Christopher Gledhill

The discourse function of collocation in research article introductions

Abstract The increasing use of computer-held text corpora containing many millions of words has allowed linguists to establish lexico-grammatical patterns in language that were previously unavailable to observers. Such patterns range from lexical collocations and idioms to the phraseology of grammatical items. Recently, collocations of high frequency words in medical research abstracts and articles have been found to be useful indicators of the prototypical phraseology of the genre. In this article we characterize the phraseology of Introductions from a corpus of 150 cancer research articles. We explain the fixedness and idiosyncratic nature of scientific phraseology in terms of discourse processes such as reformulation. We argue for the design of a representative and specialized corpus of the research article and a contextual approach to corpus work that is appropriate to the teaching of languages for specific purposes (LSP) and the ethnographic aims of genre analysis in general.

A Lexicogrammar approach to checking quality: looking at one or two cases of comparative translation

In this paper I take a tried-and-tested methodology in linguistic analysis (the ‘‘lexicogrammar approach’’) and apply it to a particular problem of translation (a comparison of two equivalent phrases in an English translation of a French text). My purpose in doing this is to raise a number of research questions which I believe should be of importance to anyone in the translation business. My first question is very general: between two potentially equivalent translations, is it possible to identify which one is best? The assessment of any translation can often be highly subjective, but there appear to be some areas which are even more di‰cult to ascertain than others. In particular in this paper I examine the traditionally murky category of phraseology. However, I shall attempt to show here that it is possible to evaluate the phraseology of a particular translation in a scientific, almost forensic way, in particular by using corpus-based evidence. By ‘‘corpus-based’’, I am referring here to the use of computer-held electronic archives of texts, whether texts found on the internet or more specifically texts prepared for linguistic analysis by ‘‘tagging’’, or marking-up the corpus. In fact, it has now become the standard position of many empirical linguists (Sinclair 1991, Coulthard 1995, Hunston and Francis 2000, Tucker 2006) that no scientific statements about the linguistic features of a text can be based on introspection or gut-feeling alone, but should rather be supported by the meticulous observation and comparison of contextualised examples from a representative corpus of texts. To many, this might sound impractical and time-consuming, but the methodology of corpus linguistics has become fairly widely-accepted in the field of translation studies (Pearson 1996, Sinclair, Payne and Pérez Hernandez 1996, Bowker 1998, Xiao and Yue 2007). Furthermore, in this paper I show that it is feasible to conduct a systematic corpus-based analysis relatively quickly, especially if the 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40

Extracting Collocations in Contexts

The aim of this paper is to develop (i) a general framework for the analysis of verb-noun (VN) collocations in English and Romanian, and (ii) a system for the extraction of VN-collocations from large tagged and annotated corpora. We identify VN-collocations in two steps: (i) by calculation of the frequent lexical co-occurrences of each VN-pair, and (ii) the identification of the most typical lexico-grammatical constructions in which each VN-pair is involved in.

Learning a ‘Genre’ as Opposed to Learning ‘French’. What can Corpus Linguistics tell us?

At undergraduate level in the UK, learners of French are expected to produce a wider variety of genres and registers than they are prepared for by their entry qualifications. This causes particular difficulties that on the surface have little to do with formal knowledge of grammar and tend to be put down to ‘style’. We argue here that corpus linguistics and an empirical perspective to language development can bridge this gap and also unlock much wider issues, in particular the ideological perspective that underpins the language of a particular discourse community. While regular patterns that are not commonly included in expository writing syllabuses may prove to be useful ‘set phrases’ for the student in a particular assessment, we emphasize the textual role of idioms and the potential of generic phraseology in the language syllabus. On a broader cultural level, the grammatical features of expository writing in French are different to those of the general language, and reveal broader issues of the extent to which French education and the media engender and reproduce their own discourse structures.