A Corpus-based Study On Lexical Chunk Errors In Chinese English Majors’ Spoken English

Since 1970s, lexical chunks have been getting increasingly popular among researchers for their values in promoting second language acquisition. Many linguists performed enormous theoretical study on definitions, classification, and function and laid theoretical basis for further lexical chunk studies. In recent years, some researchers have adopted lexical chunk in their classroom teaching and proved the importance of lexical approach in second language acquisition.Along with the development of cognitive linguistics, researchers found that errors are necessary and exist in all stages of second language learning. Lexical chunks are no exception. Based on previous researches on lexical chunks, this study applies error analysis theory to performing error analysis on lexical chunks used by English majors in their spoken English. The purpose is to find out the main types of lexical chunk errors in English majors’ spoken English and discover the causes for these chunk errors. Then, some pedagogical implications are made.This study applies corpus research method by selecting 100 pieces of transcripted texts from TEM-8 oral test in 2007 in SWECCL 2.0. All lexical chunks in these texts are identified and labeled out according to Nattinger and DeCarrico’classification. In order to make such identifying and labeling more accurate, two postgraduates who had performed relevant studies on lexical chunks are invited to check all lexical chunks in the whole samples. Then, by applying BNC and two dictionaries, all misused lexical chunks are identified and picked out. All these chunk errors picked out are analyzed from lexical and grammatical perspective.The result turned out that, on the whole, among the four categories of lexical chunks, phrasal constraints have the largest number of errors which are far more than polywords and sentence builders. Few errors occur in institutionalized expression. From the lexical perspective, overrepresentations are most frequent and then follow redundances and additions. In each category of the lexical chunks, the frequency of each type of errors is similar. Among all lexical errors, the most frequently misused is preposition. From the grammatical perspective, the most frequent errors are singular and plural forms used in polywords. In phrasal constraints, the error frequencies of the use of singular and plural form, tense, grammatical voice and gerund are similar. Sentence builder errors are mainly in the use of singular and plural forms and gerund. On the whole, students make more lexical errors than grammatical errors.Based on the analysis of chunk errors in all samples and by referring to the actual situation of current English majors’spoken English learning, this study concludes five causes for their chunk errors:interlingual transfer, intralingual transfer, communication strategies, carelessness and limitation of time in speaking.Concerning these above causes of chunk errors, some effective teaching and learning strategies are proposed. Firstly, in order to reduce the interference of native language and overgeneralization, the input should be enhanced by using such methods as intensive and extensive reading and corpus. Secondly, teachers should improve their teaching methods and enhance students’practice in speaking English. Lastly, correct attitudes should be held on lexical chunk errors for both students and teachers. Error correction can be applied in classroom to help students to realize and understand their errors so as to avoid repetition on previous errors.