Summary of findings

Overall speaking, the humanities writers use the Elaboration connectives significantly more than the engineering and social sciences writers. The Elaboration connectives denote appositive relations, either expositing or exemplifying the propositions. The other function is clarification, linking the arguments by means of particularizing (e.g. 其中 qízhōng ‘in particular’), summarizing (e.g. 總而言之 zǒngéryánzhī ‘to sum up’), and verifying (e.g. 其實 qíshí ‘in fact’). The significantly more use of Elaboration connectives by the humanities writers suggests their greater concern about how to reinterpret and clarify the knowledge content in domains which deal with abstract ideas and philosophical theorization, and thus the writers need to exploit more connectives to bridge the logico-semantic connection within an abstract in which the disciplinary knowledge is presented holistically and reiteratively. On the other hand, the scientific knowledge is cumulative, relying on commonly assumed methods and leaving not much room for individual interpretation, and therefore the engineering writers exploit less linguistic resources to clarify the logical linking

121 among propositions.

The other finding is that writers in all disciplines use the Enhancement the most frequently, followed by the Extension and in turn the Elaboration. This finding is partially supported by Liu’s (2001) analysis of Chinese cohesive conjunctives in five written genres, suggesting that the more use of Enhancement is not attributed to the disciplines or genres but to the needs in Chinese writing. Moreover, the consistent pattern (Enhancement > Extension > Elaboration) across disciplines may result from Halliday’s classification, where the Enhancement involves more subcategories than the other two categories. The following sub-sections summarize the disciplinary variations within each category.

Subcategories of Elaboration

There is an overall significant difference across disciplines in the Clarification subcategory, with the humanities writers using the most connectives and the

engineering writers using the least. This can be justified by the nature of disciplinary knowledge and the style of disciplinary writing. In the humanities abstracts, the writers’ line of reasoning is more reiterative and the writing style is river-like, thus the humanities writers need to exploit more connectives to agglutinate the comparatively implicit and loosely-structured logical relations among arguments. In the engineering abstracts, on the other hand, the writers depend on the use of Particularizing 其中 qízhōng ‘in particular’ to a great extent and do not use the Summative and Verifactive connectives at all. This can be explained by the fact that the engineering writers concern the investigation of physical environment and often need to specify the experimental conditions under scrutiny. With the help of 其中 qízhōng ‘in

particular’, writers build a connection between the general goal of the study and the experimental specifications.

122

The other observation is that the humanities writers use only 其實 qíshí

‘actually’ but not 事實上 shìshíshàng ‘in fact’ in the Verifactive subcategory.

However, the social sciences writers use more 事實上 shìshíshàng ‘in fact’. In the humanities examples, writers use 其實 qíshí ‘actually’ to counterclaim and show disagreement with the previous arguments. In the social sciences examples, 事實上 shìshíshàng ‘in fact’ denotes writers’ assertion and certainty of arguments. The subjectivity expressed by the social sciences writers suggests that they prefer to argue with ‘mental fact’ in mind (Wang et al. 2011). On the other hand, the humanities writers prefer to argue with ‘addressee-oriented’ 其實 qíshí ‘actually’, showing a rhetorical interaction closer to the spoken register and a greater involvement of the readers in the line of argumentation (Chang et al. 2012).

Subcategories of Extension

The across-discipline variation shows that the humanities writers use the

Variation connectives (e.g. 反之 fǎnzhī ‘instead’, 除了 chúle ‘except’) significantly more than the engineering writers. With a closer look into the subcategories of

Variation, the humanities writers use the most Replacive connective among the three disciplines, while the engineering writers do not use the Replacive (e.g. 而 ér

‘instead’) at all. This can be explained by the semantic function of the Replacive. The Replacive connectives usually link a counterclaim with the preceding arguments, and the counterclaim conveys a tone of refusal or denial to the established facts in the domain knowledge, therefore the use of Replacive connectives is less preferred by the engineering writers who seek to have a common ground with the previously

established knowledge in accordance with the scientific conventions.

The within-discipline variation shows that writers in all disciplines use the Addition connectives (e.g. 此外 cǐwài ‘in addition’) significantly more than the

123

Variation (e.g. 反之 fǎnzhī ‘instead’). This can be accounted for by the three reasons.

First of all, the additive relations are more straightforward and direct in conjoining arguments, thus the Addition connectives are used commonly by all writers with a large quantity. Second, the Addition connectives express a ‘positive’ tone to the connection among arguments, while the Variation connectives convey a rather

‘negative’ prosody, which is less preferred in the academic argumentation. Third, in the subcategories of Variation, all of the Alternative 或者 huòzhě ‘or’ denote

propositional ‘or’ meaning in the databank, thus being excluded in the metadiscoursal use of connectives.

Subcategories of Enhancement

Writers do not use the Enhancement with an across-discipline statistical difference. However, the three disciplines present a similar pattern in terms of the subcategories of Enhancement, viz. the Causal-conditional > Manner > Spatio-temporal > Matter. In the engineering abstracts, in particular, the number of Causal-conditional connectives (e.g. 因此 yīncǐ ‘therefore’) outweighs that of other three subcategories statistically.

In the Spatio-temporal subcategory (e.g. 首先 shǒuxiān ‘at first’, 其次 qícì

‘next’), writers do not have disciplinary differences in statistics, suggesting that the use of Spatio-temporal connectives is a universal feature of academic writing.

In the Manner subcategory (e.g. 相較於 xiāngjiàoyú ‘compared with’, 透過 tòuguò ‘by means of’), writers in the three disciplines have near frequencies in the Means (e.g. 透過 tòuguò ‘through’, 藉由 jièyóu ‘by means of’) and use the Means much more frequently than the Comparison (e.g. 相較於 xiāngjiàoyú ‘compared with’), suggesting that the elucidation on the methodological move is obligatory in the dissertation abstracts, while giving a comparison between two points of view is

124

relatively optional (Bitchener 2010, Koltay 2010). The other observation is that the engineering writers do not use the negative Comparison connectives (e.g. 不同於 bùtóngyú ‘different from’). This can be accounted for by the conventions of scientific studies in which the engineering writers seek for common grounds with the previous literature, while prefer not to single out the differences with the established

viewpoints.

In the Causal-conditional subcategory, the engineering writers use the most Conditional connectives (e.g. 於…條件下 yú…tiáojiàn xià ‘under…circumstances’) among the three disciplines, while the humanities writers do not use at all. This is not surprising because the specifications of experimental conditions are important in the hard-applied disciplines. In the humanities abstracts, however, writers prefer to argue in a holistic and integrated way, thus there is fewer opportunity to state discretely the circumstances under which the findings are generated. The other observation is that the engineering writers use more Purpose connectives (e.g. 為了 wèile ‘in order to’) than writers in the other two disciplines. This is due to the purposive nature of the disciplinary knowledge in which the engineering writers concern the scientific exactness and the actual practices (Becher 1987, 1994).

The last subcategory is the Matter (e.g. 在…方面 zài…fāngmiàn

‘in…respect’). Writers do not use the Matter connectives statistically differently across disciplines. Most of the Matter connectives are spatial metaphors in origin.

They serve cohesive functions when referenced to a particular aspect of the arguments, helping the readers know to which aspect the argumentation pertains.