A Corpus-based Study Of The Differentiation Of English Synonyms

Corpus-based approach, a new research method in modern linguistics, has been more and more widely employed in language researches of various fields and perspectives since the flourishing of computer technology and the appearing of lots of large-scale and well-designed corpora and concordance software. The present study attempts to conduct a case study of the differentiation of English synonyms on the basis of the British National Corpus (BNC). The present study is to differentiate 12 pairs of synonyms, chosen according to personal studies and teaching experience, and with the help of WordNet as well. The present study will conduct corpus inspection, and analyze the concordance lines of these synonyms in BNC, from the perspectives of register, collocation, colligation, and semantic prosody, aiming to discover the distinctive features of the two words in each pair, and to expose the effectiveness of corpus-based method of distinguishing synonyms. The present study attempts to bring about some innovations in vocabulary teaching approaches, that is, to offer both language teachers and learners a brand-new method of synonym differentiation, and meanwhile, introducing corpora and concordance tools into classroom teaching will promote students to conduct data-driven learning (DDL), thus, this method can effectively stimulate the motivations of students, and further improve their ability of analyzing and solving problems.The present paper consists of five chapters. Chapter 1 is Introduction, which presents the history of corpus linguistics and its development in China as well as the achievements of Chinese scholars, and this chapter also illustrates the purpose and significance of the present study. Chapter 2 is Literature Review, which introduces some relevant theories and studies concerning the present study, including the role of corpus in the present study, theory of lexical grammar, concept of synonyms, register distribution and illustration of the studies of colligation, collocation and semantic prosody both at home and abroad. Chapter 3 is Research Methodology, which firstly illustrates the characteristics of The British National Corpus (BNC) employed in the present study, then explains the method of choosing the subjects of the present study--the 12 pairs of synonyms, and finally introduces the research procedure and approach of register distribution, colligation, collocation and semantic prosody respectively. Chapter 4 is the body of the present paper, which displays the quantitative results of the corpus concordancing, including precise data and randomly selected concordance lines. In this chapter, a qualitative analysis of these results is made as well. Chapter 5 draws a conclusion that corpus-based approach is quite effective in distinguishing synonyms with its unique advantages.Corpus-based approach is not a replacement, but an indispensable supplement to the traditional method of differentiating synonyms through dictionaries and reference books. Corpus-based method provides both qualitative and quantitative analysis, which produces more reliable and convincing results. Synonym differentiation in this way surely is of significant pedagogical implications.