A Corpus-based Study Of College Nonenglish Major Students’ Anaphora Proficiency In Writing

In the long period of the study of anaphora, many scholars have done lots ofresearches, both theoretically and empirically. However, most of their works focusedon theoretical analysis and English-Chinese comparative study. The definition ofanaphora is that anaphora is a linguistic use where an expression is implied to referback to an entity or item indicated by another expression within the same text ordiscourse. From this definition, the conception of anaphora actually contains variouscohesive relations including reference, substitution, and ellipsis in discourse asdiscussed in Halliday and Hasan (1976). Judging the linguistic embodiment ofanaphora, anaphora is composed of two components: the anaphor and its antecedent.The interpretation of anaphora is of great significance to the understanding of Englishdiscourse on account of the fact that, in natural discourse, the conciseness, coherenceof the discourse are mainly realized through anaphoric relations. The literature reviewis summarized in this paper from formal syntactic approach, discourse functionalapproach and pure cognitive approach. The empirical study about anaphora of FL(Foreign Language) learners is rarely seen, particularly in China. It is well known thatChinese college students often make mistakes when using anaphora, which can bespotted very often. But the reasons why students make such kind of mistakes have notbeen illustrated systematically. The present research aims to examine Chinese collegenon-English majors’ proficiency of using anaphora, as well as to explore the reasonsthat cause the mistakes of anaphora, in both qualitative and quantitative ways.The main body of this paper, the corpus data analysis and teaching experiment,aims to study anaphora from input and output aspects. On one hand, in term of output,Chinese Learners’ English Corpus (CLEC) is used as the tool of analyzing anaphoraphenomena in the composition written by Chinese English learners, in which anaphoricmistakes frequently occurred. CLEC can roughly represent English level of Chinesestudents. It has a frequency list, and is in error tagged text version. According to thefrequency list, these anaphoric mistakes are classified and the percentage of eachmistake is computed. The errors which most frequently take place should be paid attention to, and this thesis intends to explore the causes which lead to students’ suchkind of mistakes and seek to explain the error patterns. According to the data collected,we can conclude that the anaphoric errors can be divided into types. The present studyexposes some types of anaphoric phenomena existing in Chinese non-English majors’use of it, including misuse of reference, confusion of antecedent, and disagreement.Besides, through analyzing the results of corpus study, the reasons why students makeerrors of anaphora are generalized, such as negative transfer of Chinese, overgeneration and so on.On the other hand, in term of input, the paper does researches on the teachingmaterial used by the non-English major college students. Another corpus, Cecorpus, ischosen to make collection and statistics more convenient. The corpus is designed withintent to help both teachers and students use the contained course books with acorpus-based approach to vocabulary development. In this study, one course book-NewCentury College English is chosen because the subjects in the teaching experiment usethis one as their textbook. Through the study of the teaching material, we can checkwhether college students learn the anaphoric uses to the full or not. The author searchesthe same word in the hints of Cecorpus, and finds out some results. In the erroranalysis, one of the reasons why students make anaphoric errors is that: the distance ofthe antecedent and the anaphor increases the reader’s reading span and also influencesthe readers’ understanding to the anaphors. The study also shows that there are fewconditions in which some other nouns appear before the anaphora and make it difficultto identify the correct antecedent in the teaching material. Instead, students often makelong sentences in which more nouns confused with the anaphor are included in theirwriting.According to the corpus study, some pedagogical implications and suggestions areput forward. In order to check whether the pedagogical suggestions put forward by theauthor can work effectively in the daily teaching and learning, and the corpus CLEC isalso out of date in some way, a teaching experiment is designed. This part is performedas a supplement in the whole paper, not playing the leading role. With the statisticsfrom the teaching experiment analyzed by SPSS software package, the data ofcomparative samples test proves that there are significant differences between thepretest and posttest in the experiment class. That is to say, the teaching suggestions put forward by the author are feasible. The present study also put forward some limitationsand recommendation for the future anaphoric study.