A Corpus-based Study On The Developmental Patterns Of MAKE In Terms Of Colligation Used By Chinese English Majors

This study was conducted to explore the developmental patterns of MAKE in terms of five categories of colligations (v+n/pron, v+part, v+n+prep, v+n/pron+c, v+adj+that/of) used by Chinese English majors (CEMs) in their argumentative writing. The theoretical foundations employed in this study involve Contrastive Interlanguage Analysis (CIA) and L1 transfer.The data employed in this study come from two corpora: Written English Corpus of Chinese Learners (WECCL) as a learner corpus, and The Louvain Corpus of Native English Essays (LOCNESS) as a reference corpus. The learner corpus in this study is made up of three sub-corpora which are all argumentative writings by CEMs from Grade One to Three. The reference corpus involves argumentative writings by American university students. Three control variables: genre, grade and length were taken into account when the data were selected. The data analysis consists of five steps: tagging the selected data, extracting the concordances, retrieving the data, categorizing the data and comparing them with Chi-square test and Z-score.The major findings yielded from the study are as follows:The first finding is that quantitatively, the pattern in terms of frequency and percentage in v+n/pron colligation in English majors'argumentative writings decreases gradually with their increased proficiency, but qualitatively, they tend to use this colligation properly compared with native speakers. This shows that the advanced learners become more competent in using this colligation despite the lower frequency.The second finding is that CEMs have a preference or inclination to overuse some L1-related colligations like"make sb. become/feel/get…"in v+n/pron+c colligation which takes up more than 50% across three grades. However, they tend to employ smaller number of this colligation as they progress to higher grade. This indicates learners'L1 has a negative transfer to their acquisition of colligations but they are inclined to reduce their dependence on Chinese way of thinking as their language proficiency improves.The third finding is that though both learners in GD1 and native speakers use v+adj+that/of colligation less, it does not mean that they are equally proficient in this aspect. The former most probably have an incomplete knowledge of the target language and the latter have more options to resort to when they try to express themselves. In addition, learners in GD2 and GD3 are more inclined to use"that-clause"compared with native speakers in this colligation to demonstrate their mastery of complex syntactical structure.The last finding is that the acquisition of colligation is influenced by many factors, such as topic effect, L1 transfer, the difference between L1 and L2, learners'awareness or consciousness of the colligation, the complexity of the colligation itself, test anxiety and so on. Therefore, it is these factors working together that lead to the result of the colligations in this study.The findings in this study offer some implications for the teaching and learning of English for Chinese English majors:Firstly, English teachers can employ concordancing software to improve the effectiveness of using colligation.Secondly, special attention should be paid to L1 interference in teaching and learning colligation knowledge.Finally, both teachers and students should be more aware of some complex colligations which have no equal equivalents in their mother tongue.The present study has its own limitations and requires improvement in furture related studies.Firstly, all the data in this study are selected from argumentative writings within several topics. However, the number of compositions under the same topic in each grade is not the same. This might produce the topic effect. To some extent, it would affect the objectiveness of the results. Furture research needs a larger corpus to provide sufficient writings with the same topic so as to reduce the topic effect.Secondly, the categorization of the colligations of MAKE is inevitably subjective, which may to some extent have an influence on the objectiveness of the findings. Lastly, due to the limited time and the size of corpora, v+n/pron+c colligation in this study is conducted as a whole rather than respectively. Future related studies can investigate each complement in detail to deepen and broaden the study.