Not for minds bigger than bellies

Jean Aitchison gets a taste of a language handbook that is a little hard to digest.

"The study of language variation and change, the core of the sociolinguistic enterprise, has a relatively short history, but it is a burgeoning history," note the editors of this chunky handbook, which they claim "reflects the vitality and growth" of the field and represents the discipline in its "multifaceted pursuits". Until some time in the 1980s, it was possible for an enterprising graduate student to read virtually everything in the field of sociolinguistics, they point out, though this is now impossible as the discipline has faced "exponential growth".

What, then, is a "handbook"? According to a recent dictionary, it is "a guidebook that lists brief facts on a subject". This hefty tome is one of a number of Blackwell linguistics handbooks, which, according to the cover blurb, aim to form a multi-volume series that "covers all the major subdisciplines within linguistics, and, when complete, will offer a comprehensive survey of linguistics". So far, ten have been published. This one sits somewhat uneasily between two others, The Handbook of Sociolinguistics (1997) and The Handbook of Historical Linguistics (in press). A number of the authors of the handbook under review feature in these other books, as do several of the topics.

Why, then, do we need this book? It may be a gap-plugging exercise intended to supplement existing volumes. The so-called sociolinguistics handbook contained mostly broadish macro-sociolinguistic papers and might perhaps have been better titled The Handbook of the Sociology of Language , especially as it included relatively little micro-sociolinguistics - the nitty-gritty variation studies that occupy most practising sociolinguists. This new volume was possibly needed to reflect the discipline of sociolinguistics more realistically. "Language variation" is an accurate summary of the subject matter, but presumably to cater for the growing interest in the relationship between variation and diachrony, the word change needed to be included: change cannot occur without variation, as is now well known - though variation is sometimes found without change.

This volume, say the editors, is "a convenient hand-held repository of the essential knowledge about the study of language variation and change... a resource for readers to turn to for garnering basic information about any of the sub-fields and for getting directions on where to learn more".

The 29 contributors have been invited to discuss "the ideas that drive their branch of the discipline, and to illustrate them with empirical studies". This, we hear, is something of a balancing act. They have been selected partly to "strike a balance between the first two academic generations, the founders and their intellectual offspring". The topics they were asked to write about are "balanced between the relatively mature and the relatively recent", and the ideas hopefully "find a balance between the tried-and-probably true and the potentially productive". The book is dedicated to William Labov, widely regarded as the father of modern sociolinguistics. His books and papers are quoted extensively.

The result is a challenging hybrid. If a platypus is a duck put together by a bunch of scholars, and a camel is a horse designed by a committee, then this book is maybe a "platycam", a not entirely coherent collection, though with some worthwhile and interesting components. Despite the book's title, only a few of the papers deal with both variation and change, though these are often the strongest and most interesting.

The handbook is divided into five major parts, and two of these are subdivided. Each subsection has its own introduction, written by a different person. The book appears to have been organised as a sandwich. First comes methodology, which includes old and new ways of handling the data. The sandwich filling is a survey of classic sociolinguistics, with particular attention to topics that bridge both variation and change. The volume finishes with bits and pieces that do not fit in elsewhere.

Part one, "Methodologies", will undoubtedly be useful for anyone who wants to know how to "do" sociolinguistics. It starts with a section titled "Field methods", which begins with a sensible shortish paper by Crawford Feagin outlining how to set about sociolinguistic fieldwork, including mention of the well-known rule of thumb that one-third of the project time will be spent in fieldwork, one-third in analysis, and the final third writing up the work. Next comes a paper by Dennis Preston on popular beliefs about language, such as the folk view that "US Southerners are laid-back and lazy; just listen to their lazy drawled vowels". (The topic of attitudes to language is a burgeoning one within linguistics, so this paper is a welcome contribution.) Further useful papers in this section are those by Edgar Schneider on the relationship between a speech event and its written record, and Laurie Bauer on the use of electronic corpora. The second section of part one takes a steep step up in technicality, although it contains useful papers by Robert Bayley on the quantitative paradigm, with particular attention to so-called Varbrul (variable rule) analysis, the most common method of multivariate analysis in quantitative sociolinguistics, and one by John Rickford outlining the pros and cons of implicational scales (if a language has x, then it will also have y).

Part two, "Linguistic structure", is somewhat anaemic, containing only four short papers, of which one on "Discourse variation" fits somewhat uneasily into the topic. As the introduction to this part notes: "It has proven difficult to incorporate variation into linguistic theory or theoretical linguistic models into variation study", though it optimistically concludes that "important inroads are being made".

"Social factors" forms part three and is the core of the book. It contains almost 300 pages, divided into three sections. A key paper in the first section on "Time" is that by Guy Bailey, who explores the relationship between real time and apparent time. Following Labov, he shows how differences in the speech of different generations (apparent time) might mirror actual diachronic developments (real time), but also discusses the importance of being on one's guard against confusing demographic change with true linguistic change. Julie Roberts looks at child language variation and discusses possible reasons for the relative neglect of this topic within sociolinguistics, pointing out the difficulty of distinguishing between variation that children may have acquired from their caregivers, and developmental change.

The second section of this part, "Social differentiation", contains discussions on "sociolinguistics proper", as it is most usually understood, with papers looking at the correlation between language and styles, social class, sex and gender, and ethnicity. Most of the findings outlined are relatively well known, though the papers contain some useful updating on key issues, such as Jenny Cheshire's paper on sex and gender, which shows how the relative importance of this field has increased over the years.

The third section, "Domains", handles "relational arenas in which variable linguistic behaviour takes place". The best-known author here is Lesley Milroy, who discusses social networks, though other useful papers are those on the domain of the family by Kirk Hazen, and "communities of practice" by Miriam Meyerhoff, with this concept being defined as "an aggregate of people who come together around mutual engagement in an endeavour" - though she admits such studies overlap with social networks.

"Contact" is another ultra-slim section and forms part four. It has only three papers, though high-quality ones; the paper by David Britain on spatial diffusion of change is particularly welcome. Also included is one by Gillian Sankoff on linguistic contact and one by Paul Kerswill on "koineization", a type of contact-induced levelling that may result in rapid change.

Part five, "Language and societies", is the anorexic of the book, with fewer than 100 pages and again only three papers. It includes various topics not yet dealt with, such as language death.

As this brief outline suggests, the editors have clearly had a hard time justifying the book's title, while not overlapping with other books in the series. One gaping hole exists: pidgins and creoles, which many regard as prototypical cases of variation and change.

The book contains a number of goodies, though identifying and evaluating them may prove difficult for some readers - the brief introductory sections do not provide quite enough help. It brings to mind some of those "eat all you can" buffets. Let us hope that the readers select wisely, otherwise they might end up with indigestion.

Jean Aitchison is professor of language and communication, University of Oxford.