Grammatical variation in British English dialects: A study in corpus-based dialectometry by Benedikt Szmrecsanyi (review)

This book analyzes the relationships between a range of dialects of British English using an innovative approach—the dialectometrical analysis of morphosyntactic variation in a corpus of recorded speech. Prior to Szmrecsanyi’s research, dialectometry—the simultaneous mathematical analysis of many geographically different linguistic varieties across a wide range of linguistic features—had largely been directed at lexical and phonetic variation (see Maguire & McMahon 2011, though see Spruit 2006 for a notable exception). S’s research is the first large-scale dialectometrical investigation into morphosyntactic variation and, to add to its novelty and rigor, it uses as its database an extensive corpus of naturalistic conversational speech. As such, it is a very welcome addition to our understanding of the relationships among British English dialects, to geographical morphosyntactic variation, and to our understanding of (the limits of) dialectometrical and quantitative linguistic methods more generally.

The book is divided into nine chapters. Ch. 1 introduces the rationale for the study and surveys previous analyses of relations among English dialects, both traditional analyses based on isoglosses and dialectrometrical analyses based on lexis and phonetics. Ch. 2 describes the database underlying S’s study, the Freiburg English dialect project and corpus (FRED; see Kortmann & Wagner 2005) and the basic principles and methods of corpus-based dialectometry. Ch. 3 describes the fifty-seven morphosyntactic features investigated, showing their geographical patterning and comparing this to previous accounts of these features. Ch. 4 goes beyond the analysis of individual features and examines aggregate morphosyntactic similarities and differences among all thirty-four dialects, and between them and Standard British and Standard American English. Ch. 5 investigates whether the relationships between varieties are gradient, and the degree to which the linguistic differences between them correlate with geographical distance. Ch. 6 takes this analysis further, exploring the extent to which dialect clusters are apparent in the data, using hierarchical agglomerative cluster analysis. Ch. 7 seeks to determine whether there are any correlations between the distributions of the different features, using principle component analysis. Chs. 8 and 9 summarize the findings of S’s analysis, including an assessment of the significance of the geographical dimension in morphosyntactic variation. These are followed by three appendices, giving summary statistics and—a major feature of the book—a series of twenty-nine color maps and four dendrograms illustrating the results of the analysis.

Grammatical variation in British English dialects is a clear and extremely informative account of dialectometrical methods. Chs. 2 and 4 in particular are an excellent introduction to dialectometry to anyone new to the field, while Chs. 5, 6, and 7 provide a very readable survey of advanced dialectometrical techniques, though the mathematical details of multidimensional scaling, hierarchical agglomerative cluster analysis, and principle component analysis are skirted over, which may be a good or bad thing, depending on your perspective. These chapters are really brought to life by the full color maps and dendrograms in Appendix C, which beautifully and clearly illustrate the results of the techniques used by S. In addition to the many grayscale figures in the rest of the book, they form a mini linguistic atlas of S’s data. The book is worth it for these aspects alone, and I would certainly recommend it to students as essential reading for an understanding of dialectometry. Ch. 3 is also a useful summary of key patterns of morphosyntactic variation in British English dialects, though, given the large number of features described, it cannot go into a great deal of depth; however, S very helpfully provides plenty of references to previous research for each of the features for the interested reader to pursue further.

Despite the many excellent aspects of this book, there are issues with the analysis and interpretation of the data, which means that the conclusions drawn by S raise as many questions as they answer. First, although S includes fifty-seven morphosyntactic features in his analysis in order...