All 12 chapters, the preface, and list of publications were specially prepared for this volume and relate to the work of Granger and colleagues, based at the Centre for English Corpus Linguistics (CECL) of the Université Catholique de Louvain. The chapters cover a wide range of language-acquisition issues, mainly from a corpus linguistics perspective.
The most practical chapters directly associate corpus findings with language-learning contexts. By combining collocations and concgrams—that is, not necessarily contiguous lexical bundles—that can include fine or general part-of-speech labels as placeholders, Wible and Tsao present a corpus-based tool called StringNet (Wible & Tsao, Reference Wible and Tsao2012) that they claim can directly assist learners by presenting them with emergent networks of meaning-based lexical bundles. For instance, the semantically similar spend time [Vvg] and take time [to V] contrast grammatically. These patterns can be presented in a dynamic network, highlighting common realizations such as the 12 typical nouns for the pattern leaving/setting aside the [Noun]. Neff van Aerstelaer and Bunce integrate corpus findings from the Spanish section of CECL’s International Corpus of Learner English (ICLE) into a pedagogic program to enable learners to satisfy can-dostatements for stance taking in academic writing. From pre- and posttreatment measures, students show improvements in the range and quantity of stance-taking verbs, adverbs, and lexical devices. Schmitt and Redwood, by comparing learners’ knowledge of phrasal and multipart verbs with their frequency scores, conclude that only 20% of the variance in test results for receptive and 9% for productive knowledge can be attributed to frequency in general corpora, suggesting how pedagogic programs can be better informed by corpus studies.
Many of the chapters provide proposals for language acquisition theory from the perspective of corpus linguistics. The ICLE is employed by Hasselgård and Johansson to quantify a range of language learner errors across four languages, and, thus, to propose a methodology for contrastive interlanguage analysis. Tribble’s contrastive analysis of a range of corpora—representing learner language, target or exemplar varieties, corpora similar (i.e., analogue) to exemplar corpora, and general English—focuses on n-grams. An examination of bundles of three or more regularly co-occurring words produces a wide range of productive language for general and discipline-specific academic language, most of which is common to all corpora but rarely included in language-learning curricula.
A few chapters appear to have little relation to language acquisition. For instance, Mair studies varieties of cyber-Jamaican using a Web-based corpus methodology to detect linguistic features that develop community values. Although this topic is of interest to corpus linguistics and sociolinguistics in general, it is unlikely to offer much of value to language-learning contexts. Rayson and Baron describe a computational tool designed to correct spelling errors in learner corpora. Their Variance Detector (VARD2) identifies about 90% of all spelling errors but, even with manual intervention in its training process, barely exceeds a 20% correction rate and, so, may add little value to learner corpus comparison.
Most of the remaining chapters are directed more toward research than pedagogic applications. Mauranen questions the standard practice of comparing learner corpora with native speaker norms, recommending instead that English as lingua franca (ELF) is a more appropriate target for the majority of language learners who may interact only with other foreign language users. Providing examples in which communicatively effective ELF differs from native speaker norms does not, however, address the question of the status accorded to language varieties. In most ELF contexts, it is likely that standard varieties still carry greater status. Rundell and Kilgariff review the increasing use of computers in lexicography and, as exemplified by the compilation of the Macmillan English Dictionary for Advanced Learners, explore the possibility of fully automating the process of dictionary building—a project that is still a number of steps from completion.
All chapters exhibit a high standard of innovation and scholarship—only Tribble’s chapter requires noticeable further proofing—and address significant issues for corpus linguistics and language acquisition, with the best contributions offering new methods for optimally identifying target language, redefining and refining the role of lexis and grammar in acquisition, and providing valuable data for improved curricula. There are clearly many directions that the research presented here can take, but the majority of these directions were inspired and enabled by Granger’s foresight, diligence, and determination in building ICLE and other corpora designed to advance language acquisition research and practice.
