A Study Of Use Of Discourse Markers In Narrative And Non-narrative Discourse By Chinese EFL Learners

This study seeks to describe the use of discourse markers (DMs) by Chinese EFL learners in the narrative and non-narrative discourse identified in their monologic production. The main questions addressed are: (1) What are the major types of use of DMs by Chinese EFL learners that distinguish narrative from non-narrative discourse in their monologic production? (2) What are the differences between the learners of different levels in the use of each type of DMs?The materials employed in the study were retrieved from SWECCL, a Chinese learner corpus, from which 30 cases were selected. On the basis of the scores that the test-takers obtained in the TEM Band-4 Oral Test, the cases were divided into higher- and lower-level groups. The conceptual framework of narrative and non-narrative distinction was based on Labov and Waletzky, Bruner and Yu. The classification of DMs was based on Halliday and Hasan, and Schiffrin. The major findings generated in the data analysis can be summarized as follows:Firstly, five types of DMs were identified that characterize narrative and non-narrative discourse in the learners'monologues, namely additive, adversative, temporal, causal and filler-type, among which temporal, additive and adversative were the frequently used DMs to signal the start of narrative talk (89%, 76%, and 75% respectively), and temporal was most frequently used by the learners to introduce their narration, implying that they were cognitively shaped in story schema or story grammar. The relatively high frequency of use of additive and adversative suggests that the learners were aware of using DMs as a discourse strategy to organize their narrative talk.Secondly, while four types of DMs were used in varying degree to introduce the non-narrative part in the monologues (11%, 24%, 25% and 32% respectively in the sequence of temporal, additive, adversative, causal), filler-type was used most frequently to signal the non-narrative part (60%). The less use of DMs to introduce the non-narrative talk is obviously related to the topic effect that the testing task itself produces in that the test-takers were asked to tell a story about the personal past experiences.Thirdly, the independent-samples t-test analysis reveals that there is significant difference between the groups of different levels in the use of temporal and filler-type in the narratives, and in the use of adversative and causal in the non-narratives, which suggests that higher-level learners are more capable of producing the well-organized story structure (or story map). Such explanation could be further supported by the type/token ratio analysis conducted in this study, which indicates that the higher-levels tended to use more varieties of non-narrative DMs. The type/token ratio analysis also shows that the lower-level test-takers used more varieties of narrative DMs, suggesting that they tended to use DMs as a communicative strategy or a strategy of veiling their deficiency in oral proficiency.The findings yielded in this study may have two pedagogical implications. In the first place, the learners'use of DMs may potentially become an effective means employed in L2 oral assessment; however, it does not mean that the high frequency or more varieties of DMs may have a predicting power to the learners'oral proficiency. In the second place, L2 teachers should use oral composition, and story telling in particular, as a way of raising the learners'awareness of using DMs to develop their discourse competence rather than using them merely as a communicative strategy.