Corpus-based Study Of Collocations And Bilingual Learner's Dictionary

The single most important task facing language learners is acquiring a sufficiently large vocabulary. It is now recognized that much of the vocabulary consists of prefabricated chunks of different kinds, the single most important of which is collocation, which provides a basis for variable and interesting lexical structure of a language system and plays an important role in language learning. Self-evidently then, collocation should be a top priority in the research of vocabulary, language learning/teaching as well as dictionary making.Among various dictionaries, bilingual learner's dictionary is more generally employed in the initial stages of learning a language. To language learners, the knowledge of how to make correct collocations between words is as important as the knowledge of word meanings. Correspondingly, a high-quality bilingual learner's dictionary should not only give word meanings and explanations, but also include comprehensive information about collocation, the presentation of which is closely linked to various aspects in dictionary designing. This is undoubtedly a large-scale work, the efficiency and accuracy of which can be guaranteed largely through a corpus method, because a computer corpus enables lexicographers to examine large text corpora in a systematic manner and allows them access to a large quantity of evidence that has not been available before. Only computer corpus method perfectly co-ordinates the resources of man, machine and data, which will lead to new postulates in the research in collocation and bilingual dictionary making.Two points that need to be called attention to in this thesis are: (1) Computer data collection does not rule out lexicographer's intuition and subjective judgment. (2) However comprehensive the collocation information is in a dictionary, a dictionary can never, for the limitation of its space, enumerate all the illustrative examples as a corpus does. Therefore, the computer corpus is not only a tool for dictionary making, but also a good complement to the dictionary.