This thesis aims to contribute to the design and implementation of genre-based academic writing courses intended for learners of English in Taiwan by (a) examining the ways in which Taiwanese speakers of Chinese as a first language (both experienced and novice writers) typically structure texts, in both Chinese and English, in response to prompts intended to elicit one of four discourse modes (recount, argument, explanation and classification) and (b) comparing the structuring of their texts with (i) prototypes relating to each of the four discourse modes that are based on a sample of published academic articles, and (ii) texts written in English by experienced and novice writers who are speakers of English as a first language. In response to one of four prompts, each intended to elicit a different discourse mode, experienced and novice writers - Taiwanese citizens whose first language is Chinese - were asked to write texts of between 200 and 250 words. They produced 240 texts, of which 120 were written in Chinese and 120 in English. Analysis of these texts indicated that, in terms of both overall rhetorical structure (often referred to as ‘schematic structure’) and internal discourse structure (the occurrence and co-occurrence of particular discourse relations), the texts written in both Chinese and English by the experienced writers were (a) very similar to one another, and (b) also very similar to texts written in English by experienced writers for whom English is a first language. Although the structuring of the texts written by the novice writers was less consistent overall, the texts written in Chinese by these writers were structurally closer than were the texts written in English to (a) the texts written in both English and Chinese by the experienced writers, and (b) texts written in English by experienced writers of English for whom English is a first language. Furthermore, analysis of 60 texts (20 written in Chinese and 20 written in English by expert/ experienced Taiwanese writers and 20 written in English by New Zealand teachers for whom English was a first language) revealed a marked tendency, in all cases, towards simple linear or complex linear textual development, with cyclic development being evident in only one of the 60 texts. There is a widespread belief that experienced writers whose first language is Chinese and experienced writers whose first language is English tend to structure texts in very different ways. So far as the writers involved in this study are concerned, this would appear not to be the case. It may therefore be that the difficulty that learners of English in Taiwan often experience in relation to the structuring of texts in English has less to do with their cultural and linguistic heritage than it has to do with the complexities that are inevitably associated with attempts to structure texts in a language in which one has limited proficiency. If this is the case, it has implications for the ways in which genre-centered writing programs intended for learners of English in Taiwan are organized and implemented.