The treatment of quantity approximation in reference grammars and business English textbooks

The expression of quantity is one of the key elements in business discourse. Quantity can be expressed in a precise or imprecise/vague manner. This paper reports on some of the results of a large-scale corpus-driven investigation of quantity approximation in a variety of business genres in English, French and Dutch. The focus of this paper is on the linguistic devices used to express quantity approximation identified using a number of search methods in three corpora representing three different business English genres: the 1-million-word Business English News corpus (BENews, business news reporting genre), the 1-million-word business subcorpus of the Louvain Corpus of Research Articles1 (LOCRA_Business, a corpus of academic publications on a number of business topics), and a 500,000-word corpus of spoken business meetings (BNC_meetings, the ‘business meetings’ component of the British National Corpus). The methods used include using part-of-speech tagged versions of the corpora (tagged with the Tree Tagger, Schmid 1994) to automatically retrieve numbers from the corpora and identify the approximators that occur around them, (2) using semantically tagged versions of the corpora (UCREL Semantic Analysis System (USAS)) (see Rayson, 2003) and analysing the different tags potentially signalling quantity approximation (e.g. A13.4 for ‘degree: approximators’), (3) using WordSmith Tools’ keyword analysis (Scott 2008) to identify potential quantity approximation devices that are specific to the business genres under study, comparing the business corpora with each other but also with a 1-million-word ‘fiction’ reference corpus (the ‘Fiction Baby BNC’ corpus). These inductive search methods made it possible not only to uncover a wide variety of approximating devices but also to shed light on genre-specific approximators and on some approximators’ preferred collocational patterns. The main aim of the paper is to compare the results of the corpus-driven analysis with the treatment of quantity approximators in a number of corpus-informed general reference grammars of English (e.g. Collins Cobuild English Grammar or the Cambridge Grammar of English) and business English textbooks and reference books that target intermediate to advanced learners of English (e.g. Intelligent Business, The Business, ProFile, Grammar for Business). The analysis is both quantitative and qualitative as it examines, on the one hand, the amount of information and the various types of approximators included, and, on the other, the type of information displayed (e.g. style labels, collocational patterns) and where and the way in which it is presented and discussed (e.g. in a special section/unit devoted to quantity approximation). The paper also offers an evaluation of the relevance and usefulness of the approximators and related information featured in the publications under scrutiny in the light of the results of the corpus-driven analysis.