A Corpus-based Study On Lexical Bundles In Research Abstracts

As an essential component of languages, lexical bundles are considered as important language material in both spoken and written forms of fluent linguistic production. To acquire the use of the lexical bundles in their professional field is rather important for the researchers.With the development of Corpus Linguistics, lexical bundles have become a hot focus in the linguistic research. In recent years, more and more attention has been paid to compare the use of lexical bundles in different languages. However, the studies comparing the use of lexical bundles between English and Chinese are rather rare. Based on academic corpus, the present study compares the use of four-word lexical bundles frequently used in academic abstracts to find the similarities and differences between the two languages.Using language processing softwares to analyze the academic corpus on transportation, the present study compares the use of English and Chinese lexical bundles from the perspectives of structure and function, respectively. The results indicate that lexical bundles are frequently used in academic research abstracts on transportation. According to the defining standard for lexical bundles in the present study,90types of English lexical bundles are found, while there are only63types in the Chinese versions. But the Chi-square test does not show any significant differences. For the distribution on structural level, both of English and Chinese have more phrasal bundles. Prepositional phrases are much more in English. Adverbial bundles are identified in the present English corpus, while there are no such bundles in the present Chinese corpus. Proper noun bundles associated with transportation can be found in both of the languages. For the distribution on functional level, most of the bundles perform as referential frames, and Chinese has more types of bundles performing as discourse organizers. Passive voice is used in English to avoid subjectivity, while it is achieved by omitting the subjectives in Chinese. Chi-square tests show the stuctural and functional distributions in the two languages are statistically different.