The Mouton World Atlas of Variation in English (WAVE), edited by Bernd Kortmann and Kerstin Lunkenheimer, is an impressive and significant achievement. Consisting of almost 1,000 pages of detailed accounts of morphosyntactic patterns in 55 varieties of English from around the world (with some analysis of a further 19) by an international range of experts (many of whom are native speakers of the varieties they describe), WAVE represents a major step forward in our understanding of dialect variation in English and illustrates in fine detail a vast array of linguistic systems that fall under the umbrella ‘English’ across the world today. Using a list of 235 morphosyntactic features, WAVE explores diatopic variation in L1, L2 and pidgin/creole varieties across the globe, illustrating the results in 96 full-colour maps, and investigates the deeper relationships between the varieties in 23 colour-coded phenetic networks. It is a veritable feast for the eyes, but there is so much detail in it that it is much more than that; it is, in fact, a huge database echoing in many ways The World Atlas of Language Structures (Haspelmath et al. Reference Haspelmath, Matthew, Dryer and Comrie2005) in its depth and breadth.

WAVE consists of an Introduction, which explains the background to the study, the varieties and features investigated, the methods used for gathering and coding the data, and the organisation of the atlas. This is followed by seven parts containing the descriptions of the dialects under investigation (The British Isles, North America, The Caribbean and South America, Africa, South and Southeast Asia, Australia and the Pacific, Isolates) and three parts exploring patterns in the data (Regional profiles, Typological profiles, Global profile).

Impressive though it is, it is inevitable in a work of this size and scope there are some things which are problematic or missing. Two methodological principles discussed in the Introduction are particularly worthy of comment. Firstly, the varieties included in WAVE are divided into five ‘types’: Low-contact traditional L1 dialects (L1t); High-contact L1 varieties (L1c); L2 varieties (L2); Pidgins (P); Creoles (C). The authors tell us (p. 4) that ‘it was one of the express aims of the WAVE project to test the extent to which these distinctions … are reflected in the morphosyntactic similarities and differences between the varieties’, but they also note (p. 4) that these types ‘are of little importance for the present undertaking of identifying the structural profiles of varieties in the Anglophone world (and possible larger groups thereof)’. It seems, in fact, that the idea of dividing the varieties into these different types was problematic in various ways. For example, the authors state (p. 4) that the decision as to which type each variety belonged to was not decided by the WAVE editors, but by the individual contributors for each variety. This has given rise to some rather unexpected results. Whilst it would be difficult to argue against the categorisation of the traditional dialects of East Anglia or southwest England as being L1t varieties, the same cannot be said for varieties such as those of Orkney and Shetland, or of Newfoundland, all of which have developed in situations of intense language and dialect contact in recent centuries (even if they are now in situations of low contact), but which are also categorised as L1t varieties. Conversely, Irish English is defined as an L1c variety, even though English has been spoken in parts of Ireland since the medieval period and for many varieties of Irish English contact with Irish is long in the past. It is not surprising then that this categorisation of varieties into types is not pursued further in WAVE, but it does feel like something of a lost opportunity which might have been remedied by editorial involvement in the assignment of varieties to types. The second methodological decision described in the Introduction which raises further questions is the rating of each morphosyntactic feature in every variety according to six categories (p. 5): A, feature is pervasive or obligatory; B, feature is neither pervasive nor extremely rare; C, feature exists, but is extremely rare; D, attested absence; X, not applicable; ?, no information available. These decisions, which are the primary answers to the WAVE questionnaire and are the basis of the further analyses in it, were of course made by the specialist contributors. However, the editors note (p. 6) that ‘some variation remains in how features were interpreted by contributors, how A, B and C ratings were assigned to features not categorically present or absent in a variety, and in whether certain features were considered to be applicable at all in a variety … Thus it has to be said that the ratings … have to be taken with a pinch of salt’, even though the editors provided further guidelines on this issue to the contributors during the course of the survey (p. 5). Given this uncertainty, the further analyses in WAVE based on these uncertain categorisations must inevitably be treated with caution.

The meat of WAVE is of course the individual dialect descriptions in Parts I-VII. Although there is a fair amount of variation between chapters in terms of their organisation and content, most of them have an introductory section or sections (covering topics such as sociohistorical and geographical background, data sources and particular issues with the variety under discussion), an analysis of key features captured (or not) by the WAVE questionnaire, followed by the questionnaire results and examples, and relevant references. One of the most pleasing things about these chapters is that they constitute mini introductions to varieties of English around the world – not just to their morphosyntax, but also to their history, geography and sociolinguistic variation. For example, Trousdale's account of Northern England is an excellent introduction to the issue as to whether there is a ‘linguistic north’ in England and what kinds of evidence might be brought to bear on the problem.

As with the Introduction, there are issues with this part of WAVE. Notably, these descriptions are not necessarily of single linguistic varieties at all, but instead may cover a range of related dialects. For example, in the chapter by Markku Filppula it is noted that the term Irish English ‘is used here as a cover term for a wide range of the non-standard regional and social varieties of English spoken in the Republic of Ireland and in Northern Ireland’ (p. 30). There is no sense in which this is a single dialect with a single set of morphosyntactic features, and it is not surprising, then, that the questionnaire results for this ‘variety’ are rather different for what I would give for my own (Mid-Ulster) Irish English variety. Likewise, the chapter on Scottish English and varieties of Scots by Jennifer Smith covers everything from Standard Scottish English to traditional Scots dialects from Lerwick (Shetland), Buckie (northeast Scotland), Cumnock (southwest Scotland) and Culleybackey (Northern Ireland). Although Smith concentrates on traditional dialect features and carefully identifies their geographical provenance, it is not at all clear that these data can (or should) be taken as a single data point in the WAVE survey and its analyses. Although generalisation of this nature is probably unavoidable in a work such as WAVE, descriptions of specific, narrowly defined dialects would potentially have made the results of the survey more focussed and accurate if not more geographically exhaustive.

Furthermore, whilst the linguistic and geographical coverage of WAVE are impressive, there are unfortunate lacunae in the data and descriptions which it would have been nice to have seen filled. In the questionnaire appendices for each variety, only features categorised as A, B or C (see above) are given, and it is not made clear whether the questions which have been left out received the answers D, X or ?, which have rather different meanings from each other. And although most authors liberally illustrate the responses for their variety with examples, some do so more than others, and questionnaire responses are not infrequently given without them so that it is unclear how the feature in question is realised. Furthermore, chapters for a number of varieties included in the WAVE survey and analyses are absent entirely, including ones of crucial interest such as Southeast English English, Colloquial American English, Appalachian English, Ozark English, and New Zealand English. Although the data for these varieties is included in the analyses, descriptions and discussion of the questionnaire returns for them are absent and are sorely missed.

Despite the issues I have outlined in this review, The Mouton World Atlas of Variation in English is a book that anyone interested in morphosyntactic variation, English dialects, World Englishes and language contact will find to be an indispensable resource and a real goldmine of information on a wide range of varieties and their features. It will doubtless act as a foundation for a great deal of new research on morphosyntactic variation in English and other languages and will be the yardstick against which further research of this nature will be measured.