Academic discourse: cross-cultural comparisons of the rhetorical structure of the research article

The significance and the relevance of the theories and methods of discourse analysis in language teaching and teacher training will be illustrated here through the example of academic discourse and within that the research article genre. The results of some of the cross-linguistic research conducted internationally and nationally motivated by Swales’s (1990) genre theory and Kaplan’s (1966) contrastive rhetorical analyses will be reviewed briefly to be able to highlight the practical implications these provide for the teaching of academic writing in the EFL/ESL context.

3.1. Cultural differences in academic writing

Swales’s and Kaplan’s ideas motivated abundant research in the field. Cross-cultural comparisons of academic writing reveal important structural and rhetorical differences across languages. Research on the Anglo-American conventions of scientific writing (e.g., Connor 1996; Hyland 1996, 1998, 1999;

Swales 1990) shows that writers tend to focus on form, the linear organization of thoughts, and the direct presentation of arguments. English academic discourse is characterized by strong writer responsibility in conveying meaning.

German academic writing (Clyne 1981, 1987) has been shown to place greater emphasis on the presentation of knowledge than on the form in which this know-ledge is conveyed. It is often characterized by digressions, linearity is considered less essential and writers expect the reader to play an active part in inferring meaning, i.e., reader responsibility in meaning making is important.

99 Discourse analysis and language teaching

Polish (Duszak 1994; Golebiowski 1999) and Czech (Čmejrková 1996;

Čmejrková and Daneš 1997) scientific writing conventions resemble the German tradition in that they lack strict linearity, focus on content, pay less attention to form and organization, tolerate digressions more, use metadis-course less frequently, and require reader responsibility.

Research on Oriental cultures, for instance on the Chinese, Korean, Japanese, Indonesian (Hinkel 1997; Chinese: Taylor and Chen 1991) traditions also demonstrates the dominance of reader responsible texts. Text analyses give account of so called “indirectness” strategies, as a result of which writers do not engage in a direct presentation of arguments.

3.2. EFL academic writing in the Hungarian context

In the Hungarian context, Árvay and Tankó (2004) investigated the move structure of theoretical RA introductions using Swales’s (1990) theory. Their analyses suggest that the structure of Hungarian theoretical RA introductions is different from the norms recorded by the Swalesian CARS model. They propose certain modifications to the model, such as the inclusion of additional steps: “Examples, which illustrate the topic” and “Analytical details of the present research”. Furthermore, they found considerable differences between the structural characteristics of the Hungarian and the English corpus:

the introductions of the English articles are longer than those of the (1) Hungarian articles;

English introductions are typically divided into paragraphs, while this is (2) not the norm in Hungarian introductions;

English introductions contain a larger number of rhetorical steps on the (3) whole than Hungarian introductions;

Hungarian introductions deviate more from the CARS model than (4) English introductions.

Motivated by the findings of the research presented above, Károly (2006) conduct-ed a study to describe the rhetorical move structure of the introductory sections of theses written by MA students in contrast to that of published research articles (RAs). It was found that the two corpora differ most in Move 3 (“Occupying the niche”). While the writers of the RA corpus place greater emphasis on “Outlining purposes” (Step 1A) and concentrate less on “Announcing present research” (Step 1B), thesis writers work equally with both steps. Furthermore, none of the thesis introductions announce the principal findings (Step 2), while some RA introduc-tions do. Step 3 (“Indicating research article structure”) distinguishes the two

100 Krisztina Károly

corpora greatly: it seems to be a “compulsory” component of the introductions in the Thesis corpus, whereas it does not typically occur in journal articles. As regards the proportions of moves, while in the RA corpus the dominant move is Move 1 (Establishing a niche), in the Thesis corpus Move 3 dominates (Occupying the niche). This suggests that while expert writers find it more important to justify the topic of their choice, students put more emphasis on what they are going to actually do in relation to the given subject.

The results of Károly’s (2006) study confirm the findings of previous research with regard to the sequencing of moves and steps (e.g., Árvay and Tankó 2004; Duszak 1994; Golebiowski 1999; Swales 1990). Move 1 is usually the opener in both corpora. Move 3/Step 3 (Indicating RA structure) closes the introductions of the Thesis corpus (with one exception only), but this step is atypical in the RA corpus. Move avoidance can only be witnessed in the Thesis corpus: in the case of Move 2/Step 1A (Counter-claiming) and Move 3/Step 2 (Announcing principal findings). The rest of the moves and steps occur in all sorts of sequences, no typical pattern may be identified.

Qualitative analyses of the data reveal that the Thesis corpus contains a number of steps that do not occur in the RA corpus and also fail to form part of the CARS model. In their analysis of Hungarian and English theoretical RAs, Árvay and Tankó (2004) also made note of the occurrence of such steps.

Such steps are as follows:

Clarifying terminology (in Árvay and Tankó’s (2004) and Károly’s (2006) (1) corpus, too),

Bringing examples (to illustrate the topic) (in Árvay and Tankó’s (2004) (2) and Károly’s (2006) corpus, too),

Stating motivation (for conducting the research) (in Károly’s (2006) (3) corpus),

Stating the research question(s) (in Károly’s (2006) corpus; this step (4) does not equal Swales’s (1990) “Question-raising” (Move 2/Step 1C) as

the latter involves more general types of questions),

Stating hypotheses, preliminary assumptions (in Károly’s (2006) (5) corpus),

Referring to research design (in Károly’s (2006) corpus).

(6)

Based on the results of the cross-cultural comparison of genres one may argue that there are important rhetorical/structural differences between languages/

cultures. Raising students’ and teachers’ awareness of these differences may considerably enhance the efficiency of learning and thus success in intercul-tural communication.

101 Discourse analysis and language teaching