An Analysis On Distribution Of Errors In Compositions Of Non-English Majors

Language is a complicated tool used for communication.Language learning is a slow and arduous process of trial-and-error testing,so learners will inevitably make various errors in the process of learning.Among them,writing error is a major sort.There are different understandings toward error.Traditionally,errors were regarded negatively and must be eradicated.But with the development of linguistics,applied linguistics, psychology and other relevant subjects,people's attitude toward error has gradually changed to be more scientific.Errors are no longer seen as 'unwanted forms',but as evidence of how learners are setting about the task of learning.The fact that learners do make errors and that these errors can be observed,analyzed,and classified to reveal something of the operation system within the learner,led to the study of learners' errors.This thesis,with CA,EA and Interlanguage(IL) as its theoretical guidance,analyzes the compositions written by non-English majors from Weifang Medical University in Shandong Province.The distributive features of writing errors in terms of language knowledge,linguistic categories and error sources are analyzed.At the level of language knowledge,the errors identified in the present paper are categorized into word errors, phrase errors,clause errors,sentence errors and discourse errors;concerning linguistic categories the errors are classified into phonological errors,lexical errors,syntactic errors, semantic errors and pragmatic errors.As to the error sources the paper focuses on intedingual errors and intralingual errors.The data analysis generated the following findings:First,intralingnal errors(57.4%) take higher percentage than interlingual errors (42.6%) though the overall ratios of them are close.The evidence of target language causes can be found in each sub-category of the three perspectives.Intralingual errors,especially errors in word(68.7%,288 occurrences) and in syntax,including clause errors(58.0%,375 occurrences) and sentence errors(53.2%,167 occurrences),take majority.That is to say,it is not the mother tongue Chinese influence,but intralingual errors caused by violation of English language rules,that mainly interfere with these English learners' language production in their compositions.Therefore,target language causes,intralingual errors, should be considered to play a decisive role in the selected sample compositions of the subjects.The result yielded in the present study confirms Taylor's(1975) assumption that advanced learners commit more intralingual errors than beginning learners.Second,errors related to meaning at each perspective still take a high percentage, such as phrase errors(24.1%,486 occurrences) at the scope of language knowledge and errors for semantic transfer(27.8%,239 occurrences) in terms of error source.Therefore, the different understanding of meaning between Chinese and English should be another vital factor,which influences the quality of these English learners' compositions.Third,from the analysis of the distribution of errors in terms of error source,we find that interlingual errors take a relative high percentage(42.6%,859 occurrences).Therefore, Chinglish is still a very serious problem among the samples.Even in some sub-categories, the frequency of interlingual errors is much higher than that of intralingual errors.Finally,these learners stay in the transitional stage in the continuum of IL development.They are at the transitional stage between elementary and intermediate level of the IL development continuum,on which L2 learners might have some knowledge of certain L2 rules,but sometimes fail to apply them in a proper way.At last,the proper attitudes,principles and techniques toward error correction and some pedagogical implications on language teaching,especially writing teaching are proposed.The study has the following pedagogical implications:reinforcing L2 learners' language knowledge;learning the target language thought patterns;employing discourse teaching method;creating more contexts for language performance;and receiving amounts of Standard English input.