Genre Analysis: Discussion Sections Of English Medical Research Articles Written By Chinese Medical Writers

The teaching of academic writing has become more and more demanding for language teachers. Boosted by efforts to improve the efficiency of the teaching of academic writing, ESP genre analysis has attracted a great deal of interest from discourse analysts and language teachers for it contributes to a better understanding of what actually constitutes the structure and linguistic features of a certain genre such as research articles (RAs) and to a clearer idea about the effects of other factors such as writers' language-using environment on these features. ESP genre analysis on RAs can be divided into two groups: the studies on the macrostructure of RAs as a whole or the microstructure of separate sections of RAs, and the studies on particular linguistic features of RAs such as hedging, modality, verb tense, voice and the first personal pronoun. Quite a number of studies have been conducted on the structure and linguistic features of RAs in natural and social sciences, including the field of medicine. However, few studies have been reported on the genre analysis of English medical research articles (MRAs) written by Chinese, let alone the discussion section, which is known as the most troublesome and mostheavily hedged section in an MRA.The present study was designed to explore the generic structure and some linguistic features of MRA discussion sections written by Chinese medical writers (CMWs) by analyzing and comparing the generic structures and distribution of hedging devices used in MRA discussions written by Chinese medical writers or native English speaking medical writers (NESMWs) for any possible differences between the discussion sections written by CMWs and those by NESMWs.A pretest of generic structure analysis was first conducted in 4 MRA discussions, 2 written by CMWs and 2 by NESMWs with an aim to attesting the applicability of ESP genre analysis to the present study. Then two corpora of MRA discussions were established, with one corpus (CCMW) of 50 MRA discussion sections written by CMWs and the other corpus (CNESMW) of 20 written by NESMWs. All the medical journals containing the MRA discussions under study were chosen by 2 criteria — representativity and accessibility. All the sample MRAs, randomly chosen by a two-round selection from the latest issues of the selected journals had a macrostructure of "Introduction-Methods-Results-Discussion". The MRA discussion sections in the two corpora were first analyzed by ESP genre analysis, employing the 5-step procedures suggested by Nwogu (1997) with moves and steps as the analytical units. The distribution of lexical hedging devices of five categories in the discussion sections was also investigated by Varttala's method (1999). Chi-Square Test, Fisher Exact Test and Wilcoxon Rank Sum Test were used to compare the analytical results of the MRA discussions in the two corpora for any possible significant differencesin their generic structure and hedging devices.The major findings of our study were as follows: 1) In the 50 MRA discussion sections written by CMWs, besides the two moves in Nwogu's model for MRA discussion (1997) (Explaining Specific Research Outcome 50; 100% and Stating Research Conclusions 31; 62%, respectively), other three moves [Background Information (38; 50%), Restating Research Objectives or Methods (25; 50%), Reference to Previous Studies for Support (35; 70%)] were found to be frequently used by CMWs. In the 20 MRA discussion sections written by NESMWs, Explaining Specific Research Outcome (20; 100%) and Stating Research Conclusions (10; 50%), the two moves in Nwogu's model, were found to be frequently used. The difference between the two corpora was significant (P<0.05). 2) The two moves of Explaining Specific Research Outcome and Stating Research Conclusions were found to be frequently realized by several steps in the two corpora with some minor differences between them. 3) Some of the moves and steps identified in 32 (64%) and 15(75%) discussion sections in CCMW and CNESMW respectively were found to be cyclic, which accorded with the findings of previous researchers in other disciplines (Holmes 1997 & 2001). 4) The occurrence frequency of the hedges of the five categories in the discussion sections written by CMWs, like that in the discussion sections written by NESMWs, decreased in the order of modal auxiliary (45.17% and 45.23%), main verbs (31.29% and 21.16%), adverbs (11.84% and 16.67%), adjectives (8.80% and 14.55%) and nouns (1.90% and 2.39%), which is roughly consistent with the findings of Varttala (1999). Except for the main verbs used as hedging, no significant difference wasfound in the overall occurrence of lexical hedges between the two corpora.From the results above, we can conclude that: 1) The MRA discussion sections written by CMWs have a relatively set pattern with 5 obligatory moves and 5 obligatory steps. Though the discussion sections in both CCMW and CNESMW share some of the obligatory moves and steps in Nwogu's model for MRAs, some other obligatory moves, much less frequently used in the discussion sections written by NESMWs but present in the RA discussion sections in other disciplines , are frequently used in the discussion sections written by CMWs. These results, as suggested by Taylor and Chen (1991), indicate an impact of the writers' cultural or ideological factors on their choices of moves or steps. 2) Some moves and steps in MRA discussions are presented in cycles, which have also been reported in other disciplines such as history, political science, sociology, and agriculture by Holmes (1997 & 2001). The Chinese medical writers, like their native English speaking counterparts, tend to present their discussion sections in cyclic patterns, for this cyclic patterning is needed due to the complex nature of medicine and quite a number of unexpected outcomes to be explained in an MRA. 3) As for the use of hedging, the Chinese medical writers are generally as proficient as the native English speaking medical writers in employing lexical hedging devices of the five categories. The Chinese medical writers tend to use more main verbs as hedging, which is probably influenced by their first language, as the use of verbs is predominant in Chinese — their first language.Though the present research has identified the Move-Step model for...