| The human society is full of power relations. There exist various kinds of power relations among people and everyone is a complexity of power relations. The creation of power and its maintenance or change occur in and through discourses. Two kinds of discourses are used in this process: primary discourse and metadiscourse. The selection of linguistic devices is based on the power relations between the participants, therefore, the choice of metadiscourse can reflect their power relations in the discourse and how the powers are distributed.This thesis explores, from the perspective of critical discourse analysis, the relationship between metadiscourse and the construction of power relations through analysing the Country Reports on Human Rights Practices 2002(China Part), with the aims of revealing what, how, and why metadiscourse constructs the power relations in the report.The thesis starts with the introduction to the subjects of discussion, in which, the relationship between power and language and illustrations of metadiscourse will be reviewed roughly. The writer of this thesis will give certain introductions to the data to be analyzed in the following parts. Then the writer reviews the related literature to power, language and metadiscourse. A variety of researches on power, language and metadiscourse will be presented. The theories of Gramsci, Foucault, Fairclough, NG and Bradac and Hyland, which illumine the finishing of this thesis, will be introduced. After that, the description to the research process and methodology will be given. The research focus and how the data will be analyzed are shown. This thesis adopts a combination of quantitative analysis and qualitative analysis. And qualitative analysis is utilized more since the analysis is not based on a corpus. In analysis and discussion part, the writer of this thesis analyzes Country Reports on Human Rights Practices 2002 (China Part) in great details. The analysis reveals that in the report metadiscourse is frequently used in the construction of power relations. Specifically, the research is designed to explore the following questions: 1. What specific metadiscourse construct the power relations in discourse. 2. How metadiscourse construct the power relations in discourse. 3. Why metadiscourse can be used to build the power relations. Finally, the writer will argue about the implications and limitations of the study. |
