This study takes a step towards answering two questions concerning L2 writing: 'Why do the Japanese write English expository essays differently in terms of discourse structure from their native English counterparts?' and 'What are the differences?' The study compares discourse structures of Japanese and English school texts from 30 Japanese junior high school pupils (14-15 years), who wrote school compositions on the occasion of their graduation, and 30 British secondary school pupils (14-16 years), who wrote GCSE coursework essays for Religious Education and English. Other written materials were collected from a Japanese school, and classrooms at a British school were observed. There are three research questions: whether the school texts are different in terms of topical progression (consisting of parallel, extended parallel, sequential and extended sequential progressions), what implications the results have in terms of genre and what implications the results have for Contrastive Rhetoric (CR) research. A review of CR literature suggests a need for further studies in this area, particularly for the exploration of the cultural and educational contexts, where preferred L1 discourse structures are likely to be taught and practised. It also raises an issue of genre, questioning whether expository writing, which has been the yardstick genre of CR research, is universally practised. Another issue is the search for a methodological framework for text analysis which can relate textual features to 'rhetorical organisation'. The apparently universal discourse structure of a meaningful utterance, topic - comment, and the central role of information structure in the construction of coherence, suggest Topical Structure Analysis (TSA) as the framework. The application of TSA to Japanese is discussed and found to be appropriate. The examination of the materials from the Japanese school and the observation of English and Religious Education classes in the British school reveal that expository writing is not practised in the Japanese school; instead, writing is used for self-reflection and self-growth. In the British school, expository writing is the major means of learning and assessment. The writing used in the Japanese school is characterised by parallel progression while English expository writing is likely to contain many topical progressions due to its clear focus on a limited number of topics and its logical nature. The TSA results show substantial differences of occurrence of topical progressions: the English texts contain more topical progressions, suggesting that they rely more than the Japanese texts on repetition of important items in conveying a sense of coherence. Parallel progression is the only type of progression that occurs more in the Japanese texts than the English texts. Examination of the results reveals the roles of topical progressions: parallel progressions for accumulation of information/descriptions, sequential progressions for development of argument, extended parallel and extended sequential progressions for repetition of important items throughout the text. The differences in the mean occurrences of topical progressions are found to be due to the difference of genre or function between the two data sets: the Japanese texts are expressive and the English texts transactional. The use of parallel progressions is associated with description within the expressive texts and the use of the other progression types is associated with the argumentative nature of English school essays. The potential of the topical marker was in Japanese for the construction of coherence is discussed. By providing a clearer coding scheme, this study contributes to TSA studies and CR research in general. The findings support a number of earlier CR findings. Although the limited sample size prevents generalisation, the findings hint at an association between lack of experience in argumentatiy writing among Japanese secondary pupils and the L2 writing performance of Japanese students at English speaking universities. Analysis based on TSA, while unable to detect every feature of coherence in text, has the potential to provide useful data for L2 writing classrooms and for future CR research relating textual features to coherence construction.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:634432 |
| Date | January 2004 |
| Creators | Tadaki, Toru |
| Publisher | University of Manchester |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0022 seconds