This article investigates the utility of Eckert & McConnell-Ginet's concept of “community of practice” for an analysis of the language used by women in a sexual assault tribunal. It is shown how the questions asked by two tribunal members (a male and a female faculty member), in a university sexual “harassment” tribunal, function to (re)frame and (re)construct the events in question as consensual sex. Although the female complainants (i.e. victims) in the tribunal characterize their experiences as sexual assault, two of the tribunal members – one of whom is a female faculty member – ask questions that presuppose the inadequacy and deficiency of the complainants' signals of resistance, suggesting that their so-called lack of resistance was tantamount to consent. Clearly, any homogeneous notion of “woman's speech style” or “woman's point of view” would fail to account for the differences between the discursive patterns of this woman tribunal member and the women complainants in this context. However, if (as argued by Eckert & McConnell-Ginet 1992a,b) our linguistic practices arise out of the kinds of community of practice with which we are involved, then an understanding of such local practices and activities should provide greater insight into the differential linguistic behavior of the women involved in this sexual “harassment” tribunal.

Substantial thanks are due to van Dijk for assembling his two-volume “Discourse studies” collection, which introduces this multi-disciplinary field to a wide readership. In this first volume, which focuses on theory and method, linguists, semioticians, communication theorists, and social and cognitive psychologists from the Netherlands, Austria, Germany, the UK, the US, and Australia have contributed to van Dijk's project of promoting dialog and collaboration in the development of discourse studies as an emerging field. Each chapter offers a useful introduction to the literature in its area and an exposition of theoretical perspectives, usually illustrated by one or more examples of text analysis. The volume is a treasure for newcomers to discourse studies, or to discourse analysis methodologies; in addition, it challenges specialists to connect their work more fully with approaches from other perspectives.

This ambitious volume succeeds in pulling together diverse approaches to surveying and analyzing human culture through human linguistic ability, as viewed by linguists and anthropologists. It does not, by any measure, take an extreme Whorfian view – or, for that matter, an extreme anti-Whorfian view – of the relationship between language and culture; rather, it sets forth a series of ideas about the mediating role of cognition in the ways in which we view and talk about the world around us. This is not a surprising stance for the editors; Taylor has long been a proponent of the semantics-based theory of cognitive grammar (cf. Taylor 1995), and MacLaury has recently published a study (1997) of color terminology within an offshoot of cognitive grammar called vantage theory.

Almost forty years ago, the “Whorf Hypothesis” was one of the things that attracted me to linguistic anthropology. George Orwell's 1984, with its dark vision of a world made safe for power by bureaucratic euphemism, was one of my favorite books, and Whorf's claim that novel and valuable ways of understanding the world might be encoded in small stateless languages struck me as a particularly telling and precise statement of a large anthropological commitment. But my own Whorfianism, and everybody else's as well, was soon to be seriously challenged. In the late 1960s and early 1970s, new scholarship on language universals and linguistic typology undercut the theory of linguistic relativity. Whorf's own best-known descriptive claims, especially those about Hopi, were challenged by knowledgeable field workers (Voegelin et al. 1979, Malotki 1983). By the early 1990s, Steven Pinker could confidently write that Whorfianism was “wrong, all wrong” (1994:57), “outlandish” (63), and “bunk” (65) – and this is a mere subsample of Pinker's characterizations.

This collection aims to offer, in the words of its editor, “a first introduction” to the study of discourse. While its companion vol. 1, Discourse as structure and process, is mainly concerned with cognitive issues, the contributors to vol. 2 focus on social organization and social processes.

This article considers the usefulness of the community of practice framework (Eckert & McConnell-Ginet 1992) to an analysis of women's pregnancy stories. The study reveals that the operative communities of practice in the lives of pregnant women are the communities comprised of doctors, health professionals, family members etc., all of whom have specific and well-defined opinions about pregnancy. Although pregnant women do not themselves constitute a community of practice, the framework leads to a variety of insights about women and pregnancy. From analyzing women's stories, it becomes apparent that others routinely construct for women what their pregnancy is to consist of – often trying to make of it a monolithic experience, whether or not this coincides with a woman's own experience. Women's stories provide data that explicitly address the intersection and collision of disparate communities of practice.I would like to thank Janet Holmes for suggesting that I apply the community of practice framework to my corpus of pregnancy stories. Thinking about this has led me to insights that I might not have otherwise had. I would also like to thank my colleague and friend, Alice Greenwood, for the very careful reading that she did of several earlier drafts of this paper. Her comments contributed significantly to the development of the ideas for this article. Of course I alone am responsible for what is presented below.

This collection of Ferguson's articles brings together his research over four decades. When Ferguson started writing about topics related to language and society, the term “sociolinguistics” was barely used and had no recognized status as a field of inquiry. Today, the term seems too diffuse for all the strands of specialization that range from the examination of micro-socio-phonetic detail to the consideration of broad-based macro-sociological and linguistic institutions. Perhaps better than those of any other individual scholar, Ferguson's research interests reflect the breadth of the field. His topics of study have ranged “from Arabic linguistics to applied linguistics, from child language acquisition to language planning, from language and religion to language universals, from Bengali syntax to American sports announcer talk” (p. 3). His impact on the field is undeniable; but because of his expansive interests in an age of increasingly (and sometimes myopic) specialization, few readers are familiar with the full range of sociolinguistic topics that bear this scholar's mark. In Huebner's selection of articles, Ferguson's overall contribution to the field is put in perspective, although the collection is hardly exhaustive of his impact. (See now also Huebner's 1999 obituary of Ferguson.)

This book is the first in a series, edited by Stephen Levinson, which (according to the dust jacket) “focuses on the role that language, in both its universal psychological aspects and its variable cultural aspects, plays in human cognition ... [as well as] the relation of speech production and comprehension to other kinds of behavior in cultural context.” Each chapter focuses on one point of contact between linguistic and conceptual representation. Since the goal of the series is of central concern to readers of LiS, the first book in the series should be especially interesting, as a harbinger of relevant studies to come.

Simply put, MacLaury's new book on color terminology systems in a hundred Mexican and Central American languages is a groundbreaking study on how the human mind apprehends the physical universe; it is by far the most important work on color nomenclature theory to appear in the past thirty years. M's insights, which are the culmination of some two decades of fieldwork and analysis, will be of tremendous value to linguists, anthropologists, psychologists, and cognitive scientists of many different backgrounds and interests.

Singh criticizes sociolinguistics, whether in its Labovian or its Gumperzian guise, for failing to take into account what he describes as linguisticality and sociality. Thus variationism fails to take into account the underlying phonological laws that not only are a part of all languages, but are also specific to certain languages; and interactionism fails to take power and inequality into account.

The search for explanatory coherence in language and gender research has fostered a variety of research methods and analyses; this article evaluates the contributions of the Communities of Practice approach, with its focus on the constructive practices of a group – especially mutual engagement of learning a jointly negotiated practice of gender. Rather than presupposing gender differences as a starting point, CofP emphasizes the learning and mutability in gendered linguistic displays across groups; CofP theory thus naturalizes intragroup variation, not marking it as deviant. However, while the CofP approach focuses much-needed attention on the social construction of gender as local and cross-culturally variable, gender research must be augmented by critical study of two other facets of gender: ideology and innateness, which are critical components of a more comprehensive theory of gender for language research. I would like to give thanks to Janet Holmes for organizing the original forum in which the paper developed, and for her very helpful comments; to Sally McConnell-Ginet and Penny Eckert for their sustained support, through their work, of me and others in the field; and to all three of them – plus Alice Freed, Miriam Meyerhoff, and Susan Ehrlich – for stimulating comments and discussion of these issues. Thanks also go to Mary Talbot and Craig Waddell for comments on an earlier draft. All errors remain my own responsibility.

Turns at talk in conversation, and in any other form of spoken interaction, are designed to enable speakers to be understood in the way they wish to be understood by their co-participants. Talk is meaningful insofar as speakers design their turns so as to be recognizable as making an offer, closing a topic, agreeing or affiliating, being ironic, finishing their turn, continuing, being surprised or astonished at news, “just” asking a question, reproaching, indicating that there is something problematic about what the prior speaker has just said, and so on. (I take these activities from those that are variously studied in the chapters in this collection.) So “turn design” is at the heart of how we mean what we say, what we communicate, in interaction.

Tonkin treats a complex and timely set of ideas when she studies the relationships between oracy and literacy, oral narrative performance and written texts, memory and history and society. To discuss these areas of scholarly debate, she employs a multidisciplinary, sometimes contentious variety of studies and assertions. Her efforts mostly succeed in promoting her claims for the necessity of blurring past distinctions and categories in the study of oral history, and for taking a much more performative/interactive view of its construction in a living context.

Linguistics, as one of the last academic disciplines in the US to adopt guidelines for gender-inclusive language (Thorne 1998), has not been in the forefront of the analysis of language and gender. Indeed, it has been slow to give – either to the social context of language use, or to the socially constructed concept of gender – the kind of serious analytic consideration that it regularly gives to the mental representation of language. This collection of articles edited by Bergvall, Bing & Freed signals a change within the profession itself in its re-examination of basic concepts to be taken into account in future research on language and gender.

What is alloglottography? A diaeresis? A digamma? Whose writing system has kanamajiri writing and kokuji? How would you start to find any of these in a conventional writing system text/reference, unless you knew where (in the world) to start? What about opisthograph, ostracon, quoc-ngu, and tugra? None are in the index of Daniels & Bright 1996, which I consider the best book to date on the world's writing systems. But all are entries, cross-referenced to other entries, in Coulmas's Encyclopedia. The reader can also look up Bamum writing, Djuka syllabic writing, the Hatrene script, Hsi-hsia writing, the Loma syllabary, Peguan script, Tifinagh, Urartian writing, and the Wolof alphabet directly, without having first to know what set of writing systems, geographical or typological, they belong to. My personal favorite is Sogdian writing (471–74), an Aramaic-derived script used by Persian colonists in Chinese Turkestan; the cursive form of this writing system is attributed to Ahriman the devil, because it is so hard to distinguish the letters. What a pleasant surprise, for one satiated with discussions of the weaknesses and unnecessary complexities of Japanese writing!

The book under review addresses the central concern of this journal: the nature of the relation between language and society. Hasan's conception of this relation is a wide-ranging one; it includes the relation between different social positionings and different styles of speaking, the ways that language contributes to the communication and maintenance of ideologies, and the relation between the language of a text and the structure of that text. In addition, one finds contributions to what some readers will consider “pure” linguistic theory. Hasan devotes one chapter to an argument for including lexis as a part of grammatical description, and a second to the necessity for having semantic networks in addition to having grammatical networks. Both “theoretical” issues form critical steps in Hasan's dream of describing the “continuity from the living of life on the one hand right down to the morpheme on the other.” It is worthy of note that her argument regarding semantic networks is presented within the framework of a theory of linguistics that tends to semanticize grammar; thus this is a more significant claim than it would have been if she had been using a formalist approach.

The central idea that informs Macaulay's book is aptly expressed in a passage he quotes from Sapir (1929:214):

Language is primarily a cultural or social product and must be understood as such ... It is peculiarly important that linguists, who are often accused, and accused justly, of failure to look beyond the pretty patterns of their subject-matter, should become aware of what their science may mean for the interpretation of human conduct in general. (138)

Language use and attitudes in Egypt and the larger Arab world pose a challenge to much of the received wisdom among sociolinguists about the role of gender in sociolinguistic variation. More generally, they cast doubt on widespread assumptions in sociolinguistics about the nature of linguistic markets and language change. Haeri's recent sociolinguistic investigations in Cairo show clearly the difficulties raised by viewing linguistic resources as based on a vernacular/standard continuum, with women leading men in use of established “standard” forms – and also (on this view, somewhat “paradoxically”) leading in use of “non-standard” incoming variants introduced in the speech of working-class and lower middle-class speakers. As Haeri points out, there is a paradox here only if one assumes that some unchanging social meaning is attached to the “standard/non-standard” dichotomy. But she points to many problems both with the dichotomy and with the way it is often understood in linguistic discussions. Haeri cogently argues against the common assumption that members of a speech community recognize a single “standard” that (a) is required for performance in high-level jobs, (b) is taught and promoted in educational institutions, (c) is the medium of highly valued literature and other culturally valued activities, and (d) is associated with upper-class speech. Although such a conception may not seem too distorting in many Western urban settings (some of my students spontaneously came up with “richspeak” to designate a so-called American English “standard”), it is woefully inadequate for understanding the sociolinguistic dynamics of a place like Cairo.

Hedging has become a topic of renewed interest recently among discourse-oriented linguists. Recent publications attesting to this fact include Salager-Meyer 1994, Hyland 1994, 1996, 1998, Crompton 1997, and Skelton 1997, not to mention work on modals (e.g. Stubbs 1996, chaps. 5 and 8), evidentials (e.g. Barton 1993), and “vague language” (Channell 1994). The present volume continues this trend, providing a wide-ranging and truly international set of perspectives on the phenomenon of hedging.

Until the last few years, linguists' interest in the language of the neurologically impaired has been primarily from two orientations: psycholinguistic and neurolinguistic. The former has applied theories concerning the mental processing of language, using acquired language disorders as a test bed for exploring and expanding these theories. The latter attempts to correlate language (disordered or not) with functional lesion sites in the brain; it has recently received a major boost from the technical developments of functional brain imaging, but its main theoretical base remains that of psycholinguistic processing.