A corpus-based study on author stance in academic English

Stance refers to the lexical and grammatical expression of attitudes, feelings, judgments, or commitment concerning the propositional content of a message (Biber and Finegan, 1989). It is the way academics annotate their texts to comment on the possible accuracy or creditability of a claim, the extent they want to commit themselves to it, or the attitude they want to convey to an entity, a proposition or the reader (Hyland, 2005). According to Biber (2006), it can be conveyed through value-laden word choice (love, hate, and etc.), paralinguistic devices (e.g. gestures) and grammatical stance devices (modal and semi-modals, stance adverbials and stance complement clauses). The present study was intended to cover the epistemic stance devices used in conveying author stance in Academic English. It investigated whether there is a statistically significant difference across the doctoral dissertations authored in English by native, Spanish-speaking and Turkish-speaking candidates with respect to the use of these devices. Being corpus-based in design, the study included the results elicited through the Contrastive Interlanguage Analysis (Granger, 1996) of a total number of 136 doctoral dissertations written by native and non-native speakers of English (Native: 45; Spanish: 43 and Turkish: 48). Following Biber (2006), epistemic stance devices found in the dissertations were identified through Wordsmith Tools (Scott, 2001) and examined in four groups (epistemic adjectives, epistemic adverbs, epistemic nouns and epistemic verbs). Subsequently, these devices were subcategorized into two groups as certainty and likelihood devices in each category and their frequencies were separately measured for each corpus. Finally, Log Likelihood Test was administered to see whether there is a statistically significant difference across the groups as regards the use of these devices. Findings of the study have indicated that certainty devices were used more frequently than likelihood devices in three corpora and that the difference between the two categories in non-native corpora is higher than the one measured in the native corpus, which leads us to the conclusion that native academic authors are more cautious with their arguments and predictions while non-native academic authors are relatively more confident in their writing. Keywords: Author stance, epistemic stance device, interlanguage, contrastive interlanguage analysis.