A Corpus-Based Study Of Collocations Of High-Frequency Verb-Make

The concept of collocation has historical roots. Concentrated effort on this discipline, however, did not really begin until after the 1950's when Firth, leader of the London School, first proposed the term"collocation."Since Firth's initial identification,"collocation"has often been used indiscriminately and with various interpretations and definitions. The broadest definition encompasses everything from the inner structure of an individual lexical item to the organizational properties of a text. In contrast, the narrowest is only interested in those word combinations that are neither free nor fixed. Due to the powerful influence of the Firthian linguistic tradition, collocation studies in western countries are generally focused on restricted combinations. This thesis endeavors to probe into all types of collocation. Its primary focus is to explore the principles that underlie collocational differences between native speakers of English and Chinese learners of English by observing, describing and explaining their respective linguistic behavior, so as to offer some useful suggestions for references in foreign language instructions.On the basis of a comprehensive review of the current theoretical background, this work presents a case study on the differences between native speakers and non-native speakers concerning the high-frequency verb"make". A systematic contrastive analysis is conducted on real-life linguistic data collected from British National Corpus (BNC) and Chinese Learners English Corpus (CLEC). Considering that superficial well-formedness itself is not a guarantee of freedom from errors, manual qualitative classification has been utilized in addition to computer-based quantitative statistics to ensure the adequacy of our work which would otherwise have been inadequate.Discussion is presented in three steps. First, data collection and classification. Second, contrast and analysis of the data. Third, findings and explanations of the differences and similarities in the data. Primarily 500 hundred sentences with the collocations of"make"are collected respectively from BNC and CLEC, which are classified into free collocations, restricted collocations and fixed collocations. On this basis, further sub-classifications are made to render contrastive analysis possible according to the grammatical relations and conceptual relations shown in the collocations. Corpus evidence based on contrastive studies indicates that significant differences exist between native speakers of English and Chinese learners of English in the use of"make"in terms of choice of collocates and their frequencies. Data analysis covers such syntactic structures as make + noun, make + prep + noun, make + noun / pronoun + noun, make + noun / pronoun + verb, make + noun / pronoun + adjective, make + adjective, make + verb and make + noun. The focus of attention is, however, laid on the"make + noun"construction. The study concludes that L2 learning is affected by native language, target language and learning strategies as well.The significance of this thesis lies in its practical implications to language teaching in general and foreign language teaching in particular. It suggests that even the upper-intermediate and advanced learners are still influenced by their native language in learning English, which shows that learner's awareness of language difference should be enhanced. It is also revealed that learners attach more attention to the conceptual meaning and consequently have a better command of free collocations but score a relatively lower degree of mastery of restricted collocations and fixed collocations. Therefore, learners need to distinguish between general rules and sub-rules and give more attention to prefabs in L2 learning. Finally, learners are expected to improve their L2 learning by optimizing their learning strategies.The inadequacy of the research is, admittedly, manifested in its restriction to one frequent verb"make", the small samples of which may possibly be partial. It would be more convincing and explanatory to English learners, if larger data and computational tools and statistics were used.