A Corpus-based Study Of Near-synonyms Rather And Quite

Synonym discrimination is always a challenge in English learning. Accurately graspthe subtle difference of synonyms can help the language learners correctly understand andproperly use the synonyms, making the communication smooth. As a new perspective oflexicology research, semantic prosody is an effective method to discriminate synonyms.The traditional way of semantic prosody research rely more on the intuition of researchers,while the emergence of modern corpus linguistics provides a new approach for semanticprosody research. Since the corpus can provide a large amount of authentic language use,by analyzing the collocations of words, researchers can detect the subtle semantic prosodicdifference of synonyms, making synonyms discrimination possible.The present study takes rather and quite as examples to investigate the ChineseEnglish majors’ semantic prosodic behavior. The data are collected from the corpusWECCL and COCA. By using the software AntConc3.2.1w and the retrieval softwarebuilt in COCA, the data of rather and quite are retrieved, and then the data are classified,compared and analyzed. This study aims at finding the semantic prosodic features of ratherand quite by Chinese English majors and the possible reasons of the outcome.The data show that the Chinese English majors and the native speakers demonstratesimilar negative semantic prosody of rather. However, Chinese English majors tend tooveruse the negative semantic prosody of quite, while underuse the positive prosody of it.From the perspective of collocational behavior, the Chinese English majors have notgrasped the significant collocates of rather and quite. They have produced some unusualcollocates of the two node words. The possible reasons for the improper use of semanticprosody and collocations lie in the inadequate knowledge of semantic prosody andcollocation of both Chinese teachers of English and learners of English. The negativelanguage transfer especially the Chinese equivalents of English words is another reason.Therefore, a focus on semantic prosodic information and typical collocates should be on the agenda.This research sheds some light on vocabulary learning and teaching. Much moreattention should be paid to the semantic prosodic features and typical collocations of words.What is more, corpus-based approach should be widely applied to foreign languagelearning and teaching，as corpus can provide adequate authentic language use.