The two volumes of Creating and digitizing language corpora, edited by Joan Beal, Karen Corrigan, and Hermann Moisl, provide a survey of current corpus projects under the category of what the editors term “unconventional” corpora: corpus-based projects that exist at the interface between sociolinguistics and corpus linguistics. As the editors describe it, a major aim of their project “is to establish whether or not annotation standards and guidelines of the kind already employed in the creation of more conventional corpora on standard spoken and written Englishes … should be extended to less conventional corpora” (vols. 1 & 2, p. 1). The editors point out that their “companion volumes are unique” (2) because they do not focus on “conventional” corpora – corpora that provide data on standard language varieties using models that do not highlight social and linguistic variation within the data. While a corpus-based focus on language variation is increasingly found in the literature, no other unified source includes the wide range of sociolinguistically relevant corpus projects included here.

The first volume, Synchronic databases, begins with a chapter by the editors providing both an overview of the eight contributed chapters and a synthesis of how those chapters relate to the broader project of the two-volume set. Most of the contributed chapters discuss specific projects on language varieties: The Scottish Corpus of Texts and Speech (SCOTS; chapter 2), the Freiburg English Dialect Corpus (FRED; chapter 3), the Syntactic Atlas of the Dutch Dialects (SAND; chapter 4), the International Corpus of English–Ireland (ICE-Ireland; chapter 6), and the Corpus of Written British Creole and the Corpus of Written Jamaican Creole (CWBC and CWJC; chapter 8). Two chapters focus on broader initiatives: The Language Interaction Data Exchange System for multilingual data (LIDES; chapter 5), and the TalkBank Project, which seeks to support data-sharing (chapter 7). The first volume ends with a chapter by Sali Tagliamonte discussing her experiences with and recommendations for developing corpora from collections of speech recordings.

The second volume, Diachronic databases, also begins with an introductory and synthesizing chapter by the editors. The remaining nine chapters discuss individual projects, often focusing in particular on issues that arise in dealing with historical and textual data: The Newcastle Electronic Corpus of Tyneside English (NECTE; chapter 2), the Network of Eighteenth-Century English Texts (NEET; chapter 3), the Origins of New Zealand English Project (ONZE, chapter 4), a text-based Corpus of Irish English (chapter 5), the Corpus of Scottish Correspondence (CSC, chapter 6), the Corpus of Early English Correspondence (CEEC, chapter 7), a comparative translation project involving five historical Chinese texts (chapter 8), the York-Toronto-Helsinki Parsed Corpus of Old English Prose (YCOE, chapter 9), and the Corpus of Late Eighteenth-Century Prose (CLEP, chapter 10).

Additionally, the volumes feature an excellent foreword by Shana Poplack, whose 1989 paper “The care and handling of a mega-corpus” (in Language Change and Variation, ed. by Ralph Fasold & Deborah Schiffrin; John Benjamins) was foundational in the burgeoning field of “unconventional” corpus linguistics.