This substantial volume contains 23 papers selected from the Fourth International Conference on Teaching and Language Corpora (TALC) held in Graz, Austria in the summer of 2000. Like many edited conference proceedings, the chapters in this book vary quite widely in both the quality and quantity of the work presented. Further, the substance of some of the stronger contributions—such as Coxhead on academic vocabulary and Lee on a genre-specific index for the British National Corpus (BNC)—had already appeared in journals at the time of this volume's publication (Coxhead in TESOL Quarterly and Lee in Language Learning and Technology). In his opening remarks, McEnery salutes the widening range of topics in this volume, but I am less sanguine about the value of adding breadth. Certainly, the new emphasis on the role of corpora in translation studies and the teaching of translation in the final section is well motivated, but it is less clear whether Renouf's discussion of short-dimension diachronic change or Flowerdew's use of language learning diaries has much to do with the central issues in TALC.

There are, however, a number of papers that deal with the design and use of smaller, more specialized corpora for specific groups of learners (see the papers by Aston, Bernardini, Gavioli, and Thompson). Of these, Gavioli offers the most penetrating account. She concludes that we need to distinguish more clearly between a corpus-building design based on descriptive needs and a design based on teaching needs. She notes that “comparing results from different types of corpora helps the students to see the limits to their generalizations more clearly” (pp. 300–301). Furthermore, in a particularly well-written retrospective account of the compilation of the large BNC, Burnard even concludes that this corpus is not really representative of the language as a whole but is au fond a collection of specialist corpora.

Another group of papers explores data-driven inductive or discovery learning, for which corpus linguistics is, of course, particularly well suited. The contributions by Bernardini and Johns are both instructive in this regard, but in this section, the most interesting and unusual paper is Mair's account of the role of corpora for advanced speakers of English in Germany. He argues that skillful use of corpora can foster nativelike language awareness, can reduce the mystical authority often attached to native speakers in German departments of English, and can reduce “possible feelings of insecurity and inferiority on the part of non-native speaking language professionals” (p. 124). In this way, Mair offers some fresh insights into the complex issues with which the language teaching professions are correctly concerned surrounding the role and status of nonnative language teachers.

Some of the more technical pieces, such as Wible, Chien, Kuo, and Wang and Berglund and Mason, will be of lesser interest to readers of SSLA. This might also apply to a number of the pieces that use corpus linguistics to discuss features of the English language. For example, Mindt claims to have produced a new corpus-based grammar for ELT or ESL; however, much of the detail in this chapter seems very reminiscent of Palmer's work, such as the treatment of catenative verbs. What might be new is the use of some frequency data that was not easily available in the 1960s and 1970s. For instance, Mindt shows that start to and start–ing are both common in all registers, whereas begin –ing is much rarer than begin to. (Findings that I easily confirmed in the Michigan Corpus of Academic Spoken English and in Hyland's corpus of 240 research articles.)

As I intimated at the outset, this is an uneven volume, although it is well edited and includes an index. Indeed, I suspect that even the more substantial papers might not be of great interest to many readers of this journal. Corpus linguistics is more of a methodology than a subfield, and it is unfortunate that few papers in this volume attempt to cross over to other areas of applied linguistics. Two notable exceptions are the papers on lexical acquisition by Coxhead and Lenko-Szymanska, the latter of which I would single out for its methodological approach. Finally, I would recommend Burnard for his honesty, Mair for his sensitivity, and Gavioli for her insights.