Language in Space: The Cartographic Representation of Dialects is a thorough and fascinating text which makes useful reading for the student of dialectology. Consisting of a richly illustrated critical account of dialect mapping methods, and Wegmann's introduction of an alternative method, Language in Space represents a significant step forward in the field of dialect geography. Through tracing dialect-mapping methods from the latter half of the 19^th century to the early 21^st, and by applying them herself, Wegmann demonstrates problems with these methods. Further, she challenges the idea of the dialect boundary itself, using both diachronic and synchronic approaches to illustrate the subjectivity of the concept, and, ultimately, its unreliability.

Language in Space begins with an outline of the various sources of dialect data used by the dialect geographers whose work Wegmann scrutinizes in the study: The Survey of English Dialects (SED), Linguistic Atlas of England (LAE) and Computer Developed Linguistic Atlas of England (CLAE). Wegmann also discusses the concept of the isogloss, pointing out the different ways in which isoglosses have been defined and identified. Hinting at her conclusions, she agrees with Chambers (Reference Chambers and Preston1993) that ‘considering more than one item and/or feature inevitably reveals variability and consequently causes a boundary line to vanish’ (17); we need a new way of illustrating the distribution of dialect features.

Chapter 2 (‘A Diachronic Comparison of Maps and Methods’) gives a comprehensive history of English dialect geography, from Old and Middle English surveys to the present day, and from early attempts at dialect classification such as Ellis’ (Reference Ellis1871) and Prince Bonaparte's (Reference Bonaparte1875–Reference Bonaparte1876) to those of Wright (Reference Wright and Paul1891) and Trudgill (Reference Trudgill1990, Reference Trudgill1999) at either end of the 20^th century. Throughout, Wegmann illustrates problems with dialect mapping practices to date, showing that dialect maps will be subjective ‘as long as human beings play a part in dialect research’ (69). Dialect boundaries identified by researchers using the same material vary significantly according to factors such as choice of linguistic features, data interpretation and mapping technique. Examining in detail the aims and methods of each scholar, Wegmann also shows that any dialect map has an ‘individual aspect’ (68), and that it is relatively easy for boundaries to reflect and reproduce the expectations of the dialect geographer. Despite her criticisms of dialectological methods, Wegmann is careful to acknowledge that the dialect geographers she scrutinises were pioneers; although there are methodological inconsistencies and biases in each author's work, ‘they mark the beginnings of scientific dialect research’ (47) and thus provide the foundation for new research such as her own.

Chapter 3 (‘Mapping dialects: A Synchronic Perspective’) is concerned with describing and evaluating the qualitative methods of traditional dialectology as established by Kurath (Reference Kurath1972) and carried forward by numerous scholars including Wakelin (Reference Wakelin1972), Fischer (Reference Fischer1976), Trudgill (Reference Trudgill1990, Reference Trudgill1999), Glauser (Reference Glauser1991) and others. The author modifies Kurath's method and applies it to data from the SED, comparing her own results with Kontic's (Reference Kontic1990) and Trudgill's (Reference Trudgill1990, Reference Trudgill1999). Again, marked differences in the results expose the subjectivity of traditional dialectology and that ‘the course of dialect boundaries depends mainly on the material and methods chosen to establish them’ (1).

As noted by the author, ‘criticism requires a counter-proposal’ (1). Wegmann's answer lies in the methods of the Salzburg School of Dialectometry, which ‘responds to the characteristic problems of traditional dialectology’ (135) laid out earlier in her work by combining geolinguistics with numerical classification. A similarity-distance matrix allows the researcher to identify linguistic proximities and structural patterns in large data sets, and to visualize these in their many aspects. In Chapter 4 (‘Mapping Dialects: A Quantitative Approach’) Wegmann first explains visual dialectometry and then applies it to the same East and South East English SED data used in Chapter 3. The results allow her to draw fascinating comparisons with the studies and migration patterns surveyed in earlier chapters, such as the dialect variations commonly identified between former Anglo-Saxon kingdoms.

At times, Language in Space can be opaque and difficult to access; in order properly to follow Wegmann's arguments the reader must compare maps and tables across multiple pages and chapters, some of which are not optimally laid out. In the first half of the text, these difficulties only highlight the sheer magnitude of the author's task of untangling the fine detail of two centuries of English dialect geography. In Chapter 4, however, poor layout and insufficient explanation obscure to an extent the new methodologies and data illustration techniques laid out there. Nonetheless, applying Salzburg dialectometry to English data for the first time represents a major achievement and a practical response to Wegmann's central thesis that ‘there are no clearly defined dialect boundaries in the East and the South East of England’ (167). A more sophisticated method of cartographic representation is needed to reflect the subtleties of variation in these areas, and no doubt others – and Language in Space paves the way.