writing and speaking, instructed in the tools of the trade (web- and PC-based concordancers)
and gradually inducted into the skills needed to best exploit the data and the tools for directed
rights reserved.
*
Corresponding author. Tel.: +81 561 732111.
E-mail address: david_lee00@hotmail.com (D. Lee).
www.elsevier.com/locate/esp
English for Specific Purposes 25 (2006) 56–75
ENGLISH FOR
SPECIFIC
PURPOSES0889-4906/$30.00
2005 The American University. Published by Elsevier Ltd. Alllearning as well as self-learning. After the induction period, participants began to compile two
additional written corpora: one of their own writing (term papers, dissertation drafts, unedited
journal drafts) and one of
expert writing, culled from electronic versions of published papers in
their own field or subfield. Students were thus able to make comparisons between their own
writing and those of more established writers in their field. At the end of the course, participants
presented reports of their discoveries with some discussion of how they felt their rhetorical con-
sciousness was raised and reflected on what further use they might be making of corpus linguis-
tics techniques in their future careers. This paper gives an overview of how this course was
structured, presents the kinds of discoursal and other linguistic phenomena examined and
the sometimes surprising observations made, and reports on the pluses and minuses of this cor-
pus-informed course as a whole, seen from the point of view of both learners and instructors.

2005 The American University. Published by Elsevier Ltd. All rights reserved.A corpus-based EAP course for NNS
doctoral students: Moving from available
specialized corpora to self-compiled corpora
David Lee
*
, John Swales
Faculty of Foreign Languages and Asian Studies, Nagoya University of Commerce and
Business, 4-4 Sagamine, Komenoki-cho, Nisshin-shi, 470 0193 Aichi-ken, Japan
Abstract
This paper presents a discussion of an experimental, innovative course in corpus-informed
EAP for doctoral students. Participants were given access to specialized corpora of academicdoi:10.1016/j.esp.2005.02.010

D. Lee, J. Swales / English for Specific Purposes 25 (2006) 56–75 571. Introduction
In many ways, the position of corpus linguistics, as a powerful methodology-tech-
nology, is well-established. It has been consolidating its position in lexicography
(e.g., the CoBuild dictionaries), in grammars of languages (e.g., Biber, Johansson,
Leech, Conrad, & Finegan, 1999, the CoBuild grammar series represented by Francis,
Hunston,&Manning, 1996, 1998;Hunston&Francis, 2000), and in diachronic studies
(e.g. Hickey, Kyto¨, Lancashire, & Rissanen, 1997; Kyto¨, Rissanen, & Wright, 1994;
Nevalainen & Kahlas-Tarkka, 1998; Rudanko, 2000). Further, the value of corpus-
based approaches for language specialists such as language majors, language teachers
and translators is becoming increasingly recognized (Bowker & Pearson, 2002; Cheng,
Warren, &Xu, 2003; Granger, Hung, & Petch-Tyson, 2002; Hunston, 2002; Kennedy,
1998; Partington, 1998; Sinclair, 2004). Mair (2002) has also interestingly argued that
the availability of appropriate corpora has been of great benefit to foreign language lec-
turers, such as Anglicists in Germany, because they are no longer beholden to the
whims of native-speakers for judgments of acceptability and correctness.
Corpus linguistics has also established itself as an important tool for determining
the linguistic features of registers and genres (Biber, 1988; Lee, 2000) and, in EAP,
for elucidating and comparing how different disciplines use language in their major
genres (Bernardini, 2002; Flowerdew, 1998; Gavioli, 2002; Ghadessy, Henry, &
Roseberry, 2001; Hyland, 2000, 2003; Luzon Marco, 2000; Thompson, 1998). How-
ever, it is less clear how – and when – these research findings can best be carried over
into effective pedagogical practice. There are a number of issues here. One resides in
the comparative merits of static, pre-prepared on-paper teaching products versus
more spontaneous on-line work in front of a computer. A textbook such as Thurstun
and Candlin (1997) shows what can be done with the former approach, but in their
companion article (Thurstun & Candlin, 1998), they do acknowledge the danger of
concordancing burnout: ‘‘Over-exposure to concordance lines can conceivably tire
students if teaching of this type depends solely on deduction from concordance
lines’’ (p. 278). This in turn raises a second issue: Should concordance work be sup-
plementary to more traditional EAP instruction, or can it be made central with tra-
ditional instructor explication and exemplification reduced to the periphery?
A third factor to be considered is the level of disciplinary acculturation of the tar-
get participants. If those participants are just beginning to attempt the professional
genres of their chosen specialization, then they probably need considerable help at
the macro or structural level (such as with audience analysis and/or with the organi-
zation of the paper). This in turn suggests that a wholesale commitment to a corpus-
based approach may not be fully effective, since concordancing tends to work better
at the lexico-grammatical and phraseological levels rather than the structural level.
On the other hand, if they already possess the appropriate genre knowledge, as is of-
ten the case with non-native speaker of English students completing their doctoral
degrees, then what they may be mostly missing is fine-tuning of lexical and syntactic
subtleties, particularly in terms of their strategic and rhetorical implications.
A fourth and final issue that we raise at this time concerns the degree of special-ization in the corpora to be made available or to be constructed. What might be the