A Corpus-based Study On English Non-clausal Coordination

With its large-scale machine-readable corpora, the corpus-based approach receives increasing and influential recognition among language researchers by the fact that computers are in a position to make fast, accurate and complex analysis. As a new research method in modern linguistics, corpus-based approach has been more and more widely employed in language researches. This new method in modern linguistics observes the "naturalness" of data from genuine texts to provide better illustration of language features and thus facilitate language teaching and learning.Therefore, this thesis intends to probe into the non-clausal coordination by adopting this new approach. Through the study of the form, meaning and use of several coordinators, efforts are made to solve the riddles of the distinctive patterns of non-clausal coordination, providing a practical model on the basis of real texts for foreign language teaching and learning. Concentrating on the classification based on parts of speech of the relevant linking units, investigations in meanings and uses of coordinators such as and, or and but are conducted. Special attentions are paid to the order of the linking elements connected by coordinators, and discussions are focused on whether the linking units are reversible with the assistance of volumes of genuine texts. From such a corpus-based view, this thesis aims to make both language teachers and learners aware of the characteristics of coordination which are seemingly known but not well mastered by a majority of them, in the hope of contributing to English learners'fluency and correctness.