Patterns And Variations Of Metadiscourse Use In Chinese EFL Student Writing

Metadiscourse resources help writers to organize the discourse into a coherent text,guiding the readers through the text and facilitate their processing difficulties inunderstanding the discourse. In addition, metadiscourse resources can also aid writers toconvey their attitude in the text and engage readers into the discourse. Literature reviewreveals that previous studies on metadiscourse use in student writers’ written texts in theESL/EFL context focus primarily on published research articles, whereas less effort hasbeen put into the student learners’ assignment essays in language classrooms. In addition,the types of student assignment essays investigated so far are restricted to persuasive essays.Therefore, there is a need to probe into metadiscourse use in student essays of others modesof discourse (MDs) such as Narration and Exposition so that a more comprehensive pictureof metadiscourse use can be drawn, which, as a result, necessitates the present research.Based on the literature review on previous theoretical and empirical studies onmetadiscourse, this present research takes a corpus-based approach to explore the patternsand variations of metadiscourse use in narrative, argumentative and expository essayswritten by English majors at some Chinese universities. In addition, it also investigates thestructural distribution of metadiscourse resources in the introductions, body paragraphs andconclusions of student essays to gain further insights into the roles of metadiscourse in thedifferent parts of texts.Corpus analysis finds that the general pattern across the three MDs is that narrativeessays contain a significantly low frequency of metadiscourse resources thanargumentative and expository ones, which significantly differ from each other in theemployment of interactional resources but not in that of interactive devices. Morevariations can be observed with regard to the metadiscourse subcategories. Statistics showthat student essays across the three MDs exhibit significant differences in the use of thefollowing six subcategories: transitions, frame markers, hedges, attitude markers, selfmentions and engagement markers. Further variations emerge in the use of metadiscoursesubcategories when comparisons are made between every two types of essays.Argumentative and expository essays display similar trend in the use of metadiscoursefeatures and they only differ significantly in the use of attitude markers, self mentions and engagement markers. The expository essays display a surprisingly high frequency ofengagement markers than argumentative ones while argumentative essays show muchhigher frequencies of attitude markers and self mentions than expository ones. Narrativeessays differ significantly from both expository and argumentative ones in the use oftransitions, frame markers and hedges, although the three categories do not show anysignificant variations in the latter two MDs. In addition, narrative and expository essays donot exhibit significant differences in the frequencies of attitude markers and self mentions.Corpus analysis into the structural distribution of metadiscourse use within textsunveils that metadiscourse resources as a whole are used consistently most often in theconclusions and least in the body paragraphs in the three MDs. The distributions of attitudemarkers and engagement markers follow this pattern across the three MDs, while othersubcategories do not. In narrative essays, more instances of frame markers, hedges and selfmentions are identified in the introductions than in the other two parts, while thefrequencies of transitions, boosters, attitude markers, self mentions and engagementmarkers are much higher in the conclusions. Code glosses do not display any significantstructural differences in their distribution. In argumentative essays, hedges occur withhigher frequencies across the three parts of essays but their structural distributions do notdisplay any significant differences. The frequencies of other subcategories presentsignificant differences across the three parts, with endophoric markers, evidentials, codeglosses and self mentions occurring most often in the introductions and other fivesubcategories in the conclusions. In expository essays, the frequencies of frame markersacross the three parts are relatively high but they do not differ significantly. Transitions,code glosses, hedges and boosters are observed to occur more frequently in theintroductions, while endophoric markers, attitude markers, self mentions and engagementmarkers are found to appear most often in the conclusions. Furthermore, morestructure-dependent features are also found with regard to the distribution of certain typesof the subcategories. For instance, frame markers announcing discourse goals are mostfrequently used in the introductions across the three MDs while those labeling stages occurfar more often in the conclusions. Functional analyses reveal that some of the distributional patterns and variations inmetadiscourse use across the three MDs can be attributed to the writing purposes of eachtype of essays. The structural distribution of metadiscourse resources are closely related tothe textual functions of the three parts of essays. In addition, corpus analyses find that otherfactors such as essay topic wording and writer-reader relationship can also influence thewriters’ choice of metadiscourse resources.On the basis of the research results and discussions, the present research proposes amultidimensional contextual framework for understanding metadiscourse use in EFLstudent essays. It suggests that dimensions such as the essay task environment, learnerfactors, and writer-reader interactions are crucial in understanding metadiscourse use in EFLstudent essays.This research is original in three aspects. First, it expands the scope of empiricalstudies on metadiscourse use from persuasive essays to narrative, argumentative andexpository ones, which will consequently provide a broader view on metadiscourse use inwritten texts by EFL students and thus facilitate our understanding of the roles andfunctions of metadiscourse. In addition, examination on the structural distribution ofmetadiscourse resources within texts provides us another perspective to understand thestructural roles and functions that metadiscourse play in the text. Finally, amultidimensional contextual framework enables us to understand metadiscourse use from abroader vision so that a deeper understanding of metadiscourse use in EFL context can beachieved.