰写过程中，得到了我的指导老师上海外国语大学束定芳教授的精心指导和帮助，在此深表感谢。

我从事大学英语教学迄今已有25年。尽管时间不短，但是在实际的课堂教学中，到底是否该纠正学生的口语错误以及如何纠正一直是我心中的疑惑。尽管负反馈在二语习得中一直被认为是有一定作用的，尤其是有助于学习者的中介语向目标语靠近。但是，由于缺乏对负反馈的实质和作用方面的实证研究，它的作用一直也被怀疑。从教学法的角度，许多教师因为害怕负反馈会挫伤学生发言的积极性，也不愿意使用负反馈，而是听之任之。因此，在选题的时候，我毫不犹豫地选了针对负反馈的研究，以解答很多像我这样的大学英语教师的疑惑。

本研究的受试是某个全国重点大学非英语专业的学生。采用的数据为：课堂上课时师生交互的录音，口语预测和后测，两次语法对比考试，实验前后的关于学生对负反馈看法的问卷。研究发现，在以意义为主的英语作为外语的课堂内，学生对负反馈普遍持积极的态度，教师学生互动时所使用的非刻意负反馈能够提高学生的口语精确度。负反馈的有效性揭示，注意和意识在语言学习中是至关重要的。学习者确实有“小小的认知窗口”，用来注意师生互动中教师的负反馈以及语言意义和语言形式之间的互动。教师的负反馈有助于提高学生对语言形式的注意和意识，从而对语言学习有促进作用。同时，该研究还总结了课堂内学生口语错误的类型和频率、教师负反馈的实际使用情况等。

在论文的写作中，我要衷心地感谢何莲珍教授、庞继贤教授、张建理教授、朱晔副教授、李德高副教授的鼓励和支持，以及帮助我实验的周颂波老师和方富民老师。我还要感谢Richard Xiao，他所教授的“语料库语言学”使我受益匪浅。最后，还要感谢我从未见过面的Paul Rayson,允许我使用他们研发的Wmatrix软件，以便更加科学地对数据进行分析。蒋景阳2011年7月于浙江大学摘要　Abstract

It has long been argued that negative feedback plays a role in second language acquisition, especially in contributing to the development of learners’ interlanguage and advancing to more targetlike language. However, due to the lack of empirical studies with regard to the true nature and function of NF, its effectiveness is hanging in doubt. From the perspective of teaching pedagogy, many teachers, in fear of discouraging students from speaking, grudge providing NF, leaving the students to their own devices.

The purpose of the present research is to examine the effectiveness of incidental negative feedback in meaning-focused speaking activities in EFL classroom setting through comparing the results of oral tests between control classes and experimental classes. That is to say the paper is intended to explore whether teachers’ NF directed at students’ oral errors in the classroom setting can help improve students’ accuracy in later oral performance. The research involved three separate but related experiments to examine the effect of NF in actual classroom teaching environment. These three experiments are designed with some variations so that more comprehensive findings and implications can be drawn.

The subjects were six intact classes of about 260 freshmen non-English majors in a key comprehensive university. Two classes （one control class and one experimental class） were taught by the researcher herself and other four （two control classes and two experimental） by her colleagues. All the three experiments lasted about 6—7 weeks. The data were collected from different sources, namely: recording of teacher-student interactions in the classroom, pre- and post-oral tests, two comparative grammar tests, two before and after questionnaires.

The major findings of the present research are summarized as follows:

1） Students committed grammatical errors the most, then lexical errors, phonological errors and L1 errors. But a vast majority of errors went untreated. L1 errors were treated the most frequently, then lexical errors, phonological errors, and last grammatical errors. That is to say, even though the most frequently occurred errors were grammatical errors, they were not the ones that were treated by the teacher most. On the contrary, they were the least treated.

2） Of all the seven types of NFs employed by the three teachers, recast ranked the first. Then came clarification request, repetition, complicated feedback, elicitation, explicit correction, and metalinguistic feedback. The distribution was very uneven, with recast high up in the rank. The last four types were almost negligible. As to the types of NFs following types of errors, recast was the most often used NF for teachers to treat all four types of students’ errors. The average successful rate of uptake for the three teachers was higher than unsuccessful rate. Clarification request was the most effective in leading to successful uptake, followed by repetition, recast, and complicated feedback.

3） Recast was more effective when it was more consistently, intensely administered and focused on a certain linguistic feature. Short, timely and emphasized recast with pauses and students’ awareness helped the students to compare their own erroneous utterance and the targetlike utterance and to notice the gap, and thus led to more successful uptakes.

4） Students held positive views towards NF. They were very satisfied with the three teachers’ actual NF treatment, including frequency of NF, timing, and the types of errors being treated. They were very positive that teachers’ NF had helped them raise their awareness of accuracy and also improved their accuracy. They also held that teachers’ NF had helped clarify what their classmates had said. They did not think NF made them afraid to speak, and they thought highly of the effect of NF, and hoped to be exposed to NF in the future.

5） NF helped to improve students’ overall oral performance, especially the use of past tense which was the grammar focus of the present study. This was proved by the significantly improved oral test scores of the three experimental classes, and students’ actual use of past tense in oral tests of Experiment 1.

6） NF administered during oral interaction didn’t lead to the significant improvement of students’ grammar knowledge in written tests for both control and experimental class. This might suggest that NF targeted towards oral errors will not necessarily have significant impact on learners’ written grammar knowledge, especially when this grammatical feature is mastered by the learners and it only causes trouble when used orally.

In conclusion, incidental NF in meaning-focused EFL classroom can help improve students’ accuracy in oral production while it is administered during teacher-student interaction. Findings of the present study can help us to better understand students’ beliefs towards NF, the errors committed by the students in oral production in the classroom in relation to error type and frequency, teachers’ actual use of NF, and the effectiveness of NF in improving accuracy in oral production, and thus overall oral proficiency.

The effectiveness of NF suggests that attention and awareness are crucial in language learning. Learners do have “small cognitive windows” for teachers’ NF during teacher-student interaction, and for the interaction between meaning and form. Teachers’ NF can raise students’ attention and awareness to form and thus have a facilitative role in students’ language learning.

Key words:　error treatment, negative feedback, uptake, oral accuracyChapter 1Introduction1.1　The Orientation of the Present Study

This study focuses on teachers’ negative feedback （NF） towards students’ errors in speaking in the classroom setting. The students in the current study are non-English major freshmen taking the course of College English Band 4. The study has several purposes. First, it is intended to describe the patterns of teachers’ error treatment sequence （ETS）, including what errors to correct, when to correct, how to correct, as well as students’ responses to the negative feedback （including successful uptake, unsuccessful uptake, no-uptake and no opportunity）. Second, this paper is also intended to find out students’ beliefs and attitudes towards teachers’ NF. Finally, it wants to explore the effects of teachers’ negative feedback on students’ subsequent language development, especially oral performance, after they are exposed to teachers’ negative feedback. Since the course in the present study is not grammar course but integrated language course, the treatment of errors is not systematic or planned. Rather, it is incidental negative feedback with primary focus on meaning. Basically, the NFs occur during the teacher-student interaction in the classroom while the rest of the students are listening at the same time.1.2　The Rationale of the Present Study

Although there is general agreement that accuracy is an important classroom language acquisition goal, it is by no means clear how learners can best be assisted via classroom procedures and tasks in becoming more target-like （Doughty ＆ Valera, 1998:114）. Input and output have been regarded as two crucial components in helping develop learners’ language abilities. Negative feedback, as a form of input, has its own role to play in the process of developing learners’ linguistic ability and improving accuracy. Negative feedback itself and the way NF is administered have always been a controversial and interesting research field. The role of corrective feedback （CF） in the process of learning a foreign language is still much debated, and it is closely related to the conception of the role of different kinds of language input in language acquisition （Doughty ＆ Williams, 1998b）, i.e. whether it is positive evidence or negative evidence that has the greater impact. Different linguists and researchers hold totally different views towards the use and function of NF.

During classroom teaching, teachers discreet point presentation and feedback on error are two common characteristics （Krashen ＆ Selinger, 1975）. With the advent of communicative language teaching （CLT）, the attitudes and practices towards error and feedback on error have also changed. Some second language （L2） teachers who are trained to teach in some communicative and content-based approaches believe that errors are part of the learners’ language development. They have come to believe that, given sufficient opportunities to hear, read and use of the language, learners will eventually reach those levels without direct feedback on errors or explicit explanation of their errors. Some even believe that error correction is best avoided. It can lead to only superficial or temporary changes in learners’ language performance and is not worth the risk of negative affective reactions （Truscott, 1999）. Some nativists, for example Chomsky （1975）, hold that negative evidence doesn’t play a role in language learning. They believe what makes language acquisition possible is Universal Grammar （UG） proposed by Chomsky: “the system of principles, conditions, and rules that are elements of properties of all human languages” （1975:29）. UG advocates have argued that instruction, including negative evidence, has little impact on forms within UG anyway, since it will temporarily change only language behavior and not interlanguage （IL） grammars （Carroll, 1996; Cook, 1991; Schwartz, 1993）. In this view, human beings depend on a “Language Acquisition Device” （LAD）, which allows them to learn language on the basis of the positive linguistic evidence. The IL grammars are the result of positive linguistic evidence not the negative evidence.

Lightbown （1985:178） also holds a negative view about the effect of NF. He argues that since errors are not isolated but represent the learners’ IL. So in order for error correction to work, there must be a change in the language knowledge, i.e. there must be a restructuring of the language system itself. A random or single error correction is not able to achieve the purpose. He argues that correction tends to be unreliable in two ways: 1） Learners cannot depend on having every error corrected—even in the classroom—and this means that they cannot depend on having someone else tell them when they are wrong; 2） the “corrector” may not know—indeed probably knows only rarely—what the real nature of the learner’s error is, that is, what it represents in terms of underlying knowledge.

Krashen （1982, 1985） on the whole is not optimistic about error correction. He holds that acquisition occurs when acquirers understand input for its meaning, not when they produce output and focus on form. Krashen holds that explicit data, whether in the form of negative evidence or in the form of explicit instruction, can only affect the learning rather than the acquisition of the target language. According to the Monitor Hypothesis, conscious learning can only act as an editor, and as a monitor, in the output of the acquired system. According to Krashen’s Input Hypothesis, it is subconscious acquisition that gains dominance, and that learning does not turn into acquisition.

Krashen’s views about language learning and acquisition have been challenged because mere input, no matter how comprehensible it is, is not enough for language development. Long （1985） acknowledges that simplified input and context can play a role in making input comprehensible, he stresses the importance of the interactional modifications that occur in negotiating meaning when a communication problem arises. In Long’s （1996） updated version of the interactionist hypothesis, he proposes that “Environmental contributions to acquisition are mediated by selective attention and the learners’ developing L2 processing capacity, and that the resources are brought together most usefully, although not exclusively, during negotiation for meaning”. Negative feedback obtained during negotiation work or elsewhere may facilitate SL development, at least for vocabulary, morphology, and language specific syntax. According to this model of acquisition, interaction may facilitate L2 learning by providing learners with NF, drawing their attention to language form in the context of meaning, and pushing them to produce more complex or accurate target forms so as to make the communication successful and go on smoothly.

Many other linguists also stress the importance of drawing learners’ attention to form and the function of NF in L2 acquisition. Ellis, R （2001b） et al. hold that L2 learners need to call on general inductive learning mechanism, which uses negative evidence, and form-focused instruction that makes such evidence available is not only helpful but even necessary for adult learners to acquire an L2. According to information-processing models, due to limited processing capacity, learners have difficulty in attending simultaneously to form and meaning. They may be forced to rely on top-down strategies such as guessing and predicting. VanPatten’s （1990） experimental study found that attention to form in the input competes with attention to meaning. That is to say learners need specific activities that draw attention to form. According to Schmidt’s （1990） Noticing Hypothesis （NH）, noticing is required in order to learn language form and subliminal learning is impossible. Noticing is the necessary and sufficient condition for converting input to intake. Form-focused instruction helps learners to pay conscious attention to forms in the input that might otherwise be ignored. This can assist IL development.

In the view of SLA as cognitive skill acquisition, language acquisition includes interaction between input, the cognitive system, and the learner’s perceptual motor system. According to this model of language learning, feedback is essential. Han （2001） holds that this is due to the fact that “it has the properties of informing, regulating, strengthening, sustaining, and error eliminating”. Gass （1988, 1990） also argues against the notion that mere comprehensible input would turn it into intake and then to output. She holds that in order for the learners to internalize the input and then acquire the input, they must not only comprehend the input, but also notice the gap between the input and their IL. She holds that “nothing in the target language is available for intake into a language learner’s existing system unless it is consciously noticed”. Corrective feedback is regarded by Gass as having the function of getting attention. She further argues that without negative feedback, which helps learners to discern the discrepancies between the learner language and the target language, fossilization may occur.

There is some empirical evidence that negative evidence is in some instances necessary for second language acquisition （SLA） especially for adults （Gass and Selinker, 2001:282）. White’s （1991） study shows that negative evidence may be necessary to trigger a permanent change in a learner’s grammar. It doesn’t show, however, that positive evidence alone is sufficient. Trahey and White （1993） conducted a follow-up study to determine the effect of positive evidence. Both classes were given input flood of English adverbs （positive evidence only） for two weeks. They found that positive input was sufficient for learners to notice that certain language structures are possible in English, but it was not sufficient to detect the ungrammaticality of some structures. The study showed that positive evidence can reveal to learners the presence of information in the SL that is different from their native language, but that negative evidence is necessary to show what is not possible in the SL when it is possible in the native language.

Han （2004） listed many factors that caused fossilization. The following are of close relation to the present study: absence of corrective feedback, lack of attention, inability to notice input-output discrepancies, the speed with which, and extent to which, automatization has taken place, natural tendency to focus on content, not on form, failure to detect errors, failure to resolve the inherent variation in the IL and lack of sensitivity to input.

Different studies abroad have been done on NF with different settings, such as 1）immersion programs with young children, 2）content- or communication-based classroom teaching, 3） laboratory, 4） dyadic-interaction, 5） foreign language classroom for adult learners, and 6） foreign language classroom for university students. These researches did not fit into the context of language teaching in China. First, many studies were done in the ESL environment. The learners were learning another language in the country where the language is spoken. But in China, English is taught as a foreign language. There is little input and interaction outside the class. They have no native speakers （NSs） to talk to outside the class, not to mention negative input. What the teacher says in the class, and what they read or listen to in the class constitute the most part of their input. Second, the teachers are mostly native-speakers, or native-like speakers. Some studies are dyadic situations between NS and Non-Native Speakers （NNs） in natural settings, which is different from classroom setting. But in China, the native language of the vast majority of teachers is Chinese even though they teach English. Third, the subjects in many studies done abroad were either very young learners （i.e. primary school students） in an immersion program or older adults （20—50） who were EFL learners. Lastly, “there have been surprisingly few longitudinal studies of classroom learners” while “descriptive studies of learner language ideally need to be longitudinal to show how instructed learners’ interlanguage developed over time （Ellis, R, 2001:29-30）. All these differences make this study very necessary since the applicability of all those results done in different contexts is held in doubt in China.

In China, some studies （Tang, 2003; Hu, 2005） have also been done on NF in the EFL setting in colleges. Tang’s study was a longitudinal one that lasted three months. It explored the effectiveness of error treatment on immediate uptake and delayed uptake through collecting and analyzing actual classroom teaching data. The study was sound in its own right. But the subjects of Tang’s study were all female English majors. The fact that all the subjects were females made the results less representative and less convincing. What is more, it was an ex-post facto type of research. It was rather observational than experimental. There was no experimental class and control class. The relationship between the independent variables and the dependent variables was not of cause-and-effect relationship.

Hu’s study compared the different ways of error treatment

试读结束[说明：试读内容隐藏了图片]

下载完整电子书