Corpus-based studies of academic writing have shown that "there is a shared scientific voice or 'phraseological accent' (Gledhill, 2000: 204) which leads academic writing to polarise around a number of genre-specific recurrent word combinations and collocations (cf. Cowie, 1997; Howarth, 1998; Biber et al., 1999; Curado Fuentes, 2001; Pecman, 2004). Lexical verbs are an essential component of academic discourse: they are used "to discuss matters lying at the very heart of the scholarly process" (Meyer, 1997: 368) and enable writers to modulate their ideas and position their work in relation to other members of the discipline. However, they create a minefield of difficulties for non-native writers (e.g. Hinkel, 2002), who have to deal with the fact that each lexical verb has its own preferred lexico-grammatical company in academic discourse, viz. subjects ('this study shows that'; 'the evidence suggests that'; 'these results suggest that'), objects ('SUPPORT the view / hypothesis that …', 'PROVIDE evidence / information') and adverbs ('DIFFER significantly'; 'VARY considerably / widely'; 'APPLY equally'; 'closely related'; 'widely used'; 'generally accepted') and tend to appear in routinized structures ('as discussed in'; 'there is (no, some, little) evidence that', 'it should be noted that') (cf. Granger & Paquot, 2008). Monolingual learners' dictionaries (MLDs) have recently taken "more proactive steps to help learners negotiate known areas of difficulty" (Rundell, 1999: 47), to the point that they are now conceived as comprehensive writing tools. They now include productively oriented information in areas such as syntactic behaviour, synonymy, register preferences and prevention of errors (cf. Gilquin et al., 2007: 325). With the advent of electronic MLDs, space is no longer an issue and more attention has been paid to word combinations. Lexical entries often include collocation boxes or phrase banks. These new features, however, are usually based on huge general English corpora including various genres (e.g. journalese, fiction). In this paper, I investigate the usefulness of the phraseological information available in electronic MLDs for academic writing. The study focuses on the phraseology of a set of verbs that is particularly prominent in academic writing, i.e. verbs of 'evidence' (e.g. 'illustrate', 'suggest', 'support') as categorized in FrameNet. Using the Sketch Engine (Kilgarriff et al., 2004), I conduct a co-occurrence analysis of these verbs in the 'Corpus of Academic Journal Articles', i.e. a 90 million word corpus of research articles published in peer-reviewed journals (cf. Kosem, 2010). The results are compared with the collocations listed for these verbs in the CD-ROMs of the 'Big Five' (CALD, COBUILD, LDOCE, MEDAL and OALD) to assess the coverage of academic collocations. Preliminary findings indicate that electronic MLDs could do much better. A number of word combinations that have essential discourse functions in academic writing are missing from the 'Big Five'. More importantly perhaps, academic collocations are listed along with other word combinations that are not typical of academic writing. Non-native writers can be seriously misled by this presentation of collocations as they are not given any help to decide which collocations are most appropriate in academic writing. Put differently, the treatment of phraseology in electronic MLDs makes non-native writers believe that all collocations and phrases are good for all purposes (e.g. writing a research article or a short story, writing a letter to a friend or to human resources manager). These findings call for a genre-based approach to the lexicographical treatment of collocations and other word combinations.
Paquot, M. (2011). Towards a genre-based approach to the lexicographical treatment of phraseology in electronic monolingual learner’s dictionaries. eLexicography in the 21st century: New applications for new users, Bled (Slovenia). https://hdl.handle.net/2078.5/226394