A Corpus-based Interactional Metadiscourse Analysis Of Professional Movie Reviews In English

Interactional metadiscourse plays an important role in communicating writers andreaders through discourse medium, and in the reflection of interaction between writersand readers. Good movie reviews often attract people to the movie house. This thesistries to use Hyland’s interactional metadiscoure model to investigate professional moviereviews.The research method of this thesis mainly includes quantitative analysis based oncorpus data and statistics, through which the distributions and word frequencies ofinteractional metadiscourse markers are investigated. The process of the research can bedivided into three steps. First,80movie reviews from the Hollywood website publishedin2012were collected. Second, the movie reviews were then transferred into data filesand input into a52888-word corpus. And third, concordance lines containinginteractional metadiscourse markers were generated. Besides, the retrieved data wouldbe transferred into Micro Excel documents for calculating and all documents werepacked together according to their category, while those unfound metadiscoursemarkers would be omitted.The research results show totally2335interactional metadiscourse markers areused,4.4%of this corpus; and engagement markers are mostly used and their proportionis27.3%of interactional metadiscourse; hedges account for25.8%; boosters21.4%,attitude markers12.8%, and self-mention12.7%. Then this thesis exemplifies thosedistributions and explains their functions in communication. In addition, this thesis usesCorbett’s kind of diction to explore rhetorical features of interactional metadiscoursemarkers and the results show that most interactional metadiscourse markers are general,abstract and Latinate words.During the research this thesis finds the shortcoming of Hyland’s metadiscoursetheory, which overemphasizes the lexical meaning of interactional metadiscoursemarkers and ignores their pragmatic meaning. So this author strongly believes themeaning of metadiscourse markers should be explained from pragmatic level andsemantic level and no bias should be given to either of them.