| Research articles, as a sub-genre of academic discourse, aim to report writers' research findings. It is actually a persuasive endeavor involving interaction between writers and readers. To achieve the purpose of convincing, text writers inevitably have to adopt methodologies and rhetoric strategies to enhance the persuasiveness. Based on the view that writing represents the social interaction between writers and readers, metadiscourse focuses our attention on the ways how writers organize the propositional contents, project themselves into their work and engage the readers. Its use as a linguistic device to guide the readers through the text and display an appropriate persona is an important aspect of persuasive writing. Its role in academic discourse to maintain and enhance the contact between writers and readers makes it a crucial part of studies in this field. This thesis sets out to investigate the metadiscourse distribution and functions in English research articles.The framework of metadiscourse adopted in the present study is proposed by Ken Hyland. Metadiscourse is defined here as the linguistic resources used to organize a discourse or the writer's stance toward either its content or the reader. There are two kinds of metadiscourse resources: interactive resources and interactional resources. The former includes transitions, frame markers, endophoric markers, evidentials, and code glosses. The latter consists of hedges, boosters, attitude markers, engagement markers, and self-mentions.The corpus employed in the present research is mainly from the website elsevier, from which 20 English research articles are chosen to analyze how metadiscourse resources are used to organize propositional contents, convey writers' stance and engage readers. The research centers on three questions: 1) What are the overall features of metadiscourse distribution in the observed research articles? 2) What are the features of metadiscourse distribution in different parts of the observed research articles? 3) What are the features of metadiscourse distribution in both of the two major branches of academic community: natural sciences and social sciences? The author employs the software - Systemic Coder to make statistical analysis of the distribution of the metadiscourse resources in the research articles and further explores relevant and reasonable explanations.Some findings have been discovered in this research. First, the features of the global distribution are that interactive resources outnumber interactional resources and that transitions, hedges and code glosses rank top three among all the metadiscourse resources. Second, the feature of local distribution is that metadiscourse use varies across different parts of the texts. The abstract part sees only eight types of metadiscourse resources; the introduction part employs far more evidentials than others; the body part features more use of endophoric markers; and code glosses and self-mentions rank higher in the conclusion part than in the other parts of the texts. Third, the comparative study concerning the two domains of knowledge reveals a lot about the conventions and norms in disciplinary communities. Social science disciplines slightly outnumber natural science disciplines in overall metadiscourse use. Writers from the domain of natural sciences use more interactive resources and those from the domain of social sciences adopt more interactional resources. From the perspective of the 10 subcategories, more endophoric markers are identified in the research articles from natural sciences and more hedges, boosters, attitude markers, self-mentions from social sciences. The findings further confirm the view that the use of metadiscourse is highly affected by the rhetorical contexts and disciplinary contexts where it occurs.This thesis is composed of five chapters. Chapter One introduces the background information on the present study, indicates the aims of the research and presents the organization of this thesis. Chapter Two reviews the previous theoretical and empirical studies on metadiscourse and indicates the working definition and taxonomy of metadiscourse. Chapter Three introduces the methodology adopted in the present study. The research questions, data collection, research instrument, analytical approaches, and the statistical procedures are presented. Chapter Four deals with the results and discussions of the study. The entire analysis centers on the three research questions. Chapter Five contains summarization of the major finding emerging out of the present study, statements of the contributions and limitations of the study and suggestions for further research.
|
PDF Full Text Request