Abstract
The aim of this study is twofold. It intends to shed light on the differences between the language of published specialised journal articles and that of ELF writing and to provide a better understanding of the impact of literacy brokering activities.. Using comparable corpora of published and unrevised materials (the economics subcorpus of the Sci-ELF corpus, Helsinki), the sudy focused on comparing ELF production with two corpora of published articles (a general one and a small corpus of the same papers once published). The analysis shows that ELF writers have a clear preference for disguising their authorial voice by adopting ‘zero stance’ reporting techniques, and that frame markers are generally used to point to topic-related propositions rather than to argumentative moves. The use of attitude markers, however, reveals that the affective attitude of the author is noticeable, though mostly related to the propotypical patterns with important, stressing the importance of a point or guiding the reader’s perspective towards their desired communicative outcome. Last but not least, unedited research article drafts show a recurring pattern of indirect self-mentions instead of personal pronouns. ELF writers seem to prefer avoiding direct references to their authorial identity and maintaining a more neutral stance, based on prototypical language forms which may be more easily acceptable and understood in an international community (see also Bondi and Borelli 2018) in the logic of the “cooperative imperative”. Published articles, on the other hand, are characterized by a higher level of textual cohesion, as shown by the more frequent use of the consequential conjunction thus, presenting arguments as logical consequences of other propositions. Another interesting feature of published economic research articles involves the more frequent reference to visual representations within the body of the text, which might suggest a preference for more objective and data-supported claims, also confirmed by the use of metadiscursive verbs show and argue as boosters to emphasize the role of the authors and their contribution to knowledge production and dissemination. The most salient feature observed is the prevalence of a stronger and more present authorial voice. Published articles display a strikingly more frequent use of the first person, thus indicating a more intrusive authorial voice. Nevertheless, the increased use of boosters and personal self-mentions resources is compensated by hedging devices and evidential markers, which contribute to moderating the tone of the claims and recognising the value of existing literature. The comparison between unedited and published ELF, despite the small sample of published versions, allowed some further observations. Editors, proofreaders and other figures involved in the process of editing and publishing research articles seem to apply changes to the language of original drafts to convert what could be perceived as an impersonal presentation of data into a more ‘persuasive endeavour involving interaction between writers and readers’ (Hyland 2005). In particular, they emphasize the relevance of the research and its outcome, whereas they don’t seem to worry about other matters of style which may simply reflect cultural diversity and openness to diversity.