Liu Zhenwei

1. Introduction

In the early 1970s， however， some misgivings about the CA began to arise. Moreover， many of the predictions of target language（TL）learning difficulty formulated on the basis of CA turned out to be either uninformative or inaccurate（Larson-Freeman， 1991）. In an attempt to reconcile that dilemma， Wardhaugh（1970） proposed a distinction between a strong version and a weak version of CA. The strong form claims that all second language（L2） errors can be predicted by identifying the differences between the target language and the learners first language. The weak form of the hypothesis claims only to be diagnostic. A contrastive analysis can be used to identify which errors are the results of interference. Implicit in the weak version is the assumption that not all errors are the result of interference， and thus claims a less powerful role for the L1. Thus， although CA might not be useful a priori， it was still claimed to possess a posteriori explanatory power. As such， it was useful in a broader approach to detecting the source of error， namely error analysis （EA）. And an article entitled “The Significance of Learns Errors” published by Corder in 1967 marked the threshold of EA.

2. Major Viewpoints

Based on cognitive psycholinguistics and Chomskys Universal Grammar， EA is a type of linguistic analysis that focuses on the errors learners make. Unlike CA， the novelty of EA was that the first language （L1） of the learners was not at all involved and errors could be fully identified in terms of the TL.

2.1 Classifications of errors

Errors can be classified into many categories by various criteria， and the following classifications are summaries from their own perspectives of different researchers.

2.1.1 Errors and Mistakes

Corder（1967） distinguished between errors and mistakes. Mistakes are akin to slips of tongue. That is， they are generally one-time-only events. It may be called behavioral mistake as well. The speaker who makes a mistake is able to recognize it as a mistake and correct it if necessary. An error is， on the other hand， systematic， and thus called systematic mistake. That is to say， it is likely to occur repeatedly without being recognized by the learner as an error.

According to Ellis （1997）， errors reflect the gaps in the knowledge of a learner； they occur because he does not know what is correct， and mistakes reflect occasional lapses in performance； they occur because， in particular instance， the learner is unable to perform what he or she knows. One way is checking the consistency of learners performance. Another way might be asking learners to try to correct their own deviant utterances.

2.1.2 Global Errors and Local Errors

Global errors affect the meaning of the whole sentence， or meaning relations between sentences， and can， therefore， seriously interfere with communication， while Local errors， such as article omission for instance， only affect the meaning of a single constituent within the sentence. Local errors are less likely than global errors to block communication （Larson-Freeman， 1991）.

2.1.3 Classifications by Chinese scholars

In China， Cai Longquan and Dai Weidong has proposed four principles of error classification， namely， self-sufficient， interchangeable， systematic and objective. Under such principles， errors are classified into the following three types： learner， content of learning and behavior of learning. They also put forward a set of parameters for error classification： quantity， characteristics and sequence. Totally， they have identified twenty varieties of representative errors.

2.2 Steps in EA research

Corder（1994）suggests the following steps in EA research， and more detailed information about the steps will be illustrated as the following：

2.2.1 Identification of Errors

To identify errors is a tough task. Even if you have the linguistic sensibility of native speaker， still it is difficult to identify the errors immediately for various reasons such as the implicit nature of errors.

2.2.2 Description of Errors

The description of learner errors involves a comparison of the learners idiosyncratic utterances with a reconstruction of those utterances in the target language. One of the simplest ways is to classify errors into grammatical categories. For example， we gather all the errors relating to verbs and then identify the different kinds of verb errors in our sample （Ellis， 1997）. But this taxonomy sheds little light on how learners learn an L2. Corders（1974） framework for describing errors is more promising in this respect. He distinguishes three types of error according to their systematicity：

I. Presystematic errors occur when the learner is unaware of the existence of a particular rule in the target language. These are random.

II. Systematic errors occur when the learner has discovered a rule but it is a wrong one.

III. Postsystematic errors occur when the learner knows the correct target language rule but uses it inconsistently （i.e. makes a mistake）.

2.2.3 Explanation of Errors

Explanation is concerned with establishing the source of the error， i.e. accounting for why it was made. This stage is the most important for second language acquisition（SLA） research as it involves an attempt to establish the processes responsible for SLA.

Richards （1972） distinguishes three different sources or causes of competence errors have been identified： interference errors， intralingual errors and developmental errors.

But most researchers have operated with a general distinction between transfer errors and intralingual errors， and transfer errors can be further subdivided into overextension of analogy， transfer of structure and interlingual/intralingual errors. Intralingual are also often further subdivided into the following： overgeneralization errors， ignorance of rule restrictions， incomplete application of rules， and false concept hypothesized （Richards， 1974）. Again， it is important to recognize that the concept of “transfer” is， in fact， a very complex notion which is best understood in terms of cognitive rather than behaviorist models of learning （Ellis， 1994）.

2.2.4 Evaluation of Errors

Error evaluation involves a consideration of the effect that errors have on the persons addressed. The design of error evaluation studies involves decisions on who the addressees will be， what errors they will be asked to judge， and how they will be asked to judge them.

3. Application and Implication for Foreign Language Learning and Teaching in China

3.1 Strategies of Correcting Errors

Language learners committing errors is a part of their natural learning process， which indicates language learners language is on a certain developmental stage， and it will become more precise in terms of meaning and more exact in terms of form. As for error correction， going to extremes is not recommended. On the one hand， not all errors that learners committed should be corrected at once； and on the other hand， neglecting all of their errors is also infeasible. For example， local errors affect only a single constituent in the sentence and can be neglected， whereas global errors violate the overall structure and must be corrected immediately. The same is true with mistakes and errors.

In foreign language teaching， teachers always encounter such a problem as when the appropriate time to correct students errors is， or the dilemma of fluency vs. accuracy. Generally speaking， as a beginner， accuracy comes first， while later the case is the other way round. Besides， to maintain the activity on learning of students and activate their initiatives also count， which requires teachers to pay more attention to manners of correcting errors.

3.2 Application and Implication of EA in Chinese Context

As far as foreign language teaching in China is concerned， for one thing， EA is conducive to revealing the psychological process and features of Chinese learners and further enriching theories of foreign language teaching in China. Moreover， the study of errors is essential to find out how learning goes， and errors can be regarded as a window leading to the learners inner heart and his or her learning strategies. For another thing， the study of learners errors is helpful in regard to improving teaching methods， adjusting syllabus design， and textbook compiling etc.

4. Conclusion

EA， one of the first methods used to investigate learner language， showing that many of the errors that learners make cannot be put down to interference， has achieved considerable popularity in the 1970s， replacing CA， and then lost its popularity as a result of its inherent weaknesses. But above all， it helped make errors respectable-to force recognition that errors were not something to be avoided but were an inevitable feature of the learning process. It has made substantial contribution to SLA research （Ellis， 1994）.

Reference：

[1]Corder，S.P.1967.The Significance of Learners Errors[J].International Review of Applied Linguistics，（5）161-169.

[2]Ellis，R.1994.The Study of Second Language Acquisiton.Oxford：Oxford University Press.

[3]Ellis，R.1997.Second Language Acquisition.Oxford：Oxford University Press.

[4]Larson-Freeman，D.and Long，M.1991.An Introduction to Second Language Acquisition Research.Longman.

[5]Richards，J.C.（ed.）.1974.Error Analysis：perspectives on SLA.London：Longman.

[6]Richards，J.C.et al.1992.Longman Dictionary of Language Teaching and Applied linguistics.

[7]Wardhaugh，R.1970.The Contrastive Analysis Hypothesis[A].In J.H.Schumann，&Sterson，N.（eds.）.New Frontiers in SLA[C].Rowley，Mass.：Newbury House.

[8]蔡龙权，戴炜栋.错误分类的整合[J].外语界，2001（4）：52-57.

作者简介：刘振卫，男，硕士，郑州轻工业学院外国语学院讲师。