This study on contrastive rhetoric reports on metadiscourse functions in sociology articles in Persian and English. The results have revealed a higher number of metadiscourse elements in the English texts. Among the different metadiscourse elements used, text connectors are the most frequently employed in both languages. Modality markers are the second most frequent in both languages although the English writers used nearly twice the number of these markers. Overall, it is found that the

more »

... cy of textual metadiscourse markers is greater than the interpersonal markers in both language samples. It was further revealed that the Persian writers of sociology texts are less interested in explicitly orienting the readers and some of the main points in an article, especially in the concluding section, are left for the readers to infer. This, we believe, is the result of less reliance on academic writing in the educational system of the country. Instead, the Iranians are largely encouraged to employ a flowery language and rhetoric to decorate their writing in their school years which makes them less attentive of their readers. A contrastive rhetoric of metadiscourse function 537 Although this illustration is now often criticized for being too simplistic, for it is overgeneralizing and also because it assumes English rhetorical model to be "straight", Kaplan's hypothesis is still given consideration because it points out the nature of these rhetorical variations across cultures. Leki (1991: 123) notes that although the work of Kaplan is "exploratory" and to some extent "more intuitive than scientific", it is "valuable in establishing contrastive rhetoric as a new field of inquiry". Clyne (1981, 1984) studied two different writing tasks from English-speaking and German-speaking subjects. The studies provided clues for Clyne to conclude that German discourse is less linear than English. While English scholars pay more attention to formal discourse, their German counterparts pay more attention to the content of the discourse. These contrastive studies give us better insights into writing styles in different languages than do the ones by Kaplan (1966) and others that investigated the texts in a common second language because they compare the original texts written by professionals, not just student essays in a second language. Hinds (1987) , by evaluating English and Japanese texts, claims that while English uses a writer-responsible rhetoric (i.e. it is the duty of the writer to make his/her text clear to the reader), Japanese uses a reader-responsible rhetoric (i.e. it is the duty of the reader to understand what the writer has intended to say). In other words, a reader-or writer-responsible rhetorical tradition depends on the degree that the reader or the writer involved in making inferential activity in comprehending the text. Indrasutra (1988), cited in Noor (2001, investigated 60 essays written by American and Thai students and found out that Thai students focused on mental status more than American students did. She supports this by arguing that Thai students are more influenced by religious beliefs than American students. The essence of the concept of genre, as now used in applied linguistics, ESP and rhetoric, is an emphasis on the primacy of communicative purpose and the ways in which communicative needs shape or influence rhetorical structures (Holmes 1997: 322). On metadiscourse Due to the definitions presented for the genre, some researchers tried to classify different types of texts into different genres. Consequently, research articles (RAs) of academic writing were recognized as a genre by some researchers (that "the universal character of academic literature derives, in my view, from the fact that academic papers belong to the same genre". On the same ground, a research paper can be said to constitute a genre within the scientific world with different conventions across different disciplines. These conventions are, nevertheless, realized through rhetoric. From whatever perspective we look at genre and rhetoric, metadiscourse conventions are of prime importance in the organization of a text. Vande Kopple (1985: 83) believes that metadiscourse is the linguistic material which "does not add propositional